Wikipedia talk:Verifiability/Archive 60

Archive 55

←

Archive 58

Archive 59

Archive 60

Archive 61

Archive 62

→

Archive 65

Some thoughts on WP:CHALLENGE

Not a proposal (at least not yet)... just want to share some thoughts as we continue to work on the wording for this section. One thing that has always struck me as odd about the CHALLENGE paragraph is the fact that we jump right to "removal", as if that was the only method of challenging unsourced material. And, of course, it isn't. Challenges can be made with a query on the talk page, or by tagging (or in other ways?) I think it might help to mention those other options before we mention the more drastic option of removal. That would put "removal" in proper context... as one of of a range of options.

It is, of course, always a judgement call as to which form of challenge will be best in a given situation ... but my personal take on it goes like this: The more you think it likely that the unsourced material is verifiable, the more you should lean towards a talk page query as your best option. The more you think it unlikely that the material is verifiable, the more you should lean towards removal (and, of course, we lean far more heavily towards removal in BLPs than we do in other articles).

Your initial thoughts on all of this? Blueboar (talk) 14:10, 2 January 2013 (UTC)

• The section currently doesn't say much about using the talk page and so encourages editors to shoot first and ask questions later. As a worked example of good practise, consider the current FA. I skim this and quickly spot a dubious quotation: Its call sounded like "barf barf". I inspect the reference which supports this which seems to be an inscription on the back of a drawing which was made about two centuries ago. That inscription has it as "baf baf" but someone else has decided what this really sounds like and it's not clear how reliable this is as the bird is long extinct. Anyway, my point is that the obvious thing to do here is to raise the matter on the talk page. Messing directly with the article would be impertinent and it would be better to encourage discussion in such cases. Warden (talk) 15:04, 2 January 2013 (UTC)

• While I was writing that, two other editors had already started discussion at Talk:Mauritius Blue Pigeon#The noise that it made :: hoax? Their independent action confirms that the talk page is the place to start. Warden (talk) 15:16, 2 January 2013 (UTC)

Again, the appropriate form of challenge depends on the specifics. I would agree that the talk page is often (even usually) the best place to start... but there are some situations where removal without prior discussion is called for. And it really is an editorial judgement call as to which is which. That is the problem with trying to be overly specific here in the policy... I think it accurate to say that we want to encourage discussion, while at the same time allowing editors the freedom to edit as is appropriate - given the specific situation. We have to trust our fellow editors to make the appropriate judgement call.

My concern is that (currently) the section doesn't even mention that there are other options for CHALLENGing unsourced information... it focuses entirely on removal. I think that should be changed. Blueboar (talk) 15:39, 2 January 2013 (UTC)

I have to preface what I'm going to say, by saying that it's awfully difficult to wrap my head around a new direction for this, at this time, and it might be better to wait on the idea. But that said, I think I agree, in principle, with Blueboar. Pointing out that there are also more collaborative ways to "challenge" seems potentially to be a good thing, and I like it. A red flag for me is that there could be drama over whether an editor did or did not follow "policy" versus "best practice". In other words: "I'm taking you to ANI because you deleted it instead of discussing it on the talk page." Stuff like that. --Tryptofish (talk) 15:48, 2 January 2013 (UTC)

• Agree, there's a range of options. The basic idea that needs to be conveyed is this;

• Adding BAD^TM information is WP:DISRUPTive
• Removing GOOD^TM information is WP:DISRUPTive

The goal of every article is to reach FA status and everything needs to be verified before that can happen. Ideally, all information should be cited at the time it's added, but often times that doesn't happen. How do we put that in to few simple words that people can understand? How do we make sure people don't use WP:V to be disruptive and illustrate a WP:POINT? I don't know the answer, but we should probably used a range of options. 64.40.54.118 (talk) 15:52, 2 January 2013 (UTC)

I don't disagree... The converse of this, of course, is:

• Removing BAD^TM information is NOT disruptive
• Adding GOOD^TM good is NOT disruptive.

The question is whether some specific bit of information is BAD or GOOD. That's a judgement call, and one that editors can disagree on. This is why I think mentioning the other options is important. There is a middle ground. Blueboar (talk) 14:40, 3 January 2013 (UTC)

I agree overall with your idea Blueboar. But I can't help but to think you are still not in line with WP:faith. Believing it's "unlikely" that a source can be found or not sounds to me like someone who don't have a clue about what the subject is. If you "did your homework" correctly and sought as much point of view as you could, you should be damn confident about whether a source exists or not.

Yeah ! That's how I would say it : You can challenge only if you sought for as much point of view as you could. Otherwise it would be in contradiction with WP:faith (and questionable in regard to WP:NPOV). Iluvalar (talk) 21:13, 2 January 2013 (UTC)

My ongoing suggestion is to just require some sort of a good faith challenge or questioning of the material when tagging or deleting material. And, awkwardly put, that the questioning should be on some grounds besides (in addition to) sourcing. I think that some of the wording was simply looking to go a bit this way but word it less awkwardly. (E.G. say that they think it may not be verifiable) I think that trying to specifically talk about the next stages (E.G. "if you do XXXX in case YYYY it may be considered to be wp:disrputive") are problematic for several reasons.North8000 (talk) 22:18, 2 January 2013 (UTC)

perhaps the simplest change to deal with the questions above is to make the addition shown below: "information which is in any way contentious or likely to be challenged in good faith can be removed. It is all to possible to go through the encyclopedia challenging perfectly reasonable things in order to make a point or cause trouble. I do not consider this a good faith challenge. DGG ( talk ) 19:12, 3 January 2013 (UTC)

with " "but information which is in any way contentious or likely to be challenged can be removed

Not following .....is there a typo? North8000 (talk) 19:39, 3 January 2013 (UTC)

Shouldn't we assume that all removals are made in good faith, unless we have conclusive evidence otherwise? Blueboar (talk) 20:32, 3 January 2013 (UTC)

The proposed change looks scrambled and unclear, even to the point of whether this is a replacement of "challenged or likely to be challenged" (which I would probably oppose) vs. using those words elsewhere. (which I would probably support) But idea of introducing wording to the effect of a "good faith" challenge might have some merit. North8000 (talk) 21:20, 3 January 2013 (UTC)

(ec; in reply to Blueboar's "Shouldn't we AGF") Yes, absolutely. However, while it may be "all too possible to go through the encyclopedia challenging perfectly reasonable things in order to make a point or cause trouble" we have independent policies to deal with those behaviors. However, I'd be willing (as I've said before) to at least entertain the possibility that it may demonstrate a lack of good faith to go around regularly deleting unsourced information in a patrollish kind of way even if the pattern of those edits do not demonstrate a clear POV or POINTy behavior, but the way to be sure of that is to say that plainly here, not to open up an examination into a deleting editor's motives that turns on "how good" particular unsourced information appears to be or not to be. Best regards, TransporterMan (TALK) 21:24, 3 January 2013 (UTC)

Cleanup from previous refs without reflists

Blueboar, Just for the sake of discussion, would the following change of the first three sentences of the Challenge paragraph be what you want?
From,

"Any material lacking a reliable source directly supporting it may be removed. Whether and how quickly this should happen depends on the material and the overall state of the article. Editors might object if you remove material without giving them time to provide references; consider adding a citation needed tag as an interim step."

to,

"Any material lacking a reliable source directly supporting it may be challenged by removing it, tagging it with {{citation needed}}, or questioning its verifiability on its article's talk page. The choice between these options would depend on the material and the overall state of the article."

--Bob K31416 (talk) 02:58, 4 January 2013 (UTC)

I think that the policy should mention other ways to issue a challenge (besides removal) before we say "Any material lacking a reliable source directly supporting it may be removed". And I would rather say it in the text than hidden away in a footnote. I am still pondering how to say it, however... so I don't have any suggested language... yet. Blueboar (talk) 14:21, 5 January 2013 (UTC)

The version in my previous message satisfied your request to mention other ways in the text. Here again is the part that addressed your request, "Any material lacking a reliable source directly supporting it may be challenged by removing it, tagging it with {{citation needed}}, or questioning its verifiability on its article's talk page." --Bob K31416 (talk) 14:51, 5 January 2013 (UTC)

You still mention removal before the other options... that puts more emphasis on removal. Off the top of my head, I would take your wording and swap things around:

• "Any material lacking a reliable source directly supporting it may be challenged. Challenges can be made by raising a query on the article's talk page, tagging the material with {{citation needed}} or some other appropriate tag,, or by removing the material completely. Which form a challenge should take is an editorial judgement call, and depends on the specific nature of the material in question."

Or something like that (I am still pondering how best to say all of this). Blueboar (talk) 15:06, 6 January 2013 (UTC)

While you're thinking about it, here's my previous version with the simple change of reversing the order, as you prefer.

"Any material lacking a reliable source directly supporting it may be challenged by questioning its verifiability on its article's talk page, by tagging it with {{citation needed}}, or by removing it. The choice between these options would depend on the material and the overall state of the article."

I've contributed what I could, so now I'll leave it up to you to follow up on your ideas. Regards, --Bob K31416 (talk) 03:29, 8 January 2013 (UTC)

And thank you for that contribution... it helps. Blueboar (talk) 04:09, 8 January 2013 (UTC)

Conflation of three separate concepts

I have been thinking about the entire CHALLENGE/BURDEN section... and I have come to the conclusion that part of the confusion and angst over the section is that we conflate three separate but related concepts. 1) When citations to reliable sources are needed... 2) what to do when the needed citations are not provided... and 3) who is responsible for providing those needed citations. I think this confusion can be resolved by separating these concepts into their own sub-sections ... so I propose the following reoganization (mostly just moving the existing wording around, but with some new language to link it together) Blueboar (talk) 16:50, 22 December 2012 (UTC)

When a reliable source is required

While it is not necessary to cite every sentence in Wikipedia, quotations and any material challenged or likely to be challenged must be attributed to a reliable published source using an inline citation. The citation should fully identify the source, and the location within the source (specifying page, section, or such divisions as may be appropriate) where the material is to be found. See Citing sources for details of how to do this.

What to do when sources have not been provided

Challenged material may be removed from the article. However, alternatives to removal should be considered first. Whether and how quickly removal should happen depends on the material and the overall state of the article; consider adding a citation needed tag as an interim step. Editors might object if you remove material without giving them time to provide references. It has always been good practice to try to find and cite supporting sources yourself. Do not leave unsourced or poorly sourced material in an article if it might damage the reputation of living people; you should also be aware of how the BLP policy applies to groups.

Burden of evidence

The burden of evidence lies with the editor who adds or restores material. If material lacking an inline citation to a reliable source is challenged (either with a tag or by being removed), the burden of responding to the challenge by supplying the necessary citation rests with those who wish the material to remain in or be returned to the article. The challenger does not need to "prove" that the material is unverifiable. Instead those who wish to retain the material must demonstrate that it is verifiable.

(footnotes)

Discussion of the proposed reorganization

Please think in terms of the presentation right now (ie is there consensus for the way I divide the topic up and organize it, rather than on specific wording... if there is consensus for the presentation change, then we can focus on any wording changes as a second step). Blueboar (talk) 17:03, 22 December 2012 (UTC)

• Blueboar. I am not entirely sure what I think yet. Let me read through this. I love the fact that you want to do something. My feeling is that we may not need as drastic a change (even though I understand this is less change and more rearranging).--Amadscientist (talk) 21:07, 22 December 2012 (UTC)

Take all the time you need. :>) I propose it because I think it makes sense and resolves a lot of the problems... but I am not in any rush. It is more important to do it right than to do it right now. Blueboar (talk) 21:31, 22 December 2012 (UTC)

I'll also put in a support for taking our time with this idea. Even though you may be correct that it could help resolve the issues that are being discussed about the existing version, I'd still like to come to some agreement about those issues first, which I think is entirely possible, and only then examine the possibility of reorganization. I say that because I think discussing the reorganization, we won't be able to avoid thinking about the specific wording. I'll also quibble with something you said in passing, that there is a lot of angst about the existing Burden section. I actually think that there is only a lot of discussion, which seems to be an idiosyncratic property of this talk page and those of us who talk here (how many angels can verifiably fit on the head of a pixel?). But I'm actually optimistic about the substantive issues. --Tryptofish (talk) 22:23, 22 December 2012 (UTC)

• I don't see the improvement. There is even a problem with the first phrase, "While it is not necessary to cite every sentence in Wikipedia ...". An editor reading this may think that when he adds material he doesn't have to go to the trouble of including citations until the material is challenged. --Bob K31416 (talk) 02:34, 23 December 2012 (UTC)
  • I changed "thus" back to "therefore" only because it was the last version agreed on. I will defend that only as being a little less formal than "thus". And only because it sounds like we're writing shakespeare. Thus and therefore have the same meaning. If that should be reverted or Trypto objects I wont revert or argue further on that issue.--Amadscientist (talk) 05:41, 23 December 2012 (UTC)

As a practical matter, the only thing that absolutely must be cited without waiting for it to be challenged is BLP content that does cast the subject a negative light, or even could be viewed as casting the subject in a negative light. While best practice is to cite any claim that is not common knowledge, no one is likely to start handing out pitchforks for not citing unless the topic is particularly controversial and thus subject to constant disputes. I don't see anything that would cause me to object to the proposal. Monty₈₄₅ 07:01, 23 December 2012 (UTC)

No. (this is just a mattter of fact, not an argument against you agreeing with the proposal) There is one requirement and that is that an article have at least one overarching reference or the entire article can be deleted.--Amadscientist (talk) 07:40, 23 December 2012 (UTC)

Being a (apparent) major substitution, I think that there would need to be an analysis of what changes are contained in it (besides moving/organizing existing material) before it can be really evaluated. North8000 (talk) 12:19, 23 December 2012 (UTC)

If we are giving this a logical analysis, it does not address the mother of all conflations contained in WP:burden. This is the conflation between

• Unsourced material and
• Sourced material where there is a claim that the source does not meet wp:rs criteria.

Sincerely, North8000 (talk) 12:29, 23 December 2012 (UTC)

• You are all still focused on language, when I am addressing structure and concept...What inspired my proposed re-organization was the realization that (in our current version) BURDEN is really just one single sentence (The burden of evidence lies with the editor who adds or restores material, and is satisfied by providing a reliable source that directly supports the material).

Everything else in the section really has nothing to do with WP:BURDEN... it relates to WP:CHALLENGE (when and how to challenge material). I think it would be helpful to separate the two concepts... 1) one section (WP:CHALLENGE) focused on what a challenger should do (how to issue a challenge)... and another section (WP:BURDEN) focused more on the challengee should do (how to respond to a challenge).

In the current version, we place BURDEN before CHALLENGE - which seems backwards to me (it does not make sense to discuss how to respond to a challenge before we have discussed what constitutes a challenge and how to issue one).

Does my proposal resolve all the issues? No... but I think it will separate them into bite sized chunks that will make them easier to resolve. By reorganizing the section first, separating CHALLENGE from BURDEN, we can then address the issues with each sub-section... without having those issues overlap (and I think overlap is one of the reasons why we have not been able to reach a consensus). Blueboar (talk) 14:11, 23 December 2012 (UTC)

Blueboar, I can understand your frustration, but there's a reason why all of us seem to be focused on language rather than structure and concept. That's because the structure and concept invariably will depend upon language. I agree with you that there are three distinct and identifiable concepts here. And I agree in principle that, if those concepts were somehow muddled together by the existing language or structure, that would be a problem in need of correction. But I'm not convinced that the single-section structure inherently creates such a muddle, nor can I picture whether a three-section structure would clarify anything until I know what the language would be in such a structure. --Tryptofish (talk) 15:22, 23 December 2012 (UTC)

OK... I can see your point about structure not creating the muddle... but I do think restructuring will make it easier to deal with the muddle. It would allow us to better separate concerns and address them one by one. What is happening now is that each of us has a slightly separate concern that we want addressed, but because several concepts are muddled together in the current text, as soon as one person raises his/her concern, the rest of us chime in with: "Perhaps, but that does not address my concern"... thus, we end up trying to discuss five concerns at the same time, and none of them get resolved.

If we separate CHALLENGE from BURDEN, it will allow us to better identify where our concerns actually are (I suspect that most of the concerns discussed so far actually relates to CHALLENGE and not to BURDEN)... and identifying where our concerns actually are will better help to focus our continuing discussions over language, so we can actually achieve something. Blueboar (talk) 16:05, 23 December 2012 (UTC)

Overall I like the idea.

This particular sentence (which really isn't under discussion here, but while I'm thinking of it): . The citation should fully identify the source, and the location within the source (specifying page, section, or such divisions as may be appropriate) where the material is to be found. I think doesn't really belong here. The bit meant to be about page numbers is kind of trivial for this policy, and I'm concerned that it will turn into "You've violated WP:V, because you couldn't figure out how to properly format the citation!" Someone was just at WT:CITE a couple of weeks ago making exactly the claim that he was entitled to revert cited material solely because the bibliographic citation supplied did not exactly match the style being used in the article, and despite CITE leading with a "do your best" statement that if you can't figure it out, someone else would fix it for you, that he wasn't going to clean up other people's formatting, so the only appropriate thing to do was to revert the whole thing and if someone wanted it in, then that person could have the BURDEN of proper formatting.

So, yes, proper bibliographic formatting is a beautiful thing, but no getting the formatting wrong absolutely does not change the fact of the material being verifiable and cited. WhatamIdoing (talk) 17:04, 23 December 2012 (UTC)

Suggest you start a section to discuss making that change in the current version of policy. As part of a change in a large part of the policy, such as this proposal, it may not get enough attention and consideration, for and against. Note that Blueboar represented this proposal as a "reoganization (mostly just moving the existing wording around, but with some new language to link it together)". Anything more would be beyond this representation of the proposal. BTW, there seems to be some concern that the proposal already goes beyond how it was represented, for example my previous comment and North8000's comment, "Being a (apparent) major substitution, I think that there would need to be an analysis of what changes are contained in it (besides moving/organizing existing material) before it can be really evaluated." --Bob K31416 (talk) 17:44, 23 December 2012 (UTC)

(sigh) This is why I asked you all to focus on presentation and not language. To clarify: I AM NOT proposing the specific language I have put in the green box above... I am proposing the structural change and merely created an example to show how it might look. My hope is that the proposed structural change will help us to better focus on the specific linguistic changes that we think are needed. Blueboar (talk) 18:05, 23 December 2012 (UTC)

Blueboar I think this is directly tied into what I have written above in #A reasonable challenge. I do not think that separating challenge and burden into different sections is desirable. It has been tried and led to problems which is why the were merged back together recently. I do however think that re-separating Burden and Challenge into two paragraphs is desirable because it would accomplish what you are suggesting without the misunderstandings that were created by separate sections (see this example from 20 September 2012). -- PBS (talk) 19:16, 23 December 2012 (UTC)

I am certainly not insistent on them being separate individual sub-sections (although I think it makes sense that way)... if people prefer formatting it as separate paragraphs within one section that would be fine with me.

That said... the Sept. 2012 example you give, is problematic in that it includes the structural issue I am attempting to resolve with my re-organization... the second paragraph starts off with the standard BURDEN statement:

• "The burden of evidence lies with the editor who adds or restores material"

However, the rest of that paragraph:

• "You may remove any material lacking ... aware of how the BLP policy applies to groups."

has nothing to do with that first sentence. The rest of the paragraph is logically part of CHALLENGE (directed at the person removing information) not BURDEN (which is directed at the person responding to the removal), and should come before BURDEN.

So, whether we format CHALLENGE and BURDEN as individual sub-sections or simply as separate paragraphs within one section... these two policy provisions (represented by policy shortcuts) need to be separated and re-organized. Blueboar (talk) 20:25, 23 December 2012 (UTC)

• support this type of separation and clarification. -- TRPoD aka The Red Pen of Doom 19:21, 23 December 2012 (UTC)

Arbitrary break

There is currently a good discussion ongoing above about the possible rearranging of the Burden policy prose. It is WAY to soon to start changing the actual text on the policy page. No consenus has formed at all and the discussion has not seen everyone commenting, even make an decision to support of oppose yet. Please do not change the prose on the artcile page until discussion winds down on that part.--Amadscientist (talk) 02:28, 24 December 2012 (UTC)

Sorry, but that call for no one to make BOLD edits sounds a little paradoxical to me, in context. Not that I'm making any such edits myself. But if it means that you have come to agree with me that, for a core policy page, we should have consensus in talk before making disputed changes, well then I think that's a good step forward. --Tryptofish (talk) 15:58, 24 December 2012 (UTC)

We don't even yet have a definite proposal. Blueboar's text above has repeatedly been stated to not be a proposal. SpinningSpark 11:53, 24 December 2012 (UTC)

It is a proposal... but it is a proposal about structure and not a proposal about some specific text. Blueboar (talk) 13:01, 24 December 2012 (UTC)

"[C]ome to agree with me that, for a core policy page, we should have consensus in talk before making disputed changes". I don't think it is a matter of either agreeing with you to believe this is a core policy page and that edits should be confined to where there is consensus...just that you seemed to have seen consensus when you removed "i.e." (that I agreed with and defended extensively) or the addition of something like "thus" although "therefore" means the same thing but doesn't sound as formal to me. The problem with myself and at least one other was you suddenly decided that there was no consensus, even though you had already believed there was when making edits to the "core policy page" yourself. Lets just move forward...shall we.--Amadscientist (talk) 22:14, 24 December 2012 (UTC)

Another look at the structural change proposal

OK... Keeping things in one compact section per PBS's comments... without changing a single word of the current text... and just rearranging the structure... this is what the section would look like if my proposal is accepted:

"WP:PROVEIT" redirects here. For the editing tool, see User:ProveIt GT.

Attribute all quotations and any material challenged or likely to be challenged to a reliable, published source using an inline citation. Cite the source clearly and precisely (specifying page, section, or such divisions as may be appropriate). The citation must clearly support the material as presented in the article. See Citing sources for details of how to do this.

Any material lacking a reliable source directly supporting it may be removed. Whether and how quickly this should happen depends on the material and the overall state of the article. Editors might object if you remove material without giving them time to provide references; consider adding a citation needed tag as an interim step. When tagging or removing material for not having an inline citation, make it clear that you have a concern that the material may not have been published in a reliable source, and therefore may not be verifiable. If instead you think the material is verifiable, it is better to try to provide an inline citation yourself before considering whether to remove or tag it.

However, do not leave unsourced or poorly sourced material in an article if it might damage the reputation of living people, and do not move it to the talk page. You should also be aware of how the BLP policy applies to groups.

The burden of evidence lies with the editor who adds or restores material, and is satisfied by providing a reliable source that directly supports the material.

(Footnotes)

Now... having (re)presented the structural change... I fully expect that such a structural rearrangement will inspire (or even require) subsequent text changes (indeed, I have some ideas on that). However, I would like people to deal with potential text changes as a separate issue... to be hammered out after (if) we reach a consensus on the proposed structural change. Hope this clarifies where I am coming from. Blueboar (talk) 13:52, 24 December 2012 (UTC)

Blueboar, being a structure person, I applaud structure-based efforts. But what you are writing will be confusing to most. You indicated that is just a proposed new structure, yet it has the appearance of being proposed new wording. And it really doesn't discuss structure or structural changes, it leaves those to be derived by the reader out of what looks like new policy wording. Sincerely, North8000 (talk) 14:15, 24 December 2012 (UTC)

Um... I am not sure how I can be any clearer than explicitly stating that I have NOT changed one word of text. But I will try: NOTE: The text used in my (re)proposal (in the green box immediately above) is exactly the same text that appears in the current version of the guideline. NOT ONE WORD HAS BEEN CHANGED. I have merely rearranged the text of the current version (moving the WP:BURDEN sentence that and its accompanying footnote from the beginning to the end of the section, and assigning the appropriate shortcut to each paragraph). I think this is a better structure than the current version. I think the policy should discuss what a challenge is, and what those who make challenges need to do... before the policy discusses what other editors must do to respond to those challenges. Does this clarify what I have done and why? Blueboar (talk) 19:09, 24 December 2012 (UTC)

You weren't asking me ((edit conflict), I think you deleted where you asked that), but the answer is a kind of yes-and-no. From where I am, you've expressed yourself very clearly, and I think I understand you quite well (I don't know about anyone else), so I don't need more explanation, and as with every talk thread here, more explanation tends towards tl;dr. What's really going on isn't that people disagree about this. It's just that there are exactly... a lot... of simultaneous proposals floating on this talk page, and no one (mea culpa) wants to talk about, or even read about, other proposals, until their proposal gets a good response. This talk page, like an ideal gas, has filled all available space. It's overstuffed, its eyes glazed over.

Give it a couple of days. Let stuff marinate. Let heads clear. Nothing has really been rejected. --Tryptofish (talk) 19:27, 24 December 2012 (UTC)

Yeah... as I said above, I see the same trend (everyone focusing on their proposal and not wanting to talk about someone else's. And I am probably guilty of the same thing here... but... I do think that if we attend to the structural concern first, it will better help us to wade through all the other (text) proposals. We can then go through each paragraph in an orderly way, and discuss what changes are desired. In any case, I am off for Christmas eve dinner... and will leave it for the rest of you to think about. Have a merry! Blueboar (talk) 19:38, 24 December 2012 (UTC)

I verified by cut, paste, and diff, that the only change to the text and footnotes was moving the lead sentence to the end. Also, the shortcut box was split into several. I'd suggest titling the proposal something like, "Proposal to move Burden lead sentence to end of section". --Bob K31416 (talk) 22:41, 24 December 2012 (UTC)

• Support "Proposal to move Burden (1)" (as I will call it for now). I think is a constructive change that addresses my specific concern (see above) without muddying the waters with multiple changes. -- PBS (talk)

• Support this reordering and separate shortcuts for each paragraph. The new structure narrates a story in the order that it happens: at first some unsourced content is found, then a challenge is made, and at the end it may or may not be solved with a reliable source. This makes clear what editors should do at each step in that process; and situations that don't match this pattern (such as removed content that was not unsourced to begin with) would not directly fall under this policy, thus requiring further discussion. The result is much better guidance for those reading and interpreting it. Diego (talk) 11:57, 25 December 2012 (UTC)
• Oppose, because it reinforces the status quo without addressing any of the issues under discussion. SpinningSpark 18:54, 25 December 2012 (UTC)

Spinningspark... Not sure I understand your objection. Are you saying we have to address the other issues under discussion first (if so, why?). Or do you think that my proposal would make it impossible to address the other issues? (if so, why?)... or is it something else entirely? Blueboar (talk) 13:50, 26 December 2012 (UTC)

Because it is distracting effort away from the substantive issues, and because of what I said. SpinningSpark 10:45, 27 December 2012 (UTC)

• Neutral as to the structural change per se, but pro forma oppose the inclusion of the "make it clear" sentence. Best regards, TransporterMan (TALK) 14:21, 27 December 2012 (UTC)

FYI, you can discuss removing the "make it clear" sentence at Burden of evidence sentence "When tagging or removing material...". --Bob K31416 (talk) 15:42, 27 December 2012 (UTC)

• About the order in which the three "parts" appear, now that I see it on the policy page, I feel like the "burden" sentence at the end seems to just kind of hang there. Perhaps it would be better to (1) keep the WP:UNSOURCED part where it has been moved, at the beginning, (2) move WP:BURDEN/WP:PROVEIT up to the middle position, and (3) have WP:CHALLENGE at the end. I think it may flow better that way. --Tryptofish (talk) 15:59, 27 December 2012 (UTC)

To my mind, it does not makes sense to place WP:BURDEN before WP:CHALLENGE. The burden to supply sources (the need to "Prove it") really only kicks in if there is a challenge (or the likelihood of a challenge) - so we should explain challenges before we explain how to respond to them. That said, I could see us expanding on BURDEN, so it does not just "hang there". Blueboar (talk) 14:46, 28 December 2012 (UTC)

Given how difficult every little change here ends up being, I would worry that adding to the BURDEN section just to make it longer would just make the drama longer. But, oddly enough, I read it just the other way around than you just did. In other words, I feel like, first, we tell users what they need to do for sourcing, then we explain whose responsibility (burden) that is, and only third do we go into what happens when another user believes that the burden has not been met. Sort of a how-to: what you need to do, why you need to do it, and what happens when you don't do it. --Tryptofish (talk) 14:55, 28 December 2012 (UTC)

Hmmm... I think part of the problem is that the three sections are talking to different people. UNSOURCED is directed at the person initially presenting information answering the question: when do "I" (person 1) need to provide a source ... CHALLENGE is directed at someone who comes across unsourced information, answering the question: when can "I" (person 2) remove unsourced information ... BURDEN is directed at someone who wants to keep challenged information in the article, answering the question: how can "I" (person 3) respond to a challenge. Blueboar (talk) 16:14, 28 December 2012 (UTC)

Well, we are probably getting into angels-on-the-head-of-a-pin territory here, but you've outlined one way to look at it, and I've outlined another way. As a matter of good writing style, it seems to me that you've ended the section with what is really an introductory topic sentence, and that sounds strange, coming after a typical ending sentence. But if you feel strongly that it should be this way, and no one else cares, I'll let it be. --Tryptofish (talk) 18:22, 28 December 2012 (UTC)

Suggest adding sentence to beginning of last paragraph of Burden

Suggest adding a sentence to the beginning of the last paragraph of Burden so that it becomes,

"Sometimes editors will disagree on whether material is verifiable. The burden of evidence lies with the editor who adds or restores material, and is satisfied by providing a reliable source that directly supports the material."

--Bob K31416 (talk) 12:42, 29 December 2012 (UTC)

Or perhaps:...

"When verifiability has been challenged, the burden of evidence lies with the editor who adds or restores material, and is satisfied by providing a reliable source that directly supports the material. The challenger does not need to "prove" that the material in question is unverifiable, those who wish to add, retain or restore the challenged material must demonstrate that it is verifiable, by supplying a source for it.

(suggested additions in italics). I think this (or something like it) would better tie the BURDEN paragraph into the CHALLENGE paragraph. Blueboar (talk) 13:41, 29 December 2012 (UTC)

I support this. (however, I am not sure it would receive any support from those who appear to want to include a requirement that the remover must claim that they do not believe the material is verifiable.) -- TRPoD aka The Red Pen of Doom 13:46, 29 December 2012 (UTC)

I don't see a problem... requiring an edit summary along the lines of "I think this is unverifiable" would have no impact on BURDEN ... there is a huge difference between asking editors to claim that something is unverifiable, and requiring them to prove that it is unverifiable. Blueboar (talk) 14:05, 29 December 2012 (UTC)

I might be able to support this if it were done in conjuntion with a clear statement that random large-scale tagging/deletion was considered disruptive, or deletion in the full knowledge that RS actually existed. But as it is, the wording seems to be designed to have precisely the opposite effect and encourage this behaviour. SpinningSpark 13:59, 29 December 2012 (UTC)

That is an issue that should be addressed in the CHALLENGE paragraph... as it goes towards "what constitutes a valid challenge". I don't think it belongs in the BURDEN paragraph. Blueboar (talk) 14:28, 29 December 2012 (UTC)

That might be so, but I still can't support this in isolation. Kind of speaks against the idea of splitting in the first place. SpinningSpark 16:02, 29 December 2012 (UTC)

The version I suggested is to help settle disagreements and includes the case where there is disagreement about the validity of the challenge, but does not presume there will be disagreement after the challenge. Blueboar's version presumes that there will be a disagreement after the challenge is made, which isn't necessarily the case. There may not be any disagreement after the material is challenged because the editor who wanted the addition may then agree with the reasoning behind the challenge. Also, the sentence added at the end seems too repetitive. The version I suggested is shorter and to the point, which is just to give a context for the burden sentence. --Bob K31416 (talk) 14:12, 29 December 2012 (UTC)

Correct... I do assume that there is a disagreement over a challenge. If there is no disagreement, then there is no BURDEN (because no one is arguing that the challenged material should be included). That is the whole point of WP:BURDEN. We created WP:BURDEN to address the issue of what has to happen after unsourced material is tagged or removed. We once had editors who would demand that we "prove" that problematic unsourced information was NOT verifiable ("How dare you challenge what I wrote... you say there is no source? Prove it... prove to me that there isn't a source out there somewhere that says this"). The entire point was to make it clear that this attitude is backwards... that it is not up to those who challenge information to "prove the negative"... it's up to those who want to keep the information to prove the positive (by supplying a source.) If no one wants to keep the information, then BURDEN does not apply. The material is simply removed and no one objects.

As for repetition... if it serves to clarify what was said before, a little repetition is a good thing. Blueboar (talk) 14:36, 29 December 2012 (UTC)

• Red Pen predicted, correctly, that some of us would object to the kind of wording that comes at the end of the text that Blueboar proposed. I've read Blueboar's rationales, (1) that there's a difference between proving and claiming verifiability (there is such a difference, of course, but it's Wiki-lawyer heaven to add language asking editors to parse where the difference is located), (2) that any problems should instead be addressed in the "challenge" section (nope, any problems should not be created in the first place, instead of dividing what had been a single section of the policy page into three sub-sections, then creating new problems in order to make the three subsections appear to work, and then trying to solve those new problems – paging Rube Goldberg!), and (3) assuming what Bob K correctly says should not always be assumed (again, only to justify the three-part ordering). Well, you can see what I think of those arguments, sorry. At this point, the, um, burden is on those who want to propose adding new language to the "burden" subsection, to make it able to satisfy the, um, challenges that I and others have raised.

(Inserting comment - Re: (1) that there's a difference between proving and claiming verifiability)... that isn't what I was talking about... I was talking about Unverifiablity... ie the LACK of verifiability. It is possible to prove that something is verifiable (by providing a source), It is NOT possible to prove that something is unverifiable. So, while both sides in a dispute over verifiability can make all the claims they want in edit summaries... the actual BURDEN is on those who claim the information actually is verifiable to PROVEIT. Blueboar (talk) 23:54, 29 December 2012 (UTC)

Duly noted, thanks. But the balance of the argument remains the same. --Tryptofish (talk) 17:53, 30 December 2012 (UTC)

• I'm OK with Bob K's simpler first sentence, proposed at the top of this talk thread, as a third choice. As a second choice, I'm also OK with just the opening phrase of Blueboar's counter-offer ("When verifiability has been challenged,...") Either of those would improve the flow, and they probably could be combined in a couple of ways. But my first choice would be, much more simply, to change the order, so that "burden" comes in the middle, before "challenge", which is what I said in the sub-thread directly above this one. Look: the fact that we now have a couple of editors all recognizing that the single sentence of the moved "burden" looks lonely where it got moved, is evidence that we recognize that, at a minimum, it's bad writing style to leave it hanging there. --Tryptofish (talk) 15:29, 29 December 2012 (UTC)

• The purpose of my proposal was to set the context for the Burden sentence, not to explain it. I think that the question of whether or not the Burden sentence needs explanation, should be discussed in a separate proposal.
  

Getting back to the purpose of my original suggestion, and considering the previous comments related to it, here's a version that incorporates Blueboar's wording that makes a connection to Challenge.

"When verifiability has been challenged, an editor might dispute the claim that the material may not be verifiable. The burden of evidence lies with the editor who adds or restores material, and is satisfied by providing a reliable source that directly supports the material.

--Bob K31416 (talk) 18:37, 29 December 2012 (UTC)

Although I continue to believe that it would be best to solve the "burden context" issue simply by changing the order, I agree with your approach if the order remains with "burden" at the end. There's any number of permutations of how we could word it. I agree with you that the first issue is context, after the "challenge" material, and that explanation, if needed, is really a separate question. --Tryptofish (talk) 19:13, 29 December 2012 (UTC)

Thanks. We're apparently in agreement on the usefulness of context for the Burden sentence.

Reviewing my recent version with "challenged", I have to say that I much prefer what I originally had, i.e. ""Sometimes editors will disagree on whether material is verifiable." And there hasn't been any comment specifically on the revised version. After the recent discussion, I'm not sure where everyone stands right now regarding my original suggestion; so I'll make the change on the project page and see what happens. --Bob K31416 (talk) 16:46, 30 December 2012 (UTC)

I can certainly live with that change, but I find it frustrating that we haven't, instead, gone with what I suggested as a better solution. --Tryptofish (talk) 17:43, 30 December 2012 (UTC)

Thanks. I'll open up a new section below where the subject of moving the Burden sentence can be discussed. --Bob K31416 (talk) 18:09, 30 December 2012 (UTC)

Moving Burden sentence

As mentioned above, I opened this section to discuss moving the Burden sentence.

Tryptofish, Could you offer (or repeat) a specific suggestion for how you want to change the section by moving the Burden sentence? --Bob K31416 (talk) 18:09, 30 December 2012 (UTC)

Yes. It's definitely a case of "repeat", but I do see how it's tough to keep track of everything in these discussions! My suggestion – with or without the new sentence that Bob K added at the front of BURDEN – is to change the order of the sections, from:

WP:UNSOURCED, WP:CHALLENGE, WP:BURDEN, to:

WP:UNSOURCED, WP:BURDEN, WP:CHALLENGE.

In one of the talk sections above, Blueboar and I discussed how the flow of ideas moves in each of these versions. I think that we established beyond a reasonable doubt that it's possible to have a clear logic either way – but neither one of us convinced the other that one way is more or less logical than the other. Given that equivalence, I would argue that it's just easier to read it the "to" way, by avoiding having a single "burden" sentence just hanging there at the end. That's less of an issue with Bob K's added sentence, but it still makes some sense, and it becomes urgent in the event that anyone objects to Bob K's added sentence. --Tryptofish (talk) 19:15, 30 December 2012 (UTC)

I looked at the discussion you had with Blueboar and I saw that your initial motivation for the change was that the Burden sentence just hung there. As you noted here, that's been somewhat mitigated by the recently added sentence. Considering the rest of the discussion, it looks to me that the Burden paragraph should stay at the end. As you mentioned, there are various ways to look at it, but so far it doesn't look to me that there is a preponderance of reasons to move the Burden paragraph. --Bob K31416 (talk) 04:41, 31 December 2012 (UTC)

Well, it should come as no surprise to you that I disagree. I think there's a big difference between "somewhat mitigated" and actually fixed. It seems to me that the discussion is far from resolved in that way, more like it's just at a temporary impasse with relatively few participants. And as I noted in the discussion below about renaming the section, I'm beginning to realize that there's a potential issue with making the concept of burden less prominent than it ought to be, something that is perhaps made worse by relegating it to the end. --Tryptofish (talk) 15:19, 31 December 2012 (UTC)

As you stated previously, your proposal to move the burden paragraph is compatible with the added sentence. What we are trying to decide here is whether to accept your proposal to move the burden paragraph. There is no rush to come to any judgement, so we can wait to see if there is any support for your proposal. --Bob K31416 (talk) 17:08, 31 December 2012 (UTC)

Exactly. --Tryptofish (talk) 18:30, 31 December 2012 (UTC)

Anchors

I just noticed something else. When changing the positions of the three shortcut boxes in the section (WP:UNSOURCED, WP:CHALLENGE, and WP:BURDEN), no one changed the fact that all of the shortcuts still go to the top of the section. If we end up committing to the new order in which the sub-sections appear, we really ought to use "anchor" templates to point each shortcut to the place where it applies. --Tryptofish (talk) 19:25, 30 December 2012 (UTC)

Good catch... I agree. Blueboar (talk) 15:09, 31 December 2012 (UTC)

Renaming Burden of evidence section

A few days ago Blueboar suggested renaming the Burden of evidence section. Perhaps it's time to consider various possibilities for the new name or whether it should be renamed. Suggestions? --Bob K31416 (talk) 04:59, 31 December 2012 (UTC)

P.S. The previous name of the section was "When a reliable source is required".[1] --Bob K31416 (talk) 14:34, 31 December 2012 (UTC)

The previous name would work for me... an alternative might be: "Unsourced material" Blueboar (talk) 15:07, 31 December 2012 (UTC)

Are we, perhaps, overly deprecating the concept of a burden if we do this? After all, a lot of the pushback from editors (not me) who have been concerned about the recent discussions is that we are at risk of watering down the assignment of the burden (to the editor who adds the material) too much. --Tryptofish (talk) 15:13, 31 December 2012 (UTC)

That can be addressed by splitting off the Burden of evidence paragraph to its own section, for example:

Splitting off Burden of evidence to its own section

When a reliable source is required

---

"WP:PROVEIT" redirects here. For the editing tool, see User:ProveIt GT.

Attribute all quotations and any material challenged or likely to be challenged to a reliable, published source using an inline citation. Cite the source clearly and precisely (specifying page, section, or such divisions as may be appropriate). The citation must clearly support the material as presented in the article. See Citing sources for details of how to do this.

Any material lacking a reliable source directly supporting it may be removed. Whether and how quickly this should happen depends on the material and the overall state of the article. Editors might object if you remove material without giving them time to provide references; consider adding a citation needed tag as an interim step. When tagging or removing material for not having an inline citation, you ought to make it clear that you have a concern that there may not be a published reliable source for the material, and therefore the material may not be verifiable. If instead you think the material is verifiable, it is better to try to provide an inline citation yourself before considering whether to remove or tag it.

However, do not leave unsourced or poorly sourced material in an article if it might damage the reputation of living people, and do not move it to the talk page. You should also be aware of how the BLP policy applies to groups.

Burden of evidence

---

Sometimes editors will disagree on whether material is verifiable. The burden of evidence lies with the editor who adds or restores material, and is satisfied by providing a reliable source that directly supports the material.

We can take this one step at a time and first decide what to do regarding renaming the present section. If someone brings up the issue about deprecating, I think their concerns can be alleviated by discussing the split off.

An alternative way to proceed is to first go for the split off, which would consequently require a renaming of the original section. --Bob K31416 (talk) 15:47, 31 December 2012 (UTC)

Hmmm, perhaps. But isn't this going full circle from where we started out, before the very top of this talk page discussion? --Tryptofish (talk) 15:59, 31 December 2012 (UTC)

The proposal at the top of this discussion looks very different to me.[2] The main similarity seems to be that the past proposal has a separate subsection for Burden of evidence and the present possible split off has a separate section for it. In any case, we can simply continue with the question of just renaming the section for now. --Bob K31416 (talk) 16:18, 31 December 2012 (UTC)

Strike that. I should have said: Wikipedia talk:Verifiability/Archive 59#Proposed change?. --Tryptofish (talk) 16:28, 31 December 2012 (UTC)

Those versions are very different too, so my previous comment still applies. --Bob K31416 (talk) 16:36, 31 December 2012 (UTC)

That's OK. Still looks like full circle to me. --Tryptofish (talk) 16:46, 31 December 2012 (UTC)

• "Reliable sources requirement" does it. Alanscottwalker (talk) 15:27, 31 December 2012 (UTC)

• I'm starting to think that it may be a problem if we change to a title that does not include, at least in part, the exact phrase, "Burden of evidence". --Tryptofish (talk) 16:35, 1 January 2013 (UTC)

To those reading this comment of Tryptofish's, please see my comment above that begins, "That can be addressed by splitting off the Burden of evidence paragraph to its own section..." --Bob K31416 (talk) 17:13, 1 January 2013 (UTC)

• Summary — The suggestions so far are

"When a reliable source is required"

"Unsourced material"

"Reliable sources requirement"

"include, at least in part, the exact phrase, 'Burden of evidence' ".

--Bob K31416 (talk) 15:52, 7 January 2013 (UTC)

• Considering the above suggestions, would there be any objections to changing the section heading from "Burden of evidence" to the following?

"Reliable sources requirement and the burden of evidence"

--Bob K31416 (talk) 15:52, 7 January 2013 (UTC)

I kind of object, on the grounds that it's too verbose. I think that a section rename probably rises to the level where it's not a matter of "any objections?", especially when the talk page has become so tl;dr. If you want to rename it, I think you may really need to open an RfC. --Tryptofish (talk) 01:18, 8 January 2013 (UTC)

Well, since you object, and since I was following up on Blueboar's suggestion to rename the section, I'll let someone else follow it up if anyone is interested. Regards, --Bob K31416 (talk) 02:53, 8 January 2013 (UTC)

Concern about appearing to deprecate "Burden"

As I think about the most recent discussions, I'm becoming concerned that we need to be careful not to, in effect, deprecate the long-standing consensus that the "burden of evidence lies with the editor who adds or restores material". There has already been discussion (a lot) about the sentence, "When tagging or removing material for not..." Myself, I think it's an important improvement, but I recognize that some other editors have wanted to delete it entirely, on the grounds that it might (in their opinions, not mine) weaken editors' ability to invoke the "burden". If, on top of that, we delete "Burden of evidence" from the section header, and/or continue to leave the "burden" language in a short paragraph that is now at the bottom of the section, I think those things may further push the balance to where it no longer reflects widespread community consensus. --Tryptofish (talk) 16:35, 1 January 2013 (UTC)

For those reading Tryptofish's message, please see my message of 15:47, 31 December 2012 in the previous section which begins with "That can be addressed by splitting off the Burden of evidence paragraph to its own section..." --Bob K31416 (talk) 17:07, 1 January 2013 (UTC)

Fair enough. (And then look back at Wikipedia talk:Verifiability/Archive 59#Proposed change?.) --Tryptofish (talk) 17:19, 1 January 2013 (UTC)

Visual alignment of shortcuts

Another problem also occurs to me. Right now, the shortcut box for WP:BURDEN and WP:PROVEIT is not really alongside the text to which they refer, but instead, alongside the last paragraph of the WP:CHALLENGE section. That's confusing to users who come here for the first time. I realize that we don't want the shortcut box to spill over into the section below, but I think that editors who want the BURDEN section to remain at the end really need to come up with a better way to handle this issue. --Tryptofish (talk) 16:35, 1 January 2013 (UTC)

Here's the two versions for comparison: Version 1, Version 2. The one that is currently on the policy page is Version 2. Version 1 is an earlier one with the box spilling over into the next section.

Here's the version if the Burden paragraph is moved as Tryptofish suggested. Version 3 --Bob K31416 (talk) 17:34, 1 January 2013 (UTC)

Thanks for creating the third version, but even if we don't reorder the sentences as I hope for, we still should do something about the alignment. --Tryptofish (talk) 20:21, 1 January 2013 (UTC)

Regarding the alignment of the boxes in these three versions, I think Version 2 has the best alignment IMO. I guess you prefer another alignment version which is ... (?) --Bob K31416 (talk) 22:14, 1 January 2013 (UTC)

Actually, I think the, um, burden, is on you and Blueboar to figure out how to fix it. --Tryptofish (talk) 22:45, 1 January 2013 (UTC)

Either of these alignments are fine with me... the key is for the shortcut to stay next to the appropriate paragraph. The alignment problem stems from the simple fact that BURDEN is only two short sentences long, which means the shortcut box actually takes up more lines of print than the text it points to. Is there a way to re-size a shortcut box to make it smaller? (I know we can do this with images, but I don't know about short cut boxes). Blueboar (talk) 22:31, 1 January 2013 (UTC)

But the problem is that neither of those alignments works. --Tryptofish (talk) 22:45, 1 January 2013 (UTC)

Oh, all right, I won't leave you guys wondering what to do. Perhaps, regardless of the order in which the paragraphs occur, the shortcuts should simply be moved back together into a single box at the top of the section. We can still have the text flow in whatever order we decide, but we don't need to be creating sub rosa sub-sections with shortcut boxes instead of section headers. --Tryptofish (talk) 22:52, 1 January 2013 (UTC)

PS: There's no way to re-size the shortcut box (I looked earlier). But if we combine them all into a single box, we can still use #Anchors to point the redirects wherever we want. --Tryptofish (talk) 22:54, 1 January 2013 (UTC)

Any objections to implementing this? --Tryptofish (talk) 23:51, 5 January 2013 (UTC)

If you mean implementing a single box, I have no objection.--Amadscientist (talk) 01:27, 6 January 2013 (UTC)

Yes, that's what I meant (along with using anchors). Thanks. I'll give it another day, and if there are no objections, I'll do it. --Tryptofish (talk) 23:28, 6 January 2013 (UTC)

 Done. --Tryptofish (talk) 01:26, 9 January 2013 (UTC)

Unsourced Section

Hi I deleted a section from the Orange Order article which had been tagged for citation for almost five years. This has been restored by a user, who when I pointed out that the "burden of evidence lies with the editor who adds or restores material", they rather than cite their sources has asked me to bring the issue up here, any thoughts? 188.29.86.18 (talk) 16:10, 2 January 2013 (UTC)

I looked at the link you gave, and I think that they were reacting to the fact that you deleted such a large quantity of material. Although you are correct that you do not have the burden of evidence, it still seems kind of improbable that everything in that section would truly be unverifiable, just because there's a template indicating the need for references in the section. Here's my advice: go to the article talk page, and try to list the specific sentences or facts in the section, to which you object. You may be able, then, to get agreement about how to either source or remove those specific claims. --Tryptofish (talk) 16:22, 2 January 2013 (UTC)

To give an example... I have tagged a few specific statements that (to me at least) need to be cited. To take this to the next level, what needs to be cited (throughout the article) are statements that tie the Orange Order to political trends and events. A statement along the lines of "The Order played a key role in <X event>" may well be accurate... but it needs to be supported with an inline citation to a source that says the order was heavily involved in X event. Blueboar (talk) 14:36, 5 January 2013 (UTC)

There's a similar discussion at ANI. A number of editors mistake 'lack of inline citation' with 'unverified.' As Uncle G nicely explains, routine facts that can easily be verified (or are done so by other sources) and simply don't have an inline is not a license to blank sections. Going through and identifying possibly controversial or unverified information is fine, and CN does well to do that. But in this instance it was a full section removal. It looks like the possibly unverified (although from what I can tell nobody's really arguing that they're biased or wrong) lines have been noted. Shadowjams (talk) 05:57, 8 January 2013 (UTC)

At the same time, there are some of us who believe that there is an obligation to Assume good faith when unsourced material is removed, even without explanation except when there is behavioral evidence, probably strong or even compelling evidence, to suggest that we should not AGF. The discussion above at Wikipedia talk:Verifiability#Some thoughts with an example probably best summarizes the position of those of us who feel that way. I may be unfairly maligning him, or putting words in his mouth, but I suspect that Uncle G would take the position that we should not ever engage in that AGF. I disagree with him and would argue that WP:BURDEN as presently written does not support his position. At the same time, I must admit that I do find section blanking troubling for a couple of different reasons. On the other hand, I am far more prepared to AGF and object to any inquiry into the motives of the deleting party without very strong evidence to the contrary when the material which is being removed has been fact-tagged (or various equivalents thereof) for a reasonable period of time, as was the case here. On my third hand, I think that controversies over this kind of removal are dealt with by the fact that you can't edit war over them and that WP:CONS#No consensus says, in effect, that if the removal of long-established material is contested then the removing editor must either get consensus for the removal or the material must stay in. (The opposite result might obtain if the material being removed was newly added, however.) Best regards, TransporterMan (TALK) 15:50, 9 January 2013 (UTC)

Straw poll: How would you interpret "widely regarded" and "regarded by many"?

Moved from Wikipedia talk:Manual of Style

If a Wikipedia article gave an opinion as "widely regarded", how would you interpret that? That most of the world's commentators agree? Or only some but they are widely distributed? How about "regarded by many" or "regarded by some"? I am trying to phrase something which is held by roughly half the world's experts (anything from 25 to 75 percent, no one really knows and the WP authors disagree, but there's a good chance it's below 50%). (I don't want to give away what it's about because I want a neutral opinion, but if you're curious, check my edit history). Adpete (talk) 10:09, 6 January 2013 (UTC)

I would interpret them as obvious weasel-wording if they didn't quote a reliable source as to such a claim. — SMcCandlish  Talk⇒ ɖ^⊝כ^⊙þ Contrib. 11:26, 6 January 2013 (UTC)

Sorry, I should have made it clear that this applies to the lead. There are a number of reliable, sourced claims for X, and there are also a number of reliable, sourced claims for the main alternative Y. The question though, is how to word this (in a single sentence) in the leads of articles X and Y. Adpete (talk) 11:45, 6 January 2013 (UTC)

This is really more of a WP:NPOV issue than a WP:V issue (and you might want to ask at WP:NPOVN). When it comes to the best way to phrase a statement as to the relative strength of contrasting opinions, a lot depends on the specific subject, and the specific sources (the reputation of the sources would need to be examined to know how much WEIGHT to give them). What I am getting at is this: We here on a policy page can not advise you in a vacuum... we do need to know which articles you are actually talking about, and examine the specifics.

That said: If (and this is a significant "if") the sources are of relatively equal WEIGHT, then phrases such as "widely regarded", "regarded by some", etc. are indeed weaselly. It is better to be specific, and list a few of the more prominent adherents of each view ("Scholars such as A, B and C believe <view X>, while scholars such as D, E and F disagree, and believe <view Y>"). Blueboar (talk) 14:05, 6 January 2013 (UTC)

• Blueboar is absolutely correct. In the specific case Adpete is talking about, my advice is to remove "is considered by many to be" and replace it with "has been called", which is factual, neutral, and informative without making any value judgments.—S Marshall T/C 17:52, 6 January 2013 (UTC)
• I agree with Stanton. A Wikipedia article should never be in a position to give an opinion as "widely regarded". Given any number of sources that say x or y is "regarded as the most [superlative] [poet or whatever]", we would still not be in a position to declare conclusion of any survey with any authority. The best we can do is to cite a source or an author as having said x or y is regarded as the most superlative [whatever]". Anything else is just weasel wording. -- Ohconfucius ^{ping / poke} 14:22, 7 January 2013 (UTC)

Well... we do put some conclusions in wikipedia's voice... when a significant, overwhelming majority of reliable sources all say the same thing. In such cases, I think it is occasionally appropriate to use weaselly words like "widely regarded". We routinely do this in regards to fringe theories (using sentences like: "The theory that X is Y is widely regarded by the scientific community as a Fringe theory"), even when there is no single source that explicitly states that the theory is "widely regarded" as being fringe. Blueboar (talk) 15:03, 7 January 2013 (UTC)

This falls in the huge overlap between "summarization" (which is the backbone of writing articles) and "WP:OR/Synthesis". For those "gray area" cases, the de facto standard is "if nobody objects, its summarization, if somebody objects, it's WP:OR/Synthesis". While that system has its flaws, I think that it works in this case. North8000 (talk) 16:02, 7 January 2013 (UTC)

I am lining up here with the weasel-word crew. I am beginning to revert weaselly worded additions in articles without definitive citations (in which case weasels aren't needed anyway). I'm loving every minute of it because I ask for a citation or a rephrasing if re-added. Secondly, when it is said above: "widely regarded" , the immediate retort is "by whom?" Fylbecatulous talk 17:41, 7 January 2013 (UTC)

I think the phrase should be used with caution and most cases probably avoided due to the weasel word argument. Nevertheless in particular context it may be appropriate in summarizing description. Say for instance a line like "The TV series Veronica Mars is widely regarded as teen noir piece". Now if the author has sampled the national/international reviews on the series and many of them indeed point out the teen noir aspect, I'd say such a line is justified, in particular since the alternatives are imho worse. Because as an alternative you can simply describe a single source labeling Veronica Mars as teen noir, though that is factually accurate it is is nevertheless somewhat misleading, as it is completely unclear to the reader that in fact many sources call it teen noir. The other option to give the full lists of all the sources doesn't allow any decent writing. So the best way here might be to use the "widely regarded" phrase and put the list (or part of it) into the footnote/inline citation.--Kmhkmh (talk) 01:58, 8 January 2013 (UTC)

Thanks for the responses. The issue is who is the best chess player of all time. As with any sport (ok chess isn't really a sport, but it's a comeptition) it's a matter of opinion, but we can cite experts. The main contenders are Bobby Fischer and Garry Kasparov. Since asking the question I've looked at the leads of a few other sportsmen often regarded as best in their sport (Wayne Gretsky, Jack Nicklaus, Don Bradman, Michael Jordan, Pele, etc) and what they do is have a supporting reference(s) in the lead. So I've done that at the Kasparov lead (easy, because there's a reliable source which says "most experts" think he was the best) and am working out how best to do that at the Fischer lead. This is harder, because there doesn't seem to be a reference saying "many" or "most" say Fischer was the best, just a few scattered experts saying they think he was the best. They are cited in the article, the question is how to summarise it in the lead. Adpete (talk) 23:33, 8 January 2013 (UTC)

In that case "widely regarded" may be problematic at when they are several contenders for the best player and the media opinion is split between them. I would use "regarded by some" instead or probably even better use "(widely) regarded as one of the best players" instead of "(widely) regarded as the best player".--Kmhkmh (talk) 23:46, 8 January 2013 (UTC)

Subjective opinion doesn't belong in the lede. In the snooker articles we limit it to achievements, stats and hard fact. Peer standing can be covered, but opinions should be attributed directly to the person stating it, not some unidentifiable mass. Even if you can source the claim "Kasparov is widely considered the greatest", that is still only the opinion of one person. Betty Logan (talk) 23:51, 8 January 2013 (UTC)

Exactly. The lede should only define facts about the person or subject. Opinions with who, what, where, when and why ツ definately appear in an appropriate section. The caveat however: all "opinions", no matter how awesome the sources, remain in that tenuous grey area of POV. Difficult to pin down to prove. It's more stating the facts and thus allowing the reader to draw his/her own conclusion as to "who's on first". Fylbecatulous talk 13:16, 9 January 2013 (UTC)

Self-conflicting wording in wp:burden

The rule isn't just that it be sourcable (and sourced if challenged), it is that it be sourcable/sourced to a wp:reliable source. The most common use of wp:burden to remove material is to remove sourced material, usually with an assertion that the source is not a wp:rs. Any operative statements that just refer to "unsourced/uncited" material (e.g. "when tagging or deleting material for not having an in-line citation") are thus in conflict with this. I tried fixing one of them but got reverted, which is fine....BRD, but they should get fixed. North8000 (talk) 11:39, 6 January 2013 (UTC)

I have been thinking about this... what I think is needed is a short section or paragraph on "Challenging cited information". We should lay out that this is a two step process...

STEP ONE - One challenges the source (as distinct from the information). The form that this challenge takes is similar to other challenges (ie there is a judgement call between querying the source on the talkpage, using an appropriate tag, or removing the source completely). The key is that, at this stage, the challenge is focused on the source, not the text.

STEP TWO - If the sourcing issue is not resolved after a reasonable time has passed, then (and only then) should one challenge the the text (on the grounds that no reliable source has been provided, and the material is potentially unverifiable). Blueboar (talk) 14:36, 6 January 2013 (UTC)

Excellent idea. Sincerely, North8000 (talk) 16:53, 6 January 2013 (UTC)

... except that it's perfectly OK for non-contentious material to be sourced to non-RS'es. For example, the fact that a person is mayor of a town could be sourced to the town's own website, a primary source. Jclemens (talk) 17:00, 6 January 2013 (UTC)

Um... A primary source can be reliable (indeed in the case you present, the town's website would probably be the most reliable sources possible). Blueboar (talk) 17:46, 6 January 2013 (UTC)

... but only an idiot would challenge that. WP:BURDEN would be very easy to phrase correctly if WP:EDITORIAL JUDGMENT wasn't still a redlink ten years into the project. Because it is still a redlink, it's practically impossible to phrase correctly and good attempts lead to ever-increasing amounts of text bloat. I think we should return to the just-post-RFC phrasing and write these more recent additions up as a separate guideline.—S Marshall T/C 17:44, 6 January 2013 (UTC)

North, the sentence that has had so much digital blood spilt over it the last few weeks is (in part) an attempt to address this issue, by asking challengers of material to communicate that they are challenging it because they think it isn't verifiable, as opposed to challenging it because they think the source isn't reliable. I realize that this one sentence does not completely solve everything, and I'm personally receptive to saying more. At the same time, I can observe that it's difficult to arrive at a consensus between editors who have the concerns that you've expressed here, and editors who feel strongly that we must not weaken BURDEN, and who feel that such caveats would do just that. --Tryptofish (talk) 23:35, 6 January 2013 (UTC)

Actually some very very minor changes (such as the one I attempted) would solve the conflict. Sentences offering advice/recommendations etc. to persons tagging or deleting material should merely have wording that does not accidentally artificially limit the advice to uncited material. And I think I was reverted only due to the "sound" not being as good, not due to objections on the substance of it. Sincerely, North8000 (talk) 15:52, 7 January 2013 (UTC)

I can sort of see what the intention is here, but I'm not sure that kind of detailed procedure (e.g. a 2-step process) is appropriate for a top-level WP policy. For instance, the BURDEN policy states "Any material lacking a reliable source directly supporting it ...". That sentence alone gives four reasons for challenging: (1) no source; (2) source not reliable; (3) doesn't support; (4) doesnt directly support. That seems like a pretty plain sentence, and I don't think a lot of text should be spent distinguishing the four situations and giving unique guidance for each situation. Too much detail in a top-level policy causes all kinds of unintended consequences down the road. As a compromise, maybe the steps to take in the various specific situations could be addressed in a guideline, essay, or Help page? --Noleander (talk) 02:12, 9 January 2013 (UTC)

I concur with Noleander. The detailed procedures are not appropriate for a top-level WP policy. But having a guideline, essay, or Help page with the more detailed guidance could be quite helpful when editors get down in the weeds and are looking for additional guidance. N2e (talk) 16:49, 9 January 2013 (UTC)

Which is better?

In the context of the first sentence:-

1. "people reading and editing the encyclopedia", or
2. "encyclopedia users"

From conversation on my talk page I think it's likely that the only person who objects to the simpler and more elegant phrasing is Amadscientist. Is this accurate?—S Marshall T/C 12:06, 8 January 2013 (UTC)

To my mind they are identical, and I don't really like either of them. I would prefer something more along the lines of: "In Wikipedia, verifiability means that published reliable sources exist to support what is written in our articles". Blueboar (talk) 16:25, 8 January 2013 (UTC)

Hmm. Verifiability means more than just articles. "In Wikipedia, verifiability means that our content is supported by published reliable sources"?—S Marshall T/C 17:33, 8 January 2013 (UTC)

Um... WP:Verifiability only applies to Article space... content in other parts of Wikipedia (policy space, user space, even article talk pages) does not need to be verifiable. Also... "content is supported" can be confused with "a source is cited", which is not always required. The key to Verifiability is that a reliable source must exist... a source that can be cited if desired... and must be cited when required. Blueboar (talk) 18:43, 8 January 2013 (UTC)

It includes, for example, lists (which are in article space but aren't necessarily articles).—S Marshall T/C 18:57, 8 January 2013 (UTC)

Hmmm... I consider most lists to be "articles in list format" ... we also have to consider that purely navigational "lists" (such as dab pages and "List of articles on X") don't require Verifiability... (but that open a whole different can of worms). Blueboar (talk) 20:02, 8 January 2013 (UTC)

It also includes outlines, which you may or may not consider to be a specialised kind of navigational list. Generally, I think it's best not to imply that verifiability only applies to articles, because it's possible to quibble about what an article is, so I prefer "content". However, I don't feel overwhelmingly strongly about this, so if it's very close to your heart how about: "In Wikipedia, verifiability means that our articles are supported by published reliable sources"?—S Marshall T/C 20:21, 8 January 2013 (UTC)

I'm thinking that it may be better to leave it as "people reading and editing the encyclopedia". Although a few words longer, it makes it more clear that we aren't just talking about readers, nor just talking about editors, but rather are talking about both. And that actually is how verifiability is supposed to work. Anyway, given all the sturm und stuff over the lead, and how incremental simplification of the text still should take second place to any concerns about consensus, I think it's probably best to let it be. --Tryptofish (talk) 01:14, 9 January 2013 (UTC)

Must say that I like Tryptofish don't see a big problem with the extra words, compared to alternatives.--Andrew Lancaster (talk) 21:43, 9 January 2013 (UTC)

Accusatory intro

Why do we have accusatory language in the "this page in a nutshell" section?

"Other people have to be able to check that you didn't just make things up." is very accusatory. Not everyone, and in fact most people aren't here to just make things up.

I propose something along the lines of, "Other people have to be able to check that information is true."

CTF83! 04:36, 10 January 2013 (UTC)

I hadn't thought of it as accusatory until you brought it up. I just thought it was an attempt at being colloquial. It may have a different effect on different people; it apparently had a different effect on you. In your suggestion, you might want to change "true" to "credible" since the information is only as true as the reliable source can make it. --Bob K31416 (talk) 14:00, 10 January 2013 (UTC)

I hadn't noticed it before, either, but it's an interesting question. Looking at it freshly, it certainly strikes me as colloquial, but I'm not really sure that it's accusatory, so much as getting right to the point. It really assumes that "you" did not make anything up, and just indicates the importance of Wikipedia's way of putting something on the page that prevents any confusion about the fact that something wasn't made up. Once we get to specifying "true", or "credible", or merely "sourced", we get into trying to define something, in the one sentence, that actually takes a lot of the subsequent text to explain. Maybe, though, we should get away from saying "you". We could change "you" to "editors", which would depersonalize it. --Tryptofish (talk) 00:46, 11 January 2013 (UTC)

OK, fleshing out what I just said, how about changing "Other people have to be able to check that you didn't just make things up" to: "People have to be able to check that editors don't just make things up." --Tryptofish (talk) 00:50, 11 January 2013 (UTC)

I would support Bob's suggestion, or Trypto of changing "you" to "editors". CTF83! 02:06, 11 January 2013 (UTC)

• Better yet, depersonalize it entirely. Try this on for size:
  • "Readers need to be able to check that Wikipedia articles are not just made up, but are based on established knowledge."
• Comments, suggestions, love it, hate it? --Jayron32 03:08, 11 January 2013 (UTC)
  • Works for me! CTF83! 03:30, 11 January 2013 (UTC)
  • Huh... I don't really see the "accusatory language" as being a problem but... If others do, I have no real objections to Jayron's language. Blueboar (talk) 16:56, 11 January 2013 (UTC)
  • Info in reliable sources may not be established knowledge, e.g. news media. --Bob K31416 (talk) 01:37, 12 January 2013 (UTC)

• Made a bold edit. Please feel free to add to it or edit as you please.--Amadscientist (talk) 01:54, 12 January 2013 (UTC)

• Works for me. CTF83! 11:21, 12 January 2013 (UTC)

• I made a further change, so please check whether it's OK or not. I changed "verify content" to "trust what Wikipedia says". I think it's self-evident that a verifiability policy is going to require that stuff can be verified, so I'd prefer to avoid the word "verify"; there's also an ambiguity as to whether one verifies content, or how the content is sourced. I think the new wording makes clear our aspirations for the project (albeit not what has already been achieved!). --Tryptofish (talk) 22:38, 12 January 2013 (UTC)

Update: Here's the last edit of the sentence.[3] --Bob K31416 (talk) 01:07, 13 January 2013 (UTC)

Works for me. CTF83! 01:46, 15 January 2013 (UTC)

Sorry, but the last version simply wasn't explanatory enough - the meaning of what "verifying content" is was lost in the process. I've used the version above (by User:Jayron32, I think?) that used "not made thinks up" in the third person, to avoid the original accusatory feeling. Diego (talk) 22:54, 15 January 2013 (UTC)

I agree with you. I wasn't going to make an issue of it, but "verify content" ends up being circular, as I said earlier. Are we verifying content, or verifying that the content has sourcing? I like the change you made much better, because it helps anyone reading it to quickly understand why we have the verifiability requirement. --Tryptofish (talk) 23:05, 15 January 2013 (UTC)

WP:PRIMARY versus WP:SPS and third-parties

This is kind of a philosophical question with no specific example in mind, but it's motivated by the number of times Twitter and Tubmler gets brought up at WP:RSN. Is Twitter an SPS or a primary source, or both? Criteria 2 of WP:SPS says you cannot use a self-published source for claims about third parties. WP:PRIMARY has no such restriction (unless I missed it). So, let's say that the Famous Singer A tweets that they've started recording a new song. This meets the qualifications of SPS. However, what if Famous Singer A tweets that they've started recording a new song with Singer B. Now, it involves a third-party. Let's assume that the information is non-contentious. Technically, it fails criteria 2 of WP:SPS. But if you think of it as WP:PRIMARY, it's acceptable. I have no real example here, but I suspect this sort of sourcing happens all over Wikipedia. But maybe I'm being too anal worrying about this or maybe this is a kind of WP:IAR situation. Thoughts? A Quest For Knowledge (talk) 13:23, 10 January 2013 (UTC)

IMHO structurally, it's a medium (like "pencil and paper") that doesn't inherently have a primary/secondary characterization, but is all wp:SPS.

Structurally I don't think "But if you think of it as WP:PRIMARY, it's acceptable." is correct. Wp:primary includes a self-limitation that says "unless restricted by another policy" and so if wp:sps precludes that use then wp:primary says that it's "permission to use" does not exist for that case. North8000 (talk) 15:38, 11 January 2013 (UTC)

Yes, Twitter tweets are SPS... reliable in limited situations, and not reliable in others.

Note: In the example of Singer A tweeting that he is recording with singer B... we are in a grey zone here... it is both a "claim about himself" (I am recording a song...) and a "claim about a third party" (...with Singer B). However, the claim about Singer B is not that controversial, so we could probably say the tweet was a reliable SPS.

That said, reliability is not the be-all-and-end-all of inclusion... I would probably omit mentioning the tweet on the grounds that it was trivial. We don't know if the song will ever be released (and a lot can happen between recording and release.) So, I would wait until the song actually has been released before I mentioned it in an article (and by that time, we would probably have better, even more reliable sources to rely on). Blueboar (talk) 16:46, 11 January 2013 (UTC)

Twitter usually falls afoul of context problems: 140 characters presumes a lot of context, and we aren't allowed to assume context when examining a source. I use reasoning similar to Blueboar's: if it isn't important enough to show up anywhere but Twitter, it isn't important enough to mention.—Kww(talk) 16:56, 11 January 2013 (UTC)

Well, if it's a major superstar, such as Rihanna and her duet with Britney Spear on S&M (song), it will certainly be reported by secondary reliable sources. But what if it's not a major superstar and it doesn't get covered by secondary sources? Again, I have no specific example in mind. I just get the feeling that this sort of stuff happens all over Wikipedia and am just wondering what we should tell editors if it gets to RSN. A Quest For Knowledge (talk) 04:17, 20 January 2013 (UTC)

Fast tag

I came across an article that was tagged with {{unreferenced}} one minute after it was created.[4] --Bob K31416 (talk) 05:52, 22 January 2013 (UTC)

Well, then create articles with references, or, use your sandbox. Material in the mainspace should be mainspace ready. --Jayron32 05:58, 22 January 2013 (UTC)

What do you think of this guy who started these without references and they became featured articles?[5] [6] --Bob K31416 (talk) 06:43, 22 January 2013 (UTC)

I think you can start articles without references. But then you shouldn't complain when they get tagged for not having them. A tag is an indication of something that needs to be fixed. If the tag is correct (which it is in this case) then fix the problem or leave it for someone else to fix. There's no reason that an article tagged with unreferenced cannot become a featured article, and I never said anything of the sort. What I said was that mainspace articles should be mainspace ready (not featured-ready) and part of that is expecting any inadequacies in said articles to be duly marked by tags. If tags bother you, fix the problem. --Jayron32 06:52, 22 January 2013 (UTC)

May I take it that you didn't give references when you started those two articles because you didn't know any better back then? --Bob K31416 (talk) 06:58, 22 January 2013 (UTC)

P.S. It is a bit humorous that you are acting like I created that article that was tagged, which I didn't, whereas you are acting like you didn't create those two unreferenced articles, which you did. --Bob K31416 (talk) 07:03, 22 January 2013 (UTC)

Nope, I knew both needed references when both were created. They eventually got references. I still fail to see your point here. Could you explain succinctly why indicating that an article without references needs references is a problem? --Jayron32 07:05, 22 January 2013 (UTC)

I simply thought that the situation described in my first message was strange because of the timing, so I wanted to see what editors here thought of it. Thanks for your response. --Bob K31416 (talk) 07:32, 22 January 2013 (UTC)

Update: I looked at the editor's contributions around the time of the edit, and it appears that the editor was working on newly created articles and put similar tags on other articles. Also, it looks like the editor was working on judging whether articles submitted for creation should be created, and was simply trying to encourage sourcing in articles that were accepted for creation but didn't have sourcing. So that solves the mystery for me. --Bob K31416 (talk) 14:46, 22 January 2013 (UTC)

FYI, At the template page {{Unreferenced}} is the advice, "Consider not adding this template to very brief stubs, since anyone visiting the page can see, in a single glance, that it contains no citations." Regards, --Bob K31416 (talk) 19:40, 22 January 2013 (UTC)

Are you trying to make a point? Blueboar (talk) 19:54, 22 January 2013 (UTC)

No, it was FYI as I stated. I'm just trying to exchange info, receive and give. --Bob K31416 (talk) 20:00, 22 January 2013 (UTC)

Which is very helpful, but it looks like we can wrap this thread up now before someone misunderstands someone else again. CarrieVS (talk) 20:14, 22 January 2013 (UTC)

Thanks. --Bob K31416 (talk) 20:25, 22 January 2013 (UTC)

Challenged and removed

Please see Template_talk:Unreferenced_section#Challenged_and_removed. --Bob K31416 (talk) 14:22, 22 January 2013 (UTC)

responded there. Blueboar (talk) 19:28, 22 January 2013 (UTC)

Interwiki

Please add [[el:Βικιπαίδεια:Επαληθευσιμότητα]] (interwiki). Xaris333 (talk) 20:29, 22 January 2013 (UTC)

Credit where due

{{Editprotected}}

Please change the following citation:

Hume, David. An Enquiry concerning Human Understanding, Forgotten Books, 1984; first published 1748, p. 86: "That no testimony is sufficient to establish a miracle, unless the testimony be of such a kind, that its falsehood would be more miraculous, than the fact, which it endeavours to establish; and even in that case there is a mutual destruction of arguments, and the superior only gives us an assurance suitable to that degree of force, which remains, after deducting the inferior."

to:

Hume, David. An Enquiry concerning Human Understanding, Forgotten Books, 1984; first published 1748, pp. 82, 86: "A wise man ... proportions his belief to the evidence. ... That no testimony is sufficient to establish a miracle, unless the testimony be of such a kind, that its falsehood would be more miraculous, than the fact, which it endeavours to establish; and even in that case there is a mutual destruction of arguments, and the superior only gives us an assurance suitable to that degree of force, which remains, after deducting the inferior." In the 18th century, Pierre-Simon Laplace reformulated the idea as "The weight of evidence for an extraordinary claim must be proportioned to its strangeness." Marcello Truzzi recast it again, in 1978, as "An extraordinary claim requires extraordinary proof." Carl Sagan, finally, popularized the concept broadly as "Extraordinary claims require extraordinary evidence" in 1980 on Cosmos; this was the formulation originally used on Wikipedia.

Note that the most important/relevant piece from Hume was actually missing! The Sagan version is often called the "Sagan Standard", though it really goes back to Truzzi in almost identical form. Because most people familiar with the idea know Sagan's version, it's important that it be edited into this page at least in this footnote, otherwise it's very difficult to find this here due to the original "extraordinary" version having been edited, somewhere along with wiki-way, into "Exceptional claims require exceptional sources", a string of text that virtually no one will actually ever search for. — SMcCandlish  Talk⇒ ɖ^⊝כ^⊙þ Contrib. 00:03, 24 January 2013 (UTC)

 Done. Very interesting as well. I never knew the full origins, despite having used the Sagan version many times.--Fuhghettaboutit (talk) 12:56, 24 January 2013 (UTC)

Offline web sources

Are we allowed to use offline web sources in articles, and would they pass WP:Verifiability? For example, a source by an online newspaper, but not providing the actual URL. Till 13:14, 14 January 2013 (UTC)

If you don't provide the URL it is not verifiable. I'm kind of curious by what you mean by an "offline" web source? Do you mean a source that was on the web but has since been deleted? If that is the case, an archived copy should be found or the source replaced. Betty Logan (talk) 13:25, 14 January 2013 (UTC)

Well, sort of. I knew my wording would be confusing. The problem is that there is no other source or an archived copy. When I click the link, I am redirected to a page that says I have to log in to the website. Is providing the title of the source and the publisher and date not sufficient enough? Till 13:58, 14 January 2013 (UTC)

In that case, the source does satisfy WP:V. It is, in fact, online. It's just that one has to log in to the website, possibly for a fee, to access it. The applicable section is WP:PAYWALL. The best way for you to cite it is to provide as much information as you have, including the URL, and also include either Template:Registration required (if there is no fee) or Template:Subscription required (if a fee is required). --Tryptofish (talk) 23:04, 14 January 2013 (UTC)

Hi again. I found an example of what I was talking about. Does that source meet WP:V as the title, author, date and publisher are provided, but not the URL itself? Till 13:27, 25 January 2013 (UTC)

If the article appeared in the print version of the newspaper then that source is ok (although a page number would help); if it only appears online then a url is needed. Betty Logan (talk) 13:47, 25 January 2013 (UTC)

Is it possible to find out if the article appeared in the hard copy (print)? In other words, would a reader have to (e.g.) obtain a hard-copy of that newspaper to verify the existence of the source/article? Till 14:04, 25 January 2013 (UTC)

Well, usually, when someone doesn't provide a url for a newspaper article it implicitly means that they are referencing a hard copy version, so as a rule we don't question it. Usually online newspaper articles appear in print as well (although sometimes published on different days), so it's not generally a problem. If someone were to obtain a print copy of the newspaper and the story does not appear in it then that would present a problem: assuming good faith, either they have got the citation wrong (which would then need to be corrected) or the story appeared online but not in print, in which case a url would have to be provided. An editor doesn't have to "prove" the existence of the print version though, because if the editor is in good standing we are obliged to believe them. Betty Logan (talk) 14:34, 25 January 2013 (UTC)

Till, I agree with everything that Betty Logan said, and I don't have much more to add to that. I'm not sure which source you are referring to, because the link you indicated shows a cite that includes the URL. But, in general terms, if you want to find out whether there's an online version to which you can link, a good strategy is to do a Google News search (making sure you are covering the correct time period, if it is not recent) for the title of the article that you saw in hardcopy. --Tryptofish (talk) 19:51, 25 January 2013 (UTC)

The source didn't have a URL, but nevermind, I have got the picture now. I will contact the main article contributor about this. Thanks for your help Till 01:15, 26 January 2013 (UTC)

Returning to a possible footnote for the "Burden of evidence" section

I'm pleased that the discussions about BURDEN have quieted down, and I hope that now might be a good time for a more thoughtful discussion of a possible footnote to add in that section.

Currently, WP:CHALLENGE includes a sentence with the following wording:

When tagging or removing material for not having an inline citation, please state your concern that there may not be a published reliable source for the content, and therefore it may not be verifiable.

I think that wording for the sentence is good, and it looks like it has been stable for a while. I suggest adding a footnote at the end of that sentence. The footnote is revised from versions that have been discussed earlier.

Footnote:

When tagging or removing such material, please keep in mind that other editors may object to your edits if you seem to be paying attention selectively to material of a particular POV, or if your edit history shows a pattern of chronic, frequent deletion of unsourced information without any effort to check for the existence of sources. Sometimes, the material may be sourced to a citation elsewhere on the page. For these reasons, it is advisable to communicate clearly that you have a considered reason to believe that the material in question cannot be verified.

Thoughts? --Tryptofish (talk) 23:38, 14 January 2013 (UTC)

An example for thought here: Murdoch University recently had a section on its main campus removed because it was unsourced (diff). A source could have been found with the bare minimum of effort in google. No reasonable person would think that the content in question could not be verified. Whatever happens, we need something to stop that sort of removal without any thought for improving articles. ˜danjel [ talk | contribs ] 16:53, 15 January 2013 (UTC)

Um... the section had been tagged as needing a source for four YEARS. I think there is a good argument for saying "hmmm... if no one could find a source in all that time... maybe it isn't actually verifiable". If that assumption is in fact incorrect... FIX THE PROBLEM yourself... rather than spend time and effort complaining that someone else did not bother to make that bare minimum of effort, make that effort yourself. Find a source, and simply return the section with a citation. Problem solved. Blueboar (talk) 17:05, 15 January 2013 (UTC)

If it's a war situaiotn, what you mean is "find an unusually bulletproof source that says it" North8000 (talk) 17:13, 15 January 2013 (UTC)

Does not look like that example was a "war" situation. More a case of neglect and laziness. Blueboar (talk) 17:27, 15 January 2013 (UTC)

But aside from that, the "sky is blue" statements are often the hardest to find bulletproof (per wp:rs) sources for because RS's typically don't write things that restate the obvious. North8000 (talk) 17:24, 15 January 2013 (UTC)

Blueboar: exactly what I do, thanks, for example in regards to Al Jazeera Academy which the same editor has also swooped on (fixed with a couple of seconds of google) twice (and again fixed). The fun thing was that the second time removed content which was essentially already covered and sourced in the article (courtesy of the first ref fix). But deleting is more fun than googling or even reading articles. ˜danjel [ talk | contribs ] 17:26, 15 January 2013 (UTC)

I really like Tryptofish's idea. It brings it up as a consideration, not a dictated action or restriction. North8000 (talk) 17:15, 15 January 2013 (UTC)

Just to examine this from the other side of the coin... Using Trypto's proposal as a starting point, we could re-state it as:

• Editors should assume that, when someone else tags or removes material for not having an inline citation, they are expressing a concern that there may not be a published reliable source for the content, and therefore it may not be verifiable. This concern is answered by providing a citation.

This addresses the same issue, but phrased as guidance to the Editors who are responding to a challenge, rather than guidance to the challenger. Blueboar (talk) 17:38, 15 January 2013 (UTC)

That essentially allows for drive by tagging, then drive by blanking. ˜danjel [ talk | contribs ] 17:40, 15 January 2013 (UTC)

If an editor is concerned that that's what's occurring, they can always discuss the matter at the article's Talk page...or moot the issue by finding sources. I don't personally have an issue with drive-by tagging followed by drive-by blanking as long as a reasonable amount of time transpires between the two....though I recommend moving contested material to the article's Talk page to leave a clear record of it rather than outright removal. Doniago (talk) 18:11, 15 January 2013 (UTC)

It does not address the same issue, Bluboar. Tryptofish's is much better. The issue it adresses is when the citation is right there in the article, but there is someone who does not look at the sources and doesn't appear to care to. Alanscottwalker (talk) 18:23, 15 January 2013 (UTC)

Um... we seem to have confusion over what problem Tryptofish was trying to address... I took his proposal as talking about situations when material is completely unsourced... You seem to think he is talking about what to do when the information actually is sourced. Trypto... would you clarify your intent? Blueboar (talk) 18:34, 15 January 2013 (UTC)

(edit conflict)Agree with Alanscottwalker....sorry to say it, but I think that Blueboar's completely misses the point. It's even directed to somebody completely different. I think that Tryptofish's idea goes right to the point. North8000 (talk) 18:36, 15 January 2013 (UTC)

It think that it is pretty clear that it is giving "keep in mind" advice / gentle nudge to the person doing the tagging or deleting. North8000 (talk) 18:40, 15 January 2013 (UTC)

OK.... but that raises a side concern... To my mind, a reader should be able to quickly see that given bit of information actually is verifiable (by having a citations near it.) If the source that verifies the information is buried elsewhere in the article... then I would say we have not done a proper job as writers. We should probably repeat the source. This would especially be the case when an entire section does not have a citation. Having a repeat citation would make it much clearer that the material is in fact verifiable.

I suppose my larger concern here is that there seems to be a flawed assumption that "drive by" tagging and removal is in some way harmful. It isn't. In fact, I take the attitude that "drive by" tagging and removal is helpful. It highlights areas of the article that need improvement. That improvement may come in the form of adding a citation that was not previously provided... it may come in the form of repeating a citation to make it clearer that it covers the material... it may even come in the form of a re-examination of the tagged or removed material, and a decision to omit or re-write it.

This is why I suggested an "examination" of the passage looked at from the other side. I was not seriously proposing that we adopt my language... but I was seriously asking you all to think about it from a new angle. If you start with the assumption that "drive by" removals are at least trying to be helpful... and that we should respond to that attempt at helpfulness by improving the sourcing of the article, then half of what we say in WP:CHALLENGE is unnecessary instruction creep. Blueboar (talk) 19:22, 15 January 2013 (UTC)

Personally, rather than start with any assumption, I start with the knowledge / knowing reality that (just using rough numbers) about 90% of of tagging and removals are well intended & helpful, and about 10% are to pursuing a POV'ing or pissing war effort, and that most of the latter efforts are harmful. And IMHO something that gives a bit of a nudge that helps reduce that 10% is a good thing. North8000 (talk) 19:47, 15 January 2013 (UTC)

Question... with a "drive by" edit, how do you know whether a given tag/removal falls within the 90% helpful, or the 10% POV (potentially) harmful. My answer is... there is no way you can know. At best the editor left an edit summary saying "unsourced"... and more often they left no edit summary at all. We are supposed to WP:Assume good faith, so while we might suspect nefarious motives, we don't actually know... And in the long run, it does not really matter what the intent of the tagger/remover was... the tag/removal still highlights an area that needs improving, and dealing with the tag/removal still results in an improved, better sourced article. It's a win-win no matter what the initial intent.

Perhaps your experience has been different, but when I look into the typical complaint about drive by tag/removals, I inevitably find that the tag/remove was at least somewhat justified... and that the person who got upset about the tag/removal is displaying a bit of article WP:OWNership (the unstated, underlying complaint really being: "How dare someone come out of the blue and mess with my wonderful article!). Now that is an attitude we should be discouraging. Blueboar (talk) 20:24, 15 January 2013 (UTC)

Blueboar, yes, I'm happy to clarify what you asked me. You asked whether I was referring to situations with, or without, sourcing. My principal intent is cases where there is no citation present after the material, although the concept is just as valid in cases where, instead, there is an inline citation, but the challenging editor wants to challenge it anyway. As it happens, I agree in principle with the alternative version that you presented here (but all those editors are right that the version that I proposed is better! ), and I also agree with you that there is sometimes WP:OWNership in objections to valid challenges. In fact, my thinking very much arises out of agreeing with both "sides" of what has often been argued on this talk page. I explained that earlier, here, and had hoped that you would reply at that time, but it got lost in the tl;dr. Please take a look there, and you will see that my intent is to be helpful to one kind of challenge, the kind that I think concerns you here, while cautioning against another kind, which may be different than what you are thinking of. --Tryptofish (talk) 22:07, 15 January 2013 (UTC)

Hello Blueboar I disagree with much of what you just said, if implications are included. My statement is an after-the-fact summary (which his what is relevant in this discussion) but on your quesiton, some of the 10% was know (based on specifics and history) right when it was happening and other were identified with the benefit of later information. And as an example, there is a a hard-to-write-for important article with no editors running with the ball there in an important technical field where the RW me writes the prominent sources that others quote and cite. This is a field where most useful knowledge comes from summarization. I had one of the worst such folks in Wikipedia fixated on me (and following me) tag bombing and deleting immense amount of "sky is blue" statements, and then going through a second round of such based on claiming flaws in the sources on the sky is blue stuff after I sourced it and then in the third round using lack of page numbers. Finally I thought to myself, much as I like (and normally enjoy) Wikipedia, hell with this s**t, I can either fight the person who is biting my leg at the article for hours sourcing and resourcing sky-is-blue statements, or have publishers that I've been turning down pay me $250 an hour to do the same thing with no harassment, and secondarily on a "life is too short for this s**t" basis so I abandoned the article and it is a shambles. Another common case is articles which are in a "C" or "D" grade state due to eternal POV wars, waged by capitalizing on weak spots in policies to mis-use the policies to wage POV wars. So, yes, sometimes it is used for un-Wikipedia purposes, and sometimes it DOES do damage. Sincerely, North8000 (talk) 22:49, 15 January 2013 (UTC)

It is entirely misguided to delete text merely because it has been tagged for a long time. I can accept deleting tagged text because the deleter has been unable to verify it, or even that he thinks it sounds suspicious, but not just for being tagged. Tagged for years does not mean that nobody has been able to find sources, in a backwater article it can just as easily mean no one has bothered to look, or even no one has noticed the tag. Taken to its limit (as some editors surely would if it were not considered disruptive) would effectively be enforcing "no contributions without citations". I could live with that rule if that were the community consensus, but it isn't and this policy page should not be pretending that it is. SpinningSpark 23:32, 15 January 2013 (UTC)

Re: it can just as easily mean no one has bothered to look, or even no one has noticed the tag. True... in which case the removal is a great way to draw attention to the tag... removal alerts editors who didn't notice the tag to the fact that it had been there (and they now have to deal with it)... and it tells those who did notice it (but never bothered to do anything about it) that our patience has run out and now they actually do have to do something about it.

The policy is quite clear... you are allowed to make a contribution without a citation... HOWEVER, if a citation is requested, then you are required to provide one. We will give you a reasonable amount of time to do so (depending on the nature of the material)... but don't abuse our patience! The fact that we are willing to wait is not a excuse to ignore the request or to do nothing about it. Blueboar (talk)

Well, we've certainly had a lot of discussion of the concepts here, but I'd like to get back to the specific question of whether we have consensus (not unanimity) for implementing the footnote. In my opinion, maybe we do. --Tryptofish (talk) 16:52, 16 January 2013 (UTC)

• Support Some of the discussions above were sort of off the topic of this footnote. The footnote is about advising (not requiring) that persons tagging and deleting material be communicative. Sincerely, North8000 (talk) 16:57, 16 January 2013 (UTC)

Re the part of the proposed footnote that says, "if you seem to be paying attention selectively to material of a particular POV, or if your edit history shows a pattern of chronic, frequent deletion of unsourced information without any effort to check for the existence of sources." — Is this essentially saying "if you are a biased and bad editor" ? --Bob K31416 (talk) 18:19, 16 January 2013 (UTC)

Well, no, obviously not in those words. But it is discouraging those kinds of editing patterns, while pointing out that providing an explanation for such challenges is a good way to avoid creating that misimpression. --Tryptofish (talk) 18:34, 16 January 2013 (UTC)

Re my comment, "Is this essentially saying" and your response, "not in those words" — we may have a future as a comedy team. --Bob K31416 (talk) 21:38, 16 January 2013 (UTC)

You handed me a straight line I couldn't resist! --Tryptofish (talk) 01:28, 17 January 2013 (UTC)

"you seem to be" and "you are" are essentially very different statements. Any editor acting in good faith should appreciate advice on how to avoid giving a different impression. — HHHIPPO 22:57, 16 January 2013 (UTC)

• Support I actually think it is about biased and bad editing. *shrug* ˜danjel [ talk | contribs ] 21:27, 16 January 2013 (UTC)
• Support This is fully along the lines of WP:FIES. — HHHIPPO 22:57, 16 January 2013 (UTC)

I love the different words we come up with to describe pages that are less than "official policy/guidance", but more than "just an essay"... WP:FIES is apparently an "information page"... WP:ATT became a "summary"... etc. Blueboar (talk) 23:09, 16 January 2013 (UTC)

• Ok... I'll Support - with a caveat - as simple a "tweak" to the existing language, I don't really mind the footnote. So, I will not oppose. However, the more we discuss the section, the more I am coming to the conclusion that the current language tries to address too many concerns and ideas at the same time. I am still in "thinking" stage and so don't have a counter proposal... but there is the germ of one beginning to form in my mind. Blueboar (talk) 23:09, 16 January 2013 (UTC)

Thanks, Blueboar. I appreciate that. --Tryptofish (talk) 01:28, 17 January 2013 (UTC)

Yes, thanks Blueboar. Sincerely, North8000 (talk) 02:43, 17 January 2013 (UTC)

• Support I like this idea. While I also agree with Blueboar, I also feel attempt should be to cover all the basis and a footnote seems to be the ticket.--Amadscientist (talk) 01:35, 17 January 2013 (UTC)

To cover all the concepts and issues that people seem to want this section to cover, we would need about 6 or 7 footnotes. Just saying. Blueboar (talk) 01:44, 17 January 2013 (UTC)

Footnote break 1

• Comment — Re "When tagging or removing such material, please keep in mind that other editors may object to your edits if you seem to be paying attention selectively to material of a particular POV" — There are problems with this advice because the article may be biased with unsourced material of a particular POV that needs to be tagged or removed. This footnote may discourage editors from removing such unsourced material of a particular POV, and gives editors with the particular POV a reason to argue against the tagging or removal of the unsourced material that biases the article. --Bob K31416 (talk) 11:23, 17 January 2013 (UTC)

That's why it says "...may object to..." We need to use the reasonable person approach here. A reasonable person (or in this case wikicommunity) would not object to an editor removing unsourced material of a particular POV and would reject the argument of a POV editor's argumengts thus. ˜danjel [ talk | contribs ] 11:27, 17 January 2013 (UTC)

It's not a matter of a reasonable person (or the wikicommunity) objecting, it's a matter of a person with a POV objecting. There are POV pushers among the editors of Wikipedia. They may or may not be defeated in a particular situation, and they cause trouble whether or not they are successful. --Bob K31416 (talk) 11:50, 17 January 2013 (UTC)

• Comment — Re "When tagging or removing such material, please keep in mind that other editors may object to your edits ... if your edit history shows a pattern of chronic, frequent deletion of unsourced information without any effort to check for the existence of sources." — How would one's edit history show that no effort was made to check for the existence of sources? --Bob K31416 (talk) 12:13, 17 January 2013 (UTC)

@ Bob... Re: "your edit history shows a pattern of of chronic frequent deletion without any effort to check for the existence of sources" - "Proving" this would be a two step process:

a) Look at the deleting editor's edit history... does it show a distinct pattern of multiple deletions (in multiple articles)?

b) Look at the material he/she deleted (again at multiple articles)... check to see how easy it is to find sources that support the deleted material... for example, were you able to find lots and lots of appropriate sources simply by doing a basic Google search? If so, then the deleting editor could have done the same. Chances are he/she did not even bother to check. (an important note here: since you were able to find sources so easily, it is now your responsibility to add them to the article(s)... if you don't, you are simply compounding the problem and have no right to complain.)

Of course, there is a major underlying assumption lying behind the warning... which is that deleting uncited material without first checking for the existence of sources is bad behavior (ie "disruptive"). It would be interesting to hold a wiki-wide RFC to confirm that underlying assumption. I suspect there is consensus for it... but I am not sure how strong that consensus would be. There are those who disagree with the assumption. Blueboar (talk) 13:44, 17 January 2013 (UTC)

Thanks but your method of "proof" goes beyond what the footnote says, which is "if your edit history shows...". Your method of proof requires a search for sources. --Bob K31416 (talk) 14:49, 17 January 2013 (UTC)

Sure it does. But this isn't something which is hard and fast like vandalism. The particular issue would have to be discussed and if there needs to be any remediation, then those need to be discussed too. In that discussion, the scale and severity of the issue would have to be considered. For example, if it were an article about a university, then removing an unreferenced section on a research centre wouldn't be so much cause for concern, because it could conceivably be difficult to find RS' for that subject. But blanking the information on the main campus (which should be easy to find) would be a problem. In essence it should come down to a pattern of such behaviour. ˜danjel [ talk | contribs ] 14:53, 17 January 2013 (UTC)

Thanks but that also goes beyond what the footnote says, which is "if your edit history shows...". Your method requires knowledge of what would and wouldn't be situations where the sources are easy to find and is a matter of opinion, which can differ from person to person for different situations. --Bob K31416 (talk) 15:18, 17 January 2013 (UTC)

Yep, that's why I said Pattern of Behaviour. This isn't 3RR or even NPOV. A judgement call would have to be made if there's genuinely an issue. For myself, I can think of only one editor that I've seen for whom the above would genuinely be an issue, and that's because deleting easily sourced content is kind of his thing. ˜danjel [ talk | contribs ] 15:28, 17 January 2013 (UTC)

I see the "without first checking" in its context as being a nudge of behavior and towards expectations of behavior, not itself the standard. Taken out of context it is two levels up from the type of behavior that that we'd like to reduce, which is using wp:burden for deletion (including via a tag->deletion sequence) of (usually imperfectly sourced) material which the deleter does not question in order to pursue a POV effort or a pissing war. North8000 (talk) 15:04, 17 January 2013 (UTC)

The problem with "nudges of behavior" is this... While you or I might intend something as a "nudge", our intent will be misconstrued. We should assume that someone else will not see it as a nudge, but will instead read it as being a firm and fast "rule", and any heretic who does not follow "the rules" should be hauled before ANI and excommunicated from Wikipedia. We always have to ask: how can someone else misconstrue our intent? Blueboar (talk) 15:48, 17 January 2013 (UTC)

Being hauled before ANI does not mean being banned from wikipedia. If it's clear that there's no or only very minor disruption, then it's more likely that there'll be nothing more than a gentle nudge back in the right direction for whichever party needs it. ˜danjel [ talk | contribs ] 16:08, 17 January 2013 (UTC)

I agree, and perhaps even more strongly. It seems like ANI is a bit toothless and tries too hard to be fair-ish even when there is some pretty offensive stuff going on. A few hours lost from a few lives. ClaudeReigns (talk) 16:33, 17 January 2013 (UTC)

If the goal is truly to provide a "gentle nudge", then I should hope we would encourage lower-intensity approaches such as trying to discuss the matter with the user on their Talk page in a civil manner (sadly, I feel I need to emphasize that) and perhaps an RFC/U before taking matters to ANI. Doniago (talk) 16:36, 17 January 2013 (UTC)

Ideally, the ANI is for incidents. I read that as emergency intervention. You use talk first unless there is bad voodoo going on. RfC more of a pattern-of-behavior thing. ClaudeReigns (talk) 16:43, 17 January 2013 (UTC)

This doesn't eliminate POV-pushing. It only makes room for a new kind of POV-pushing based on bad/no sourcing. The conversation will shift away from "That doesn't look right" towards "You didn't say Simon Says". The only cure for POV-pushing is community involvement - multiple points of view - in other words, more peer-review. Which is better for the project? Better sourcing and more peer-review. That enhances our reputation. This does not. ClaudeReigns (talk) 16:02, 17 January 2013 (UTC)

• Huh? I think everything should be subject to a Full Cavity Search. ClaudeReigns (talk) 12:53, 17 January 2013 (UTC)

This section of the discussion arises from two issues raised in Bob K's comments higher up: that someone who is pushing a POV might accuse someone else of pushing the opposing POV, and that someone might accuse someone else of not having checked for sources, without actually knowing whether they had checked or not. I'm sorry that what I'm going to say is harsh, but these concerns are totally without merit, and border on ridiculous.

Wikipedia has a long history of POV disputes. Of course people pushing one POV accuse their "opponents" of pushing the opposite POV. That happens all the time. We deal with it. But this footnote does nothing to make it worse. It does NOT establish a policy that failure to provide a particular kind of edit summary is disruptive conduct. Repeat: it does NOT. It provides advice that one can lower the risk of such accusations by providing an explanation. What's the worst that will happen? If someone tries to haul someone else before the powers that be for what, as ClaudeReigns (I love your user name, by the way!) says, "You didn't say Simon Says", what will result? No administrator is going to block someone for forgetting an edit summary once or twice. Maybe that editor gets advice to provide a better edit summary next time. Oh the horror! There's NOTHING that becomes sanctionable as a result of this footnote. Repeat: NOTHING.

Of course, no one knows for sure whether or not another editor has checked for references. Duh! But, as I've said more times than I can count now: this. When editors seem to go on sprees of deleting uncited content, other editors will think what they will think. This footnote does not change that. It does NOT make those thoughts a basis for seeking sanctions. Repeat: it does NOT. What's the worst that will happen? Someone gets advice to provide better edit summaries. Oh the horror!

All the footnote does is encourage better communication. I realize there's a parlor game on this talk page, of spending an excess of time obsessing over every imaginable implication of every punctuation mark on the policy page, but, really, don't over-think stuff that isn't even in the proposed language, just because you can dream it up. --Tryptofish (talk) 00:38, 18 January 2013 (UTC)

Ah! I see! There was so much talk of turning WP:CHALLENGE into a guideline that I assumed that what this was all about. That's what I'm rejecting. No. No. No. Returning the footnote is good advice. Strong Support for the footnote. Utter mortification at WP:PROVEIT-->guideline. As a side note, if deletionists want to clean up old articles in a sweeping fury, I am not opposed. They should, however, have an inclusionist teammate of a decidedly different topical bent to follow up on challenges so they can stay busy and not war. I think that would be more commonsense and somewhat avoid an appearance of POV-pushing. I'd say that would be the most effective way to cut the boilerplate population in a non-bitey way. ClaudeReigns (talk) 08:42, 18 January 2013 (UTC)

Great! Thanks! --Tryptofish (talk) 00:02, 19 January 2013 (UTC)

Footnote break 2

• I supported Trypto's footnote earlier... but am afraid I must change my mind. I offer the following scenario to explain why...

Suppose I go on an "article clean up" sweep that includes checking for WP:V issues. I review 500 random articles. 300 of them are fine and need no tagging or removing (so I do not make any edit). 200 of them do have unsourced information. When I review these 200, I find that 150 fall into the category of "Hmmm... I think this is probably verifiable" (so I make no edit).

Of the remaining 50, 25 fall into the category of "Hmmm... I have no idea whether this is verifiable or not, but I think it needs a source... unfortuately I don't know the topic well enough to tell which sources are reliable and which sources are unreliable. However, I know that other editors do know the topic, and can make that call..." So I tag the material and move on.

The final 25 fall into the category of "hmm... I don't think this is likely to be verifiable" I might be wrong, but that is my good faith judgement. I remove.

Now... you may or may not agree with my last action (some will, some will not)... but I did edit in good faith, and was not motivated by any POV. Here is the problem... My edit history will not reflect all those articles where I did NOT make any edit. It does not reflect those 300 articles where my review found no unsourced material, nor (more importantly) does it reflect the 150 articles where I decided to let the unsourced information stand without tagging or removal. All it shows are the 50 articles where I did determine to take some action and either tagged or removed. Nor does it reflect the fact that I did actually look at a few sources, but because I could not determine which were reliable and which were not, I did not add them. Now consider my edit history... it appears to have a solid pattern of tagging and removal, when in fact my actual practice is far more mixed - and actually leans heavily towards NOT tagging or removing. Blueboar (talk) 19:25, 17 January 2013 (UTC)

My main problem is where information that could be easily sourced is removed instead. For example, the main campus location and description of a major university, which should be easily found with a couple of seconds of google. A pattern of behaviour of removing such information is problematic against the proposed footnote because it is disruptive.

On the other hand, the scenario you describe is quite defensible. A second editor reviewing the removal of content could, perhaps, try to find a source, but it would quickly become apparent that sources aren't necessarily easy to find. One presumes that it isn't "the sky is blue", "water is wet", "dogs (except for dingoes) bark" sort of thing. ˜danjel [ talk | contribs ] 23:46, 17 January 2013 (UTC)

Blueboar, I really wish you'd engage with what I said here. I've pointed it out to you repeatedly. Using your example here, and focusing on those 25 pages where you tagged, let's consider the consequences of the proposed footnote. NOT what one can dream up, but what the footnote would actually do. Let's imagine that, at each of those 25 pages, you included an edit summary with your tag, in which you communicated in some way that "I don't think this is likely to be verifiable." You'd be doing exactly what the footnote suggests. Exactly! And would those edit summaries have been some sort of painful burden? Oh, the horror! Alternatively, let's say you didn't really explain your tagging. Well, you can do that now. What happens? Maybe someone responds to you and expresses concern that you tagged improperly. And you discuss it with them, and the discussion goes where it goes. What changes if the footnote is implemented? That person can also point you to the footnote. Nothing more. No policy violation, no sanction. That's it! I'm receptive to your idea of an essay that expands on WP:CHALLENGE. But that doesn't mean we cannot also do the footnote to what we have here. --Tryptofish (talk) 00:51, 18 January 2013 (UTC)

But what if I don't leave any edit summary? I completely agree that leaving edit summaries is helpful and nice thing to do... and yet, if you look at my actual edit history, you will see that I am very inconsistent about actually remembering to do so. Far too often, I forget and just hit "save page". You can nudge me about this all you want, but the reality is that I am unlikely to change my behavior. It's not an intentional behavior, it's just that I am not in the habit of leaving edit summaries (and am unlikely to get into the habit, unless there are consequences to forgetting). So are my edits disruptive simply because I am not in the habit of leaving an edit summary? 67.100.107.228 (talk) 15:52, 18 January 2013 (UTC)

As I see it, your lack of an edit summary is not disruptive in itself, not something that anyone ought to do anything bad to you for. What it is, is a less-than-optimum editing habit. Wikipedia asks editors for good faith, not for perfection (which is a good thing for me!). Maybe someone would tell you that you should have used an edit summary, and point you to the footnote. You, I assume, would reply that you didn't mean any harm, but that you just aren't in that habit of consistently using edit summaries. That's it. No harm done, other than you getting advice, advice that you ought to follow, but being a human being, might not. Basically, it's what I said to Blueboar just above, about the case where "Alternatively, let's say you didn't really explain your tagging." --Tryptofish (talk) 00:11, 19 January 2013 (UTC)

Footnote break 3

• Comment — The footnote has problems addressing those issues. At least two of the problems surfaced in my last two bulleted comments and the discussion that followed them. The issues associated with the footnote need more space to properly address them. Would it be possible to have a subpage for WP:V that addresses such issues, e.g. WP:Verifiabilty/Supplement? --Bob K31416 (talk) 15:44, 17 January 2013 (UTC)

I have been thinking along similar lines... I have already said that I think we may be trying to address too many concepts and interpretations with too few words. There are a lot of nuances involved here, and in order to cover all of these nuances properly, we need to break it down and explain them in more detail. However, I appreciate the concern that doing so in the WP:V policy may be Instruction Creep. Bob's idea of moving it to a new page makes sense to me.

So... I hereby PROPOSE that we turn WP:CHALLENGE into a new guideline. Something that advises editors as they navigate all of the nuances involved with challenging unsourced information (it could also go into the subtle differences between challenging text and challenging a source ... which might address North's continued concerns about the removal of sourced information).

If we do this, then WP:V can simply summarize the key points in that guideline, and link to it for further information. Any takers? Blueboar (talk) 16:29, 17 January 2013 (UTC)

By "WP:CHALLENGE", Blueboar is referring to what is currently the second paragraph of Burden. I was thinking of having a subpage WP:Verifiability/Supplement with a section for the topics of the proposed footnote, but Blueboar's suggestion of moving WP:CHALLENGE to a guideline page is worth considering. --Bob K31416 (talk) 16:53, 17 January 2013 (UTC)

(Whoops... yeah... I had not noticed that the separation of the various paragraphs by shortcut had been reverted. I was talking about the second paragraph)

Sub-pages and "supplements" are somewhat confusing to most editors... they don't know whether the page is "officially" part of the policy or not. Since our goal is apparently to offer advice rather than "rules", I think it better to clearly call it a "guideline". Then everyone knows where they stand. Blueboar (talk) 17:21, 17 January 2013 (UTC)

In that case, you might consider altering your proposal to keep the first two sentences of the Challenge paragraph in Burden, adding a parenthetical link to the proposed guideline, and moving the info of the rest of the paragraph to the proposed guideline. --Bob K31416 (talk) 18:06, 17 January 2013 (UTC)

• Reject no one bothered to answer any of my concerns. Despite the fact that information for verification is freer than it has ever been, we're looking to back off from WP:V. Reality check: I've been engaged to speak at a vocational center in the coming weeks about Wikipedia and critical thinking. I will have to explain the implications of the possible policy change. I will feel inclined to warn them rather than encourage. Sad times, IMO. ClaudeReigns (talk) 18:18, 17 January 2013 (UTC)

  Did my comments under "break 1" respond to your concerns? If not, maybe it's not clear what those concerns are. --Tryptofish (talk) 00:55, 18 January 2013 (UTC)

  "Did I err?" Yes I did. I was responding to something other than the footnote proposal, which I strongly support. I have responded there. In brief: WP:PROVEIT --> guideline is what got my panties in a wad. ClaudeReigns (talk) 08:50, 18 January 2013 (UTC)

• Comment – Technically, the footnote doesn't request any particular behavior wrt. editing, it only advises to explain one's edits, which is always a good thing. There is a difference between "I think this is wrong and I can't find any sources" and "I don't know if the reference at the end of the next sentence also applies to this one and don't care to check". This tells other editors how much effort to expect when trying to fix the problem.

That said, Blueboar made a valid point above that there is an implied message that can be interpreted as declaring certain behavior disruptive. I'm afraid we need a clearer description of (and consensus about) what exactly we want to discourage. — HHHIPPO 20:47, 17 January 2013 (UTC)

• Yes, you are exactly right: the footnote does not mandate any particular behavior, and only advises better communication. I'm perfectly receptive to some sort of additional sub-page, if someone wants to work on it. As for what kind of editing we want to discourage, my take on it is here. As for "a clearer description", that's fine with me. What do you suggest? --Tryptofish (talk) 01:00, 18 January 2013 (UTC)

• Tryptofish, After reading your recent comments, I don't think there is very good communication between us. Perhaps in your responses to what I wrote, it would help if you first quoted a part of what I wrote that you wanted to comment on, and then address that part. Then move on to the next part that you wanted to respond to. But whatever style you want to have for your responses is fine with me. I'll respond or not respond as I think is appropriate. Regards, --Bob K31416 (talk) 02:25, 18 January 2013 (UTC)

As for us not having good communication, yup, that's an understatement. If you want, I suggest that you look again at what I said in "break 1", just before Claude's reply to me, because that's the part that was directed at your two bullet point comments. (What I said in "break 2" was more in response to what Blueboar said.)

But let's save all of us some time. Just above, I asked for suggestions about "a clearer description" of the intent of the footnote. I've been thinking about it, and I'm going to suggest an answer of my own, as a revision to the proposed footnote.

Footnote:

When tagging or removing such material, please keep in mind that other editors may object to your edits if you seem to be paying attention selectively to material of a particular POV, or if your edit history shows a pattern of chronic, frequent deletion of unsourced information without any effort to check for the existence of sources. Sometimes, the material may be sourced to a citation elsewhere on the page. For these reasons, To avoid having other editors misconstrue your intentions, it is advisable to communicate clearly that you have a considered reason to believe that the material in question cannot be verified.

What I attempt to do with that change is to keep the basic meaning of the proposed footnote, since it has actually had a lot of support in this discussion, but make it clearer that we are talking about better communication in order to avoid misunderstandings, rather than disruptive behavior. --Tryptofish (talk) 00:24, 19 January 2013 (UTC)

This last sentence does not describe good faith efforts on my part to engage possible watching editors while making recent improvements at a BLP. The article had long been tagged for problems. I like trying for 2x article expansions necessary for DYK, so rather than simply deleting unsourced statements, I went for an overhaul. I read the sources to see what the vaguely and somewhat erroneously cited sources had to say about the subject, and inline-tagged a few namespace statements that weren't supported. I did a broad search to come up with new sources. When some sources supported namespace statements, I made sure they were properly cited. Eventually only two remaining namespace statements didn't have a source, neither of which could be considered especially contentious. I tagged those and went to sleep. Maybe they'd be supported, maybe they wouldn't. But at least I'd know where things stood when I came back. Later, I performed exhaustive searches by what means I have available and found the DOB and 2009 election margin were circularly sourced. I had actually expected to find some verification and didn't. I deleted those two statements from namespace per BLP. Was there anything wrong with doing it that way? ClaudeReigns (talk) 04:22, 19 January 2013 (UTC)

Short answer: no. Long answer: maybe I misunderstand something about what you are asking, but I'm actually rather puzzled that you would ask at all. If I understand correctly, you made a diligent search for sourcing for two bits of content: the DOB, and the election margin. All you could find was WP:CIRCULAR. Therefore, based on WP:BURDEN, you deleted those two bits of content. There's nothing wrong with that, that I can see. That's what BURDEN is for, all the more so in a BLP. You didn't say here whether you used an edit summary, or anything else, to say why you deleted those two bits of information. If you did, you complied 100% with what the proposed footnote says. If you didn't, I expect one of two things to happen: (1) nothing, or (2) another editor asks why you deleted that stuff, hopefully asking politely, but maybe not. If the second outcome happens, you will have proven the last sentence of the proposed footnote right: someone misconstrued or at least questioned why you challenged the material, and you could have avoided them misconstruing or questioning if you had, for example, used an edit summary. (Minor point: it's article space, not namespace, if you were editing the actual page.) --Tryptofish (talk) 00:02, 20 January 2013 (UTC)

I looked again at the edit summary you used for your talk page comment here, and now I wonder whether your point was that you tagged material because you hoped other editors would see the tags and help, and you are concerned that the last sentence of the footnote would imply that you should tag only after becoming certain that the material is unverifiable, whereas here, you were tagging to stimulate other editors to look for sourcing. Is that it? I'm not sure that it matters. You were convinced, reasonably, that sourcing was needed, and it wasn't there, so you tagged it. What is key is that you were challenging the material based on WP:V, not based on anything else, which is what the sentence the footnote would attach to is asking for. Remember: this is a proposed footnote, to come after a sentence already in the main text of the section! This footnote will have a context, right? The footnote talks about users who (1) seem to push a POV, or (2) go on massive tagging or deleting sprees, or (3) don't realize that the citation is provided elsewhere on the page. I think I can safely assume that none of those three was the case here. So, all you had to do was use an edit summary when you tagged, saying why you did it. I don't know, maybe some users are such a--holes that they would give you a hard time, saying that you tagged the material to ask for sourcing when, in the deep dark recesses of your mind, you didn't really doubt that sources would exist, but I don't think reasonable editors would take those complaints seriously, and neither should you. --Tryptofish (talk) 00:19, 20 January 2013 (UTC)

Footnote break 4

Re: The footnote talks about users who (1) seem to push a POV, or (2) go on massive tagging or deleting sprees, or (3) don't realize that the citation is provided elsewhere on the page.

While I do understand that the intent of the footnote is to talk about these three limited, even rare, situations... I am not sure that this intent is made clear. The fact that Claude had to ask whether the footnote applied to his actions makes me wonder whether others might misinterpret the intent as well.

Also, I have a few questions about the three situations: (1) what if the tag/removal is made as a response to POV pushing, not to push it yourself? (2) Are "massive tagging or deleting sprees" never beneficial? (3) If a citation is provided elsewhere on the page, could we not simply revert the tag or deletion, with an edit summary noting that the citation is elsewhere on the page, and pointing to it? Blueboar (talk) 14:04, 20 January 2013 (UTC)

So, yes, edit summaries of "still not sourced" at tag and "removing per BLP" at deletion. I think the thing is that some content has a stronger sourcing requirement. Some unsourced can't even be put on talk, though I am quick to put anything on talk which would not be considered libelous. I am always concerned about trying not to appear the POV-pusher, which is why I tend to do things in a way which would stimulate participation. I do edit controversial topics and I do have a view on most things. Therefore I have plenty of incentive to leave a nice clear trail of edit summaries and citations. In all, I strongly support the footnote, but it should not ignore that certain statements in Wikipedia's voice have a stronger citation requirement and a lower deletion threshold than others. ClaudeReigns (talk) 06:00, 21 January 2013 (UTC)

I an with Blueboar on this and oppose the proposed footnote. I have not been contributing to this conversation because Blueboar has been stating my position better than I and adding "me too" tags just seems like verbose mode on. If BB states something I disagree with over this issue I will make it known. I have just changed the sentence as I suggested when I last contributed to a conversation on this talk page about this sentence "please state your concern" to "you ought to state your concerns" for the reasons I gave previously. -- PBS (talk) 16:46, 20 January 2013 (UTC)

Oh, what fun to log back in and see an edit war over whether or not to say "please". PBS, please rest assured that I could not possibly care less about whether we say "please" or "ought". So whatever ends up at the end of the edit war will be fine with me. Truly, I don't care, and it boggles my mind that anyone else does. But this discussion is about the proposed footnote, not about the sentence in the main text.

Blueboar, about that other editor getting confused, you may perhaps want to look at a discussion that showed up on my talk page, because it may shed some light on your point about that. Let's focus on where you and PBS have concerns. You ask me three numbered questions, so I will try to answer them:

1. That's just fine, and all it takes is an edit summary to say why.
2. It depends.
3. Yes.

I feel like this discussion has become a sort of advice column, with editors posing questions to me about the etiquette of WP:CHALLENGE. If anyone really has questions like that, please ask them in a new talk thread, because this thread is about the proposed footnote, and whether to implement something like it or not. Otherwise, it's time to let Dear Abby rest in peace.

I've said it over and over and over and over again: the footnote does not, repeat: NOT, establish any of those three things as "disruptive behavior" that would be subject to blocking, banning, or boiling in oil. It's advice about avoiding misunderstandings, and most of the editors who have commented here understand that. So let's please drop all this discussion about "what if someone makes a WP:CHALLENGE edit in such-and-such a way, is that going to get them in trouble?" If it does, the proposed footnote has nothing to do with it.

The proposed text of the footnote says that "other editors may object". Is that choice of words a problem? Does anyone think that no editors would ever object to those things, but once we implement the footnote, all of a sudden they will? It's a statement of fact: some editors tend to object to these things. The fact that someone objects to something doesn't mean that the thing they object to is forbidden by policy. The footnote does not establish policy forbidding anything. It just points out that if you explain why you did something, people are more likely to understand why you did it.

I revised the proposed footnote by trying to make it more clear that this is about not misconstruing. Now, I'm going to revise it again, to try to make it even clearer:

Footnote:

When tagging or removing such material, please keep in mind that such edits can be easily misunderstood. Other editors may object to your edits if you seem to be paying attention selectively to material of a particular POV, or if your edit history shows a pattern of chronic, frequent deletion of unsourced information without any effort to check for the existence of sources. Sometimes, the material may be sourced to a citation elsewhere on the page. To avoid having other editors misconstrue your intentions, it is advisable to communicate clearly that you have a considered reason to believe that the material in question cannot be verified.

Does this help? If not, please point me towards how to address your concerns, as opposed to raising hypotheticals. --Tryptofish (talk) 00:09, 21 January 2013 (UTC)

Why not just say:

• When tagging or removing such material, please keep in mind that such edits can be easily misunderstood. To avoid having other editors misconstrue your intentions, it is helpful to communicate clearly what you are challenging and why. Be prepared to respond to questions on the talk page.

In my opinion this says it all. Blueboar (talk) 01:09, 21 January 2013 (UTC)

I prefer Tryptofish's version, because there are rare cases where mass removal of content is disruptive. Individuals who engage in such behaviour, if they are going to say anything in edit summaries or on talk pages do little more than say "d per tag" or selectively quote part of WP:V as justification for deleting anything. ˜danjel [ talk | contribs ] 04:52, 21 January 2013 (UTC)

OK, then perhaps we need to narrow this down... when is a mass removal of content "disruptive" (and why), and when is a mass removal of content not disruptive?

Also... the main concern seems to be that "Others may object". We should explore "objections" in more depth... because I think there are times when it's actually the objection (and not the tag/removal edit) that causes the disruption. Blueboar (talk) 13:22, 21 January 2013 (UTC)

Blueboar, to answer those questions, I'm going to ask again that you look at what I previously said here. Please do that for me, OK? And please acknowledge explicitly that you have read it, whether or not you agree with me about it. Because that's where I've already given my own opinion about how to distinguish between constructive and disruptive edits of this nature.

I think danjel is correct, and actually the large majority of editors who have commented in this talk thread agree with that. But I asked above whether your objection has to do with the language about "editors might object", and I think you are saying that it does. I'm open to alternative wording if you or anyone else can suggest it. But I've said repeatedly that editors objecting does NOT equal a determination of disruptive behavior. (Example: I've proposed wording for a footnote. You "object" to that wording. Does that mean you think I'm being disruptive, and I should be blocked?) It's a statement of fact. Editors do object to those things. They sometimes object to other things, too, but we don't have to list all those things (such as fixing an existing POV) in a single footnote. Yes, objections to a valid challenge can be more disruptive than the challenge itself. But if the challenge is accompanied by a good explanation, the likelihood of that subsequent disruption is decreased. --Tryptofish (talk) 21:52, 21 January 2013 (UTC)

Yes, I have read what you said... I disagree with your conclusion that the edits in (2) are necessarily disruptive. I don't think they necessarily are disruptive. In fact, I see them as mostly being beneficial. I understand that others (such as yourself) think such edits are disruptive, but I have yet to see an explanation as to why you consider them disruptive... the closest to a why I have read is "other editors might get upset". Is that the only reason? Blueboar (talk) 15:36, 22 January 2013 (UTC)

If it's the case that you don't object to those edits, but you acknowledge that some people do, for whatever reason... Then the footnote as it stands is technically correct: some "editors might object" if they see that sort of thing happening frequently through a person's contributions. Personally, as I have said above, I object to the removal of easily sourced information that, akin to "water feels wet", should generally not be subject to challenge, because such actions lead to articles with gaping holes, rather than genuinely informative pieces of work (hence my example above of Murdoch University having information on its main campus removed). ˜danjel [ talk | contribs ] 15:52, 22 January 2013 (UTC)

So anything someone objects to is automatically disruptive? I don't agree. I have to point out that removals do not necessarily lead to articles with gaping holes... at least not permanent ones. All it takes to return the information is a source (for your "water feels wet" example: you can find multiple potential sources at this google search). I don't find the removal disruptive at all. What I find disruptive is spending time complaining about the removal, and arguing about whether the "challenge" is valid. Blueboar (talk) 17:33, 22 January 2013 (UTC)

OK, thanks Blueboar for responding directly to what I asked you. Let me now parse a bit about the word "disruptive". What I actually said back there was: "I think that (2) can be a potentially disruptive misunderstanding of BURDEN." Therefore, I'm seeing such edits as being in a gray area, one that is neither purely disruptive nor purely beneficial. Not "automatically disruptive", not "necessarily lead[ing] to articles with gaping holes". You and I both actually agree that, sometimes and to some extent, those types of edits in (2) can do some good. Where you and I may disagree is in that you seem to be arguing that, because such edits can do some good, we must not recommend anything to make them more constructive. I, on the other hand, am arguing that such edits sometimes do a bit of good, and a larger amount of trouble for other editors, and it would actually be an easy thing to increase the proportion of good and decrease the proportion of trouble.

Furthermore, language about some editors objecting to something is not language that says that what they object to is disruptive, in the sense of the word "disruptive" that says that policy has been violated, harm has been done, and there may be a need for a block. I don't think anything in the proposed footnote could lead to someone being brought to ANI or the like and getting blocked or otherwise sanctioned, at least not so long as they eventually show some willingness to participate in communication with other editors about their reasons for tagging or deleting.

Let's say someone goes on what I've described as a "spree" of making a very large number of taggings or deletions, and they leave no edit summaries or other explanations. Do you really disagree that some other editors are likely to express objections? Now let's say that the person who did the deleting responds to those objections by saying something like: "Oh, the reason I did that is that I don't believe that any sourcing for that material exists." Do you think that this person would then be in any jeopardy as a result of this footnote? As I see it, the "worst case scenario" for that person would be other people saying something like: "Well, read what it says at WP:V, and use an edit summary in the future." Do you think something worse would happen? On the other hand, let's say that the person failed to respond to those objections. Let's say that, time after time, they are unresponsive to requests to explain their deletions. Not a one-off, but a chronic pattern. Do you really think that such conduct is so beneficial to the project that we cannot even recommend that the person communicate a reason based on policy? --Tryptofish (talk) 22:48, 22 January 2013 (UTC)

Sure, editors might object... So what? I see such objections as being an expression of WP:OWNership. If we must address the issue in terms of behavior, what I think we should be telling editors is: "If someone removes unsourced material that actually is verifiable... Don't waste everyone's time by objecting to the removal. Instead of objecting, simply return the removed material with a citation to a source that supports it." That resolves the issue with no disruption. Blueboar (talk) 14:18, 23 January 2013 (UTC)

Then I'm going to take what you said literally, in the context of what I asked. I asked about a scenario in which one editor goes on multiple sprees of deleting massive amounts of uncited content without an edit summary. Other editors ask: "Why did you delete that?" Every time, the deleting editor gives no response, no matter how many times other people ask. The pattern goes on and on. Now we know that your interpretation of the scenario is that the deleting editor is a defender of the Wiki, and the editors asking why are guilty of OWNership. Really? In fact, I find it hard to believe that you really think that!

But that's OK. I'll give more consideration to your concerns than you are giving to mine. Please see below. --Tryptofish (talk) 17:35, 23 January 2013 (UTC)

As I've said previously, I consider both please and should to be permissive, not mandatory. I prefer "should" because it is more clearly non-mandatory, but please is also non-mandatory. My real preference is to delete that sentence altogether, as I've said many times before, as I believe that it merely adds a bureaucratic gotcha. Regards, TransporterMan (TALK) 21:40, 21 January 2013 (UTC)

Continuing from my dialog with Blueboar, something more occurs to me, as I try to see it from Blueboar's perspective. How about this revision?

Footnote:

When tagging or removing such material, please keep in mind that such edits can be easily misunderstood. Other editors may object to your edits if you seem to be paying attention selectively to material of a particular POV, or if your edit history shows a pattern of chronic, frequent deletion of unsourced information without any effort to check for the existence of sources. Sometimes, the material may be sourced to a citation elsewhere on the page. To avoid having other editors misconstrue your intentions, it is advisable to communicate clearly that you have a considered reason to believe that the material in question cannot be verified. On the other hand, if you object to some tagging or deletion, please consider providing citations yourself if you have them. Withdrawn. --Tryptofish (talk) 00:42, 24 January 2013 (UTC)

The sentence at the end reflects something Blueboar has said a number of times, and perhaps it helps to present both "sides" of such situations. Thoughts? --Tryptofish (talk) 22:54, 22 January 2013 (UTC)

Based on Blueboar's most recent reply to me:

Footnote:

When tagging or removing such material, please keep in mind that such edits can be easily misunderstood. Other editors may object to your edits if you seem to be paying attention selectively to material of a particular POV, or if your edit history shows a pattern of chronic, frequent deletion of unsourced information without any effort to check for the existence of sources. Sometimes, the material may be sourced to a citation elsewhere on the page. For all of these reasons, it is advisable to communicate clearly that you have a considered reason to believe that the material in question cannot be verified. On the other hand, if someone removes unsourced material that actually is verifiable, please consider providing citations yourself if you have them. Withdrawn. --Tryptofish (talk) 00:42, 24 January 2013 (UTC)

That uses some of Blueboar's wording (minus the part about not wasting everyone else's time). --Tryptofish (talk) 17:35, 23 January 2013 (UTC)

I don't like that, it essentially encourages lazy editing and mass tagging for the sake of bumping up one's edit count. I prefer the original. No offence, but it's clear that you (BlueBoar) are in the minority, I think we have to move forward. ˜danjel [ talk | contribs ] 23:09, 23 January 2013 (UTC)

In the discussion about "please", editors have been doing some head-counting, so I think that it's not unreasonable for me to do it here as well, with all the usual cautions about WP:NOTAVOTE. I've gone back carefully through all the discussion in all of the subsections/breaks of this talk thread. There are 9 editors who support the footnote, and 3 who oppose it. As danjel correctly points out, that's really pretty lopsided. Given that there is pushback against my most recent attempt to accommodate Blueboar's wishes, and given that I really don't see anything further to do that would both satisfy Blueboar and satisfy everyone else, I'm withdrawing my most recent draft versions. I'm going to ask Blueboar and the few editors who agree with him to understand that I've been trying really hard to respond constructively to your concerns, but it just isn't working, and you need to respect the emerging consensus. Pending any further reasonable suggestions, I think that the consensus is to add the following after the page protection is lifted.

Footnote:

When tagging or removing such material, please keep in mind that such edits can be easily misunderstood. Other editors may object to your edits if you seem to be paying attention selectively to material of a particular POV, or if your edit history shows a pattern of chronic, frequent deletion of unsourced information without any effort to check for the existence of sources. Sometimes, the material may be sourced to a citation elsewhere on the page. For all of these reasons, it is advisable to communicate clearly that you have a considered reason to believe that the material in question cannot be verified.

--Tryptofish (talk) 00:42, 24 January 2013 (UTC)

After the initial support there has been considerable discussion challenging the footnote. I don't think you can say that consensus has been emerging for the footnote. --Bob K31416 (talk) 01:05, 24 January 2013 (UTC)

Given the the opinions (below) to my question about whether this policy should focus on content or behavior, I question whether we have a clear consensus to have any footnote outlining behavior. Let's establish that first... then, if there is a clear consensus that we want to have a such a footnote... then we can reach a second consensus on how to phrase it. Blueboar (talk) 01:38, 24 January 2013 (UTC)

Focusing on content (a point with which I would agree) does not mean that preclude any behavioural considerations that could be addressed in a footnote. ˜danjel [ talk | contribs ] 01:41, 24 January 2013 (UTC)

Bob and Blueboar, I counted both of you as opposing the footnote. If you can show me that editors who previously supported the footnote have been swayed by your arguments to change their minds, then I'll take that seriously. Failing that, you are wasting everyone's time. A "no consensus" result in the discussion about content versus behavior does not constitute a consensus against the footnote. We need to move on. --Tryptofish (talk) 22:08, 24 January 2013 (UTC)

I reverted your implementation of the footnote per my previous comments re consensus and challenges. 1RR. --Bob K31416 (talk) 02:39, 25 January 2013 (UTC)

And I have reverted you. There is a clear consensus, whether you agree with it or not. --Tryptofish (talk) 19:24, 25 January 2013 (UTC)

I have to challenge that, Trypto. I do not believe there is a clear consensus... a majority may have supported adding your footnote, but that is not the same as a consensus. Especially when you factor in the majority that favor this policy being focused on content instead of behavior (see below)... I think the situation is far from "clear". Blueboar (talk) 19:59, 25 January 2013 (UTC)

I'm sorry the two of you are unhappy, but I stand by everything I said. What we have here are two editors, you and Bob, plus a third who has designated you as his spokesperson, who are refusing to recognize that you aren't persuading anyone else. I fully understand WP:VOTE, but with 9 editors supporting and 3 opposing, we have to look at how the discussion is actually going. I've bent over backwards to listen to your concerns, and to try to revise the footnote to satisfy you. I, and others, keep pointing out the logical flaws in your arguments, but you both keep repeating the same arguments without dealing with those flaws. That does not constitute a change in direction of the discussion. It's just the same thing over and over. No editor who supported the footnote has changed his or her mind as a result of your arguments. Not a single one. The discussion below is about a general concept of content versus behavior, not about this specific footnote. There is a loose trend towards saying that the policy is mostly about content, but there is no consensus that it must never address behavior. To revert this footnote on the basis of that discussion, you would have to show a consensus that WP:V must never contain any mention of behavior whatsoever, and that simply is not what you have, or ever will have. --Tryptofish (talk) 20:12, 25 January 2013 (UTC)

This is easy enough to resolve... let's agree on a neutral admin, and ask him/her to review the relevant discussions and determine a) whether there is or is not a consensus, and b) if so, what that consensus is. An RFC closure should never be "declared" by the people who were most heavily involved in the RFC discussion... either its primary supporters or it primary critics. Blueboar (talk) 02:22, 26 January 2013 (UTC)

I'm fine with having a neutral administrator review this. Indeed, I have nothing to hide. Go right ahead, with my blessing. But I note that this was a discussion, rather than an RfC, and there was nothing out-of-process with the conclusion that I drew. Optionally, please feel free to open an actual RfC, if you would like to have more input. --Tryptofish (talk) 20:52, 26 January 2013 (UTC)

My mistake in calling it an RfC... it was a DwCR (Discussion with Comments Requested) :>) Blueboar (talk) 23:18, 26 January 2013 (UTC)

FYI - I have asked User:Slim Virgin to review and settle this for us (see: User talk:SlimVirgin#Neutral third party opinion needed to settle dispute at WT:V)... she used to be very involved on this policy page, and so knows the background of the BURDEN section well... Yet since she has not been active in any of our recent debates, I think she will be objective and neutral. I respect her opinion. If she says that there is consensus for your footnote, I will back off. I hope you will do the same if she says the opposite. Blueboar (talk) 00:01, 27 January 2013 (UTC)

That seems like a very strange choice. SlimVirgin was an editor with very controversial behavior here, e.g. the closing chaos at the large RfC. Did you consider that when making your decision to ask her? --Bob K31416 (talk) 00:20, 27 January 2013 (UTC)

I'm not sure that "used to be very involved" makes someone neutral. I think Bob's concern is one that will be shared by many editors here. I'll also note that I have had a history with this administrator, and that may also complicate things. --Tryptofish (talk) 00:41, 27 January 2013 (UTC)

Also, in your request, you stated that #Is WP:V primarily a content policy or a behavioral policy? relates to the consensus decision. In your opinion, it does, but that is not an opinion that is shared by all of us here, so in that regard, your request was not neutrally worded. What I agreed to was not giving an administrator a "super vote" as to whether or not to include the footnote. The only thing to be determined is whether or not I was correct in my evaluation here of the discussion consensus. --Tryptofish (talk) 01:23, 27 January 2013 (UTC)

Whoops... I was not aware of your prior conflicts with SV... given your objections, I have asked her NOT to get involved, and will find someone else (any suggestions?)

As for WT:V#Is WP:V primarily a content policy or a behavioural policy? That is the primary reason why I object to your footnote... The realization that this policy was focusing too heavily on "behavior", and shifting away from "content" is why I changed my mind from my initial support to an oppose regarding your footnote. So that section is absolutely related to our dispute, and anyone trying to figure out the viewpoints here has to read that part of the discussion (they should probably also read Bob's "bot" thread, although that is somewhat less on-topic). Blueboar (talk) 02:19, 27 January 2013 (UTC)

The best way to ask is via a neutrally worded request at WP:AN. I understand that that is the primary reason you object to what you call "my" footnote. But it is unacceptable for a closing administrator to enter a "super vote", as I already explained to you. You don't get to ask an administrator to rule on whether they agree or disagree with the nine editors who support the footnote. You only get to ask whether or not this was actually the consensus of the discussion that took place. And by the way, you may want to notice that the footnote has been on the policy page, in one form or another, for over 48 hours now, and yet only you and Bob have indicated objections to it being there; another editor has revised it without objecting to it. --Tryptofish (talk) 03:05, 27 January 2013 (UTC)

Gentlemen, are you trying to defuse all the problems associated with this footnote by making it so very long that nobody who needs to read it, will?—S Marshall T/C 03:01, 27 January 2013 (UTC)

Good question. I'm not, but Blueboar is trying to make it so long that it becomes an essay and is deleted from here. Please feel free to prune it. --Tryptofish (talk) 03:07, 27 January 2013 (UTC)

I made a pruning of my own, in response to S Marshall's comment. Only after saving that edit did I see Blueboar's null edit, indicating that he might be able to live with the footnote in its revised form. I don't know whether that changes as a result of my pruning, but we can certainly discuss that. If there's any chance that we are at the olive branch stage, I would most heartily welcome it. In fact, once we started the back-and-forth editing of the footnote on the page, I think that we have made a great deal more constructive progress in finding a middle ground than we did during the lengthy talk page discussions. --Tryptofish (talk) 03:17, 27 January 2013 (UTC)

If you two have found a compromise you can both live with, then I'm not anxious to disturb it! I also can't really help with trimming it because I don't even understand the purpose of the footnote, or at least, not in any detail.

It seems to me that footnotes 3 and 5 are groping towards similar things, and might be combined with advantage. I also wonder whether a link to WP:PRESERVE wouldn't be out of place there. But it's not the most urgent thing to fix about this page...—S Marshall T/C 03:23, 27 January 2013 (UTC)

Please

As the edit warring continues over "please", let me "please" suggest de-coupling the two places where it occurs. Perhaps the considerations are different between them. One is about bad stuff in BLPs, the other is about indicating reasons for a challenge. --Tryptofish (talk) 00:35, 21 January 2013 (UTC)

• Well, I'm being overpowered by serial reversions. The stable version of the page includes the word "please". Removing it is an undiscussed bold change, but I can't now revert again without violating 3RR, so please could a neutral editor restore the stable version while we discuss it.

  Amadscientist's view, that I have to get permission at WP:BLP for what he describes as a "change", is simply bizarre. This is the core policy page and "please" is the longstanding wording.—S Marshall T/C 00:39, 21 January 2013 (UTC)

Just so you know, I'm neutral to the extent that I don't really care about that, but I've requested that the page be full protected. --Tryptofish (talk) 00:43, 21 January 2013 (UTC)

I would support full protection of the page. The change was made by another editor to accurately relect the BLP policy. Marshal has been edit warring to retain a version that simply does not reflect the policy as written at the BLP policy page. Being stable for any length of time is no excuse to revert a legitimate edit.--Amadscientist (talk) 00:48, 21 January 2013 (UTC)

• Just to be clear on the sequence of events: It said "please" for months. PBS removed that with this edit, and I reverted. Then KWW and Amadscientist accidentally formed a tag team to change the policy wording; it needs to go back to the stable version until there's consensus.—S Marshall T/C 00:51, 21 January 2013 (UTC)

You have been warned about unconstructive edits to this page by at least three seperate editors. You toss out accusations of gaming the sytem and tag teaming. It makes little difference that you add an AGF link to the accusation, it is still an accusation of tag teaming that is baseless.--Amadscientist (talk) 00:57, 21 January 2013 (UTC)

• Only three? :) Amadscientist, I've been "warned" about "unconstructive" edits to this page by approximately 11,458 editors since mid-2009 when I started being active here, all of whom were completely convinced that I was trying to destroy the policy utterly. I'm fairly familiar with the situation.  :) The effect of your actions was to combine into a group of editors who forced through an undiscussed change to the policy, preventing me from reverting you by using 3RR. I accept that it was inadvertent.

  In fact, I think that I'm the one who's being constructive. Issuing orders will not be more effective than seeking to persuade our volunteers.—S Marshall T/C 01:07, 21 January 2013 (UTC)

You seem more concerned with the edits of others that you feel roped you, unfairly into edit warring. That didn't happen. Editors don't agrees with you. That's all. You did not need to continue to revert. I do not believe you are trying to destroy anything, but clearly if over 11,000 editors have warned you about unconstructive edits there should be some concern that you might wish to address. I will leave that to you, but is it possible you are not getting the point?--Amadscientist (talk) 01:20, 21 January 2013 (UTC)

• (ec) I personally dislike the "please" version because it makes it sound like an optional request. I prefer Amadscientist's version. Reyk _YO! 00:59, 21 January 2013 (UTC)

Reyk... it is an optional request. We can not force someone to leave an edit summary to explain their edits. All we can do is note that it is helpful for them to do so. Blueboar (talk) 01:12, 21 January 2013 (UTC)

This is really more about the BLP bit, Blueboar, I think. The stable version said "Please remove unsourced contentious material about living people at once," and PBS's version that's been restored thrice so I'm presently not allowed to revert says "Unsourced contentious material about living people should be removed immediately."—S Marshall T/C 01:16, 21 January 2013 (UTC)

Ah... sorry for my misunderstanding over which "please" we were talking about. In this case, since we are essentially repeating what is said at WP:BLP, we should use the same phrasing as that policy uses. If it uses "please", so should we... if it uses "should" or "ought", so should we. Blueboar (talk) 01:40, 21 January 2013 (UTC)

"Please" is a request that gives the impression that it need not be done at all. "Should" makes it clear it is the proper route.--Amadscientist (talk) 01:21, 21 January 2013 (UTC)

Agreed. Reyk _YO! 01:37, 21 January 2013 (UTC)

I think that I like "should be" better because it defines it as where it should end up rather than sounding like it is telling a particular person to remove it. But, from a process standpoint I think that the last stable version (e.g. "please remove") should be in there while we decide this. This should be settled in talk, not by the power of inadvertent tag-teaming. Sincerely, North8000 (talk) 01:44, 21 January 2013 (UTC)

• Amadscientist's actions illustrate exactly what I mean. Amadscientist, do you remember just before Christmas when I posted a link to the BLP noticeboard of this article? I cannot order you to follow the links and remove the unsourced contentious material. And that's why you gave me this answer at the time.

  The policy can't issue orders because editors don't have to follow it. They can simply behave like Amadscientist did and not actually read the unsourced information. It's better to say "please" because then at least editors might be persuaded...—S Marshall T/C 01:49, 21 January 2013 (UTC)

Amadscientist is not the subject and neither is your malformed question at the BPL/N.--Amadscientist (talk) 04:57, 21 January 2013 (UTC)

• You're answering things I didn't say. I didn't say you were the subject. What I said was that your behaviour is illustrative of editor conduct towards BLPs, which is what that sentence of the policy addresses.—S Marshall T/C 12:17, 21 January 2013 (UTC)

One of the problems we have is that this topic has been converted into a behavioural guideline instead of a content guideline. Language like "Articles should be based on independent sources" has been changed into "Base articles on independent sources". This is an example of the same problem: this guidelines shouldn't be requesting editors to behave in any fashion at all, it should be stating what is acceptable contents for an article and what is not. When I reverted, I thought that this was just another case of that inappropriate conversion. I see that it was not, that it was simply an old, existing case of inappropriate language. I still think that language that describes content is what we need, not language requesting behaviour.—Kww(talk) 02:02, 21 January 2013 (UTC)

• I made that change, along with the word "please". You can read the historical discussion, reasoning and supporting consensus in archive 58 of this talk page, near the bottom ("proposal concerning grammar").—S Marshall T/C 02:29, 21 January 2013 (UTC)

• The problem is that it wasn't simply a grammar change: it changed the meaning of the guideline. Most of those changes should be reverted, and we should make it completely consistent. This policy should describe acceptable content, not suggest behaviour.—Kww(talk) 02:43, 21 January 2013 (UTC)

• Really? Why do you think so?—S Marshall T/C 02:46, 21 January 2013 (UTC)

• Because verifiability is an aspect of content, not an aspect of behaviour.—Kww(talk) 02:51, 21 January 2013 (UTC)

• Verifiability is fundamentally behavioural. It's not about what the ideal article looks like, it's about how we move towards better articles from where we are now. And specifically, it's about (1) how and why editors should cite their sources when they add information to articles, and (2) when and how editors can remove uncited material from articles. The policy is primarily concerned with these two behaviours and the importance of rigour in their implementation. Sure, we could convert it to a discussion of what the ideal article looks like, but I don't see the benefit.—S Marshall T/C 12:17, 21 January 2013 (UTC)

"Verifiability is fundamentally behavioral" - Wow... I could not disagree more strongly. Verifiability is fundamentally about content (Unverifiable material is unacceptable in a Wikipedia article). WP:V is (after all) one of our core content policies. It's primary focus should be on content and not behavior. Now, the policy may need to discuss the occasional behavioral issue. However, when we do so, we should try to maintain our focus on the content. If there is a need to delve further into behavior, the proper place to do so would be in a linked behavioral guideline. Blueboar (talk) 14:06, 21 January 2013 (UTC)

You are clearly off on the wrong foot here, S Marshall, and it's the root cause of the language difficulty. It's a content issue: content about living people must be verifiable in reliable sources. No "please" or "thank you" about it. The problem is that the guideline has been rewritten to address people, and doing so without adding the niceties can, indeed, appear a little abrupt. The solution is to target the guideline back at the content that it is attempting to control: content doesn't get its feelings hurt.—Kww(talk) 15:44, 21 January 2013 (UTC)

Note: This difference of opinion about whether WP:V is "fundamentally" a behavior or a content policy needs to be explored further... It may be what underlies our debates over the Burden/Challenge section. Since it goes beyond just the BLP issue, I will start a new thread for it. Blueboar (talk) 15:53, 21 January 2013 (UTC)

I support the version preferred by Kww & Amadscientist even though I just temporarily took it out so that this be settled in talk, not by the power of accidental tag teaming. North8000 (talk) 02:27, 21 January 2013 (UTC)

• Thank you, North8000.—S Marshall T/C 02:29, 21 January 2013 (UTC)
• I am going to use my last revert on that as you are now actually tag teaming. You agree with the prose but accuse two editors of tag teaming because the both revert the same material? That is also edit warring by definition. Please use AN/I if you feel there is such tag teaming going on and I suggest you be able to demonstrate such.--Amadscientist (talk) 04:48, 21 January 2013 (UTC)

• Regardless of who prefers what or what WP:THEWRONGVERSION is, this is the last stable format of the page prior to the edit war starting. I have, after seeing this mentiomed at AN/I, reverted to that version, as it is from that point that WP:BRD starts, not the changed version. There really should be a round of trouts handed out here as all of the participants in this edit war are long-standing editors who know how BRD works, how consensus works, and how discussion works. Please (< and I realise the irony of using that word here) discuss the proposed changes, not the changing back to the stable version after the initial, WP:BOLD and good faith change was reverted. Thank you. - The Bushranger _{One ping only} 06:39, 21 January 2013 (UTC)

The edit I made is in two parts. I am surprised that the alteration to the lead was reverted. S. Marshal, your comment in the "Undid revision 534022022 by PBS (talk): this is how it's appropriate to speak to volunteers." baffles me. The BLP wording is based on a legal imperative and my change reflects the language in BLP:

"unsourced or poorly sourced – whether the material is negative, positive, neutral, or just questionable – should be removed immediately and without waiting for discussion"

As to the second change this was discussed and agreed in December see Wikipedia talk:Verifiability/Archive 59#Burden of evidence sentence "When tagging or removing material...". So what ever else is said it can not be said that the wording "please state your concern" is not a stable version.

• Does anyone other than S Marshall object to the change in wording?
• Can I suggest that we refer to the changes separately as the "BLP change" and the "Burden change"

S Marshall what do you object to in one or both changes? -- PBS (talk) 09:38, 21 January 2013 (UTC)

Personally basic civility leads me to prefer 'Please' as the more appropriate version. Saying that - it probably should be followed by a sentence outlining the consequences of not complying. (This might be covered elsewhere, but when you make a demand, albeit nicely phrased, it goes down easier if it is straight away followed by the consequences). Absent that, 'Please' doesnt really have any teeth. Only in death does duty end (talk) 09:53, 21 January 2013 (UTC)

As assuming you are writing about the BLP change, "polite notice do drink and drive" and "Police notice do drink and drive" are not the same thing, the first is a polite request and the second a prohibition. The wording in this policy should in this area summaries BLP in such a way that it does not give room to wikilawyering that this policy is not in step with BLP. Besides we can not spell out the consequences of breaking the policy dictates of BLP in the lead of this policy, because it is not a mechanical process -- it depends on circumstances (and would take paragraphs to detail), and so it would bring yet more complication into the lead that is not needed. -- PBS (talk) 11:47, 21 January 2013 (UTC)

• PBS, I feel that the way to speak to volunteers is persuasively and courteously, and that it's inappropriate to talk to them as if we were giving them orders. Badly-written policies are an editor retention issue. I object to removing any instance of the word "please" from the policy except by replacing them with other words of at least equal respectfulness.—S Marshall T/C 12:17, 21 January 2013 (UTC)

I'm with you on speaking politely to volunteers and not bossing them around. But I would argue that your preferred version is bossier than the proposed version. I read the proposed version as a statement (not directed at anybody in particular) of how the content should end up, while I read your preferred version (even with the "please" tempering it) more as "giving orders" to the person reading it, directed to them specifically. Sincerely, North8000 (talk) 12:47, 21 January 2013 (UTC)

When it comes to unsourced information in BLPs, I do think we need something a bit stronger than a mild "please". Having unsourced (and thus potentially unverifiable) information in BLPs is absolutely unacceptable. Blueboar (talk) 14:13, 21 January 2013 (UTC)

• @S Marshall: You may be pleased to hear that I was inspired to write an essay by one of your edits above. Bishonen | talk 19:38, 21 January 2013 (UTC).

As a hitherto uninvolved editor, I support the 'should' wording. As I understand it, BLP is one of the most strictly enforced rules and the removal of unsourced content in BLP articles is a requirement not a request, and should be worded as such. And in my honest opinion, the idea that a simple statement of this strictly enforced rule will offend editors is frankly patronising. Also 'should' follows the wording on the BLP page. Is it too impolite there also? CarrieVS (talk) 20:13, 21 January 2013 (UTC)

• What I support above all else in this escapade is Black Kite's protection of the page. Honestly, are all those pixels being spilled over the advantages and disadvantages of saying "please"? Is that really such a major issue? Anyway, one semi-trend I see in this discussion is that the part about not letting bad stuff stay in BLP pages is something where there's a longstanding consensus that Wikipedia really, really wants that kind of stuff deleted from BLPs without delay. On the other hand, I don't really think we have consensus one way or the other about whether we should say "please" or "you ought to" in that sentence about the conditions for a WP:CHALLENGE. So, how about we split the difference? As a peacekeeping compromise, how about using the not-please language for the BLP sentence, and the please language for the other sentence? --Tryptofish (talk) 21:31, 21 January 2013 (UTC)

That is not a compromise its a fudge. There is broad agreement here that please in inappropriate in both places. -- PBS (talk) 23:12, 21 January 2013 (UTC)

To see where we're at, I did a quick count-up of opinions. Of twelve editors (including myself) who've commented on this thread:

• three have no opinion.
• two are in favour of 'please' (though one feels that it isn't strong enough without being followed by an additional statement)
• seven are in favour of 'should'.

Whilst I don't want to suggest that this be based on a simple majority vote, more than 3:1 does look like there is broad agreement in favour of 'should'.

To summarise the arguments:

• For 'please':
  • It's more polite.
  • It's already there.
• For 'should':
  • It's more strongly worded, requirement versus request.
  • At least one editor has disagreed that 'please' is more polite.
  • It's in the main BLP article.

So the question is, is it more important to word it strongly, or to be polite. Taking into consideration that it's not unanimous that 'please' is more polite, and that one proponent of 'please' has said that although they support it for politeness they don't feel that it is strongly enough worded without being followed by a statement of the consequences.

Discuss. CarrieVS (talk) 15:24, 22 January 2013 (UTC)

Did you include Black Kite in your survey? From the history of the article "Changed protection level of Wikipedia:Verifiability: Edit warring / Content dispute: enough I think (for the record I strongly support removing 'please'". -- PBS (talk) 19:17, 22 January 2013 (UTC)

I did not, as I only looked at this thread. So if Black Kite were included the numbers would be thirteen, three, two, and eight. CarrieVS (talk) 19:32, 22 January 2013 (UTC)

My own view is that although politeness is important, sufficiently strong wording to be in line with the policy in question (that it is imperative that such material be removed) should come first. It's not as though 'should' involves any insults or abuse; it's a simple, neutral statement of what should be done, whereas even if 'please' is politer (I am inclined to agree that it isn't particularly) it, to my mind, downplays the importance and makes it sound like a request - which is incorrect. CarrieVS (talk) 15:32, 22 January 2013 (UTC)

Let's just decide on that one narrow one-sentence question and then move on. This has gotten way too rough on a few folks. North8000 (talk) 15:34, 22 January 2013 (UTC)

Having had more of a read of the section below, is this discussion still going on? If we'd agreed to leave it be, then please don't take my above comments as trying to reopen the debate - the contrary, although I have a preference on the matter I don't think it's a major issue and was only trying to help resolve the argument; if we've come to an agreement then that's great and please ignore everything I've said since then. CarrieVS (talk) 15:51, 22 January 2013 (UTC)

• I'll add a semi-!vote. For the sentence in the lead, about BLP material, I prefer the language that does not use "please", because I think that BLP policy is so important that it goes beyond just a polite request. As for the sentence in the BURDEN section, I continue to be neutral, and I'm flabbergasted that anyone would see it as a big deal. --Tryptofish (talk) 23:03, 22 January 2013 (UTC)

I think what we really need now is comments from the two users who support 'please' (version #1 below).

To them: do you still disagree with 'should' (version #2 below), in the light of the 'summary of arguments' above and subsequent comments? CarrieVS (talk) 00:10, 23 January 2013 (UTC)

Let's decide the recent specific questions and move on

Question 1

Question #1 is choosing between the two versions of the first involved sentence that it was going between just before it was locked. Which (roughly) are:

• Version #1 Please remove unsourced contentious material about living people immediately.
• Version #2 Unsourced contentious material about living people should be removed immediately.

Responses

• I prefer Version #2 but would also be fine with version #1. The main reason is that the choice of words is less bossy sounding to volunteers. (yes, even with the "Please" in it, IMHO #1 is bossier because it is telling the reader to do something, whereas version #2 is more of a less-directed "somebody should" statement. And "Please" does not negate the inherent bossiness. My second weaker reason is that IMHO #2 is a bit more oriented towards the end result than the action to be taken to get there. And I would like to add that the thought anybody doing anything "wrong" here is either non-existent or too microscopic to discuss. And everybody involved does an immense amount for Wikipedia and should be thanked for that. Sincerely, North8000 (talk) 23:22, 22 January 2013 (UTC)
• Prefer Version 2. But isn't this just one of the two sentences that are under discussion? --Tryptofish (talk) 23:25, 22 January 2013 (UTC)

I dunno. It's the one involved in the exchange that got this locked. I figured lets just simplify/separate and resolve that little question. (?) North8000 (talk) 23:29, 22 January 2013 (UTC)

Yeah, there is. It's a sentence in the BURDEN section that grew out of a suggestion that you made. Both sentences were in the edits that led to page protection. (Welcome to tl;dr!) --Tryptofish (talk) 23:32, 22 January 2013 (UTC)

OK, let's freeze this a few minutes while I tidy up. Since I'm going to refactor a tiny bit, please feel free to changes yours. North8000 (talk) 23:36, 22 January 2013 (UTC)

OK, done. Wow, was I confused! North8000 (talk) 23:45, 22 January 2013 (UTC)

• Prefer Version 2. And would actually like to see Q1 and Q2 switched on this with the BLP issue stated as "ought to be" and the unsourced removal issue stated as "should be". But this is acceptable. — TransporterMan (TALK) 17:05, 23 January 2013 (UTC)
• Prefer version 2, but no strong opposition to 1. CarrieVS (talk) 17:13, 23 January 2013 (UTC)
• Prefer version 2... it's closer to the wording of WP:BLP Blueboar (talk) 23:17, 23 January 2013 (UTC)
• prefer version 2 -- TRPoD aka The Red Pen of Doom 23:24, 23 January 2013 (UTC)
• Although it looks to me like the responses to question 2, below, are all over the place, do we perhaps have a consensus here on question 1? --Tryptofish (talk) 20:55, 26 January 2013 (UTC)

Question 2

Question #2 is choosing between the two versions of the second involved sentence that it was going between just before it was locked. Which (roughly) are:

• Version "A" When tagging or removing material for not having an inline citation, please state your concern...
• Verison "B" When tagging or removing material for not having an inline citation, you ought to state your concern...

Responses

• I prefer Version "A" It is more polite sounding, and less guttural sounding. North8000 (talk) 23:44, 22 January 2013 (UTC)
  • It is not a matter of how it sounds it is a mater of precision. "ought to", "should", and "must". Please does not convey that sort of granularity as what is being asked could be any of those three wrapped up in candy. We do not help new users by not being precise. It is partly a matter of dialect is seem that in some English dialects "should" is closet to "ought to" and in others closer to "must" (this became clear from the big fuss over at AT and an ANI over the use of should). In this case we clearly mean "ought to". -- PBS (talk) 12:53, 23 January 2013 (UTC)

IMO at least in US English the format "Please xxxx xxxx xxx" 90% of the time is a mandatory instruction, (e.g. "please refrain from smoking" "please remove your car from the loading zone") so IMO it certainly isn't weak. North8000 (talk) 13:08, 23 January 2013 (UTC)

• I prefer Version "B" "C" "Please" is not mandatory, it's a request, not a mandate. In some contexts it implies that a mandatory order will probably promptly follow if you don't comply with the request, but in it's initial form it's a request. (That's the reason that official street signs, as opposed to signs in a shop, say "No parking" rather than "Please don't park here": if they said "please" they wouldn't be enforceable.) In any event, it's ambiguous and open for the interpretation that it's mandatory, which it should not be in this case. "Ought" is clearly non-mandatory and is preferable for that reason. — TransporterMan (TALK) 17:05, 23 January 2013 (UTC) Supplement: Change support from B to C, now it has been proposed. — TransporterMan (TALK) 15:04, 25 January 2013 (UTC)
• Prefer version A, but only for 'nicer-sounding' wording - I don't think there's a big difference between these two. CarrieVS (talk) 17:13, 23 January 2013 (UTC)
• I prefer Version A, and I believe that any heretics who disagree with me should be burnt at the stake. This is the most important issue that is being raised anywhere on Wikipedia. There is no room for compromise. --Tryptofish (talk) 17:44, 23 January 2013 (UTC)

• North8000's post at 13:08, is a good example of why "please" is not very useful, because in this context as it can be misunderstood -- in Britain it would not be understood the way North8000 describes it. From what North8000 writes it sounds even more problematic than "should" (for the should saga see This AT talk page discussion (and collateral damage further up the same archive and here, here, and here) -- PBS (talk)

• Version A per Tryptofish. This is auto de fe, and ontological cri de couer stuff, people. Alanscottwalker (talk) 23:06, 23 January 2013 (UTC)

• If we have to include behavioral instructions like this (which I question), I would suggest:

Version C - When tagging or removing material for not having an inline citation, it is helpful to state your concern. Blueboar (talk) 23:14, 23 January 2013 (UTC)

• support C and Blueboars general position.-- TRPoD aka The Red Pen of Doom 23:24, 23 January 2013 (UTC)
• Oppose C, and this time I'm serious. That's going beyond "please", and saying "pretty, pretty, pretty please". But the fact that C would even come up for discussion shows that insistence on B is also kind of "partisan". --Tryptofish (talk) 00:24, 24 January 2013 (UTC)
  • Agree. No strong opinion between A or B, but not C please.CarrieVS (talk) 11:24, 24 January 2013 (UTC)
• Support C, as "please" is already prone to being misconstrued and I find "ought to" to be an awkward construction. My understanding is that what we're talking about here is a recommendation, not a mandate...which I don't believe we're authorized to issue in any case(?). Doniago (talk) 16:01, 24 January 2013 (UTC)
  • it is not a mandate in current policy or application, and a handful of people cannot make it so- although it appears that it is indeed the intention of some people here. -- TRPoD aka The Red Pen of Doom 19:19, 24 January 2013 (UTC)

Is WP:V primarily a content policy or a behavioral policy?

This question came up in the above section (relating to BLPs and "Please")... but I think it may also be what underlies our recent debates about the entire BURDEN/CHALLENGE section. Some of us (myself included) see WP:V as being primarily a content policy. Others, however, apparently see it as being primarily a behavioral policy. This disagreement is fairly fundamental (affecting both what the policy says, and how we phrase it). I think we need to explore and discuss this underlying disagreement before we can make any progress on language proposals. If we can not reach a consensus on this disagreement, we will never be able to reach a consensus on language. Blueboar (talk) 16:09, 21 January 2013 (UTC)

• As I commented above, WP:V is a content policy. It describes what standards the contents of an article have to meet before they are acceptable. WP:BURDEN is really the only behavioural section, and I could happily see it spun out somewhere.—Kww(talk) 16:13, 21 January 2013 (UTC)

• With all due respect for Blueboar and Kww, the reality is that WP:V is a behavioural policy; their position is that it should be rephrased as a content policy (with the consequent spinning out of WP:BURDEN). I've expressed counterarguments above, but rather than rejecting this idea out of hand, I suggest that Kww and Blueboar are invited to propose a draft to show us specifically what changes are intended.—S Marshall T/C 17:06, 21 January 2013 (UTC)

• Pretty much undoing everything you did to the policy between Nov 9 and Nov 11. It was a content policy with occasional behavioural direction, and this set of edits, primarily by you changed it into a behavioural policy with a little be of content direction. There's a few other changes to make to the earlier version, such as changing "Take care to avoid plagiarism and breaches of copyright when using sources. Summarize source material in your own words as much as possible; when quoting or closely paraphrasing a source use an inline citation, and in-text attribution where appropriate" to "Material from other sources should be summarized as much as possible, avoiding excessive quotation. Quotations must be clearly indicated, and both quotes and close paraphrases must be cited in order to avoid plagiarism and copyright issues". Again, the focus is content, not behaviour. The guideline needs to instruct us as to what the content has to look like. I don't think we need to present a complete proposal in order to gain consensus that the earlier changes weren't for the better.—Kww(talk) 17:30, 21 January 2013 (UTC)

While behaviour impacts how many of our policies will be implemented, those policies should not be considered behavioural policies in my opinion, and as such the weight of those policies of which verifiability is one should focus on what the content and its sources looks like rather than how an editor behaves in order to get that content into an article. At the same time, how an editor edits does impact content policies so perhaps some mention of editor action is appropriate, while not a focus. The weight of the policy would be on content but a small amount of weight could be given to how the editor impacts that content. I don't think this is a black and white issue, either one or the other, but more of a focus/weight issue.(olive (talk) 17:44, 21 January 2013 (UTC))

Which is why I phrased my question with the word "primarily". Blueboar (talk) 17:58, 21 January 2013 (UTC)

I'm adding a view that happens coincides with yours. I did assume editor input would be OK here. :O)(olive (talk) 18:34, 21 January 2013 (UTC))

It's mostly a content policy. More to the point, it sets conditions for the presence of material in articles.

It reflects a bit on actions to implement that restriction (ostensibly controls on insertions, but in reality it's about removals, not insertions). While some of these may be prima facie behavioral, in the in the end they just affect removal of material....what gets taken out and how easy it is to take it out.

Sincerely, North8000 (talk) 18:38, 21 January 2013 (UTC)

• Well, I feel a bit blindsided by all this. It was carefully discussed in November, and I specifically mentioned most if not all of the changes I made to convert the policy to a behavioural one. At that time, everyone who contributed a view, approved (see the bottom of WT:V archive 58). Nobody reverted any of the changes at the time. And now all of a sudden it's become an issue because I said "please"? Wikipedia's a bizarre place.

  I still think trying to use policies to describe the Platonic ideal of a perfect article makes the text seem vague and inscrutable, but I suppose if the consensus is to take this retrograde step then it behoves me to accede. It's still important that the new text uses courteous and respectful language towards our volunteers.—S Marshall T/C 18:40, 21 January 2013 (UTC)

Silent consensus is the weakest kind. Consensus can change. WP:V is a content policy and should not be altered unilaterally to a behavioral guideline unilaterally.--Amadscientist (talk) 18:51, 21 January 2013 (UTC)

It wasn't a silent consensus, of course; neither was it a unilateral change. There was positive agreement: diff, diff, diff; and Blueboar specifically averred that "the idea has been presented and discussed on the talk page, and there seems to be at least an initial consensus to support it" (diff). Please read and check before hitting "submit".—S Marshall T/C 19:14, 21 January 2013 (UTC)

I apologize for that S Marshall. I came in with an opinion rather than a definitive position. I'm not attached to either way of viewing the policy except that logically, it makes sense to me to clearly delineate behavioural policies from content policies and to make sure our wording reflects that delineation. Taking what seems to me to be a content policy, it is about verfiablity rather than the act of verifying content, would seem to me to make this policy a content driven one. I do tend to see things in an overarching way and then to go to specifics so my leaning is to fist decide what kind of policy we are dealing with then to decide on its wording based on the categorization. As I said, adding thoughts rather than wanting to enforce anything.(olive (talk) 18:58, 21 January 2013 (UTC))

STOP... Look, I did not mean this thread to become a referendum on any one editor's edits. Consensus can change... and sometimes a whole chain of edits (by multiple editors) can gain consensus when and as they are proposed - only to be reconsidered later (after people have had a chance to live with them for a while, and see the impact that they have). That is what I think has occurred here. It is only now that I realize how far this policy has shifted its focus from content to behavior... So now I question that shift (even though I contributed to that shift in the past). Thus, I am re-examining and rethinking some of the decisions that I (and all of us, as a group) have previously made. Blueboar (talk) 20:36, 21 January 2013 (UTC)

I just like the version promoted by Amadscientist/Kww a bit more than that promoted by S Marshall. Nothing more that that. S Marshall, I think that I remember when you put it in; I think that it was on a "good grammar" basis. I sort of looked at it and said "not my preference, but it's fine". Now with someone making a change, the discussion got re-opened. I am sorry how this played out. Sincerely, North8000 (talk) 19:00, 21 January 2013 (UTC)

• It's clearly a content policy, not a behavioural one. Reyk _YO! 21:22, 21 January 2013 (UTC)

I consider all policies to be both content and behavioral. The degree to which they are one or the other depends upon how they're worded (and how they are customarily enforced, but the wording is the primary criteria). Regards, TransporterMan (TALK) 21:31, 21 January 2013 (UTC)

My reason to lean towards content rather than behavior is more mechanistic than anything. To state it in a usefully overreaching way, the end goal is always content. So to get there you can either specify the one place that you need to end up, or you can specify what the next action (towards that end) should be for each of the million situations. I think that "one" makes for a shorter and more succinct policy than the "million" .  :-) North8000 (talk) 15:30, 22 January 2013 (UTC)

It is both content and behaviorial. Of more metaphysical concern: will this policy ever be left alone? Alanscottwalker (talk) 00:58, 23 January 2013 (UTC)

No, it never will be. :-)

I think it's primarily about what content should consist of. But it's silly to pose this question as though it's all-or-none (done, I suspect, to make a point about the footnote being discussed above). Our NPOV policy is primarily about content: that content should have a neutral POV. Nonetheless, it includes WP:NPOV#Handling neutrality disputes. WP:PG is our "policy about policies". Nowhere does it state that policies must be neatly divided into ones dealing with content only, and ones dealing with behavior only, with a strict separation. Anyone who thinks that this discussion is going to tell us what WP:V should or should not include is going to be disappointed. --Tryptofish (talk) 17:54, 23 January 2013 (UTC)

True... but this discussion will hopefully guide us in figuring out how to say what we do include. The phrasing of a policy focused on content is going to be different from the phrasing of a policy focused on behavior. More: "Information should be..." and "Material should not..." - Less: "Editors should..." and "Do not..." Blueboar (talk) 18:12, 23 January 2013 (UTC)

It does, however, bear noting that there's only one known method for information to get in articles. Editors are adding it there. So the distinction is really moot: Information doesn't materialize out of the spirit world into articles; it is added by editors. The difference is meaningless in the context of this policy. --Jayron32 18:16, 23 January 2013 (UTC)

And that's the point I hoped would emerge from this discussion. Information doesn't materialize out of the spirit world into articles; it is added by editors. I want to add that it's the editors, not the articles, who read policy pages. So it's the editors we should be speaking to.—S Marshall T/C 12:46, 24 January 2013 (UTC)

Given that this discussion was motivated by the discussion about the footnote, I think it's worth pointing out explicitly that, even though a number of editors (including me) feel that the primary emphasis of WP:V is about the kinds of content to include or exclude, there is clearly consensus against the position that WP:V must be purged and scrubbed clean of every last scrap of language that could possibly be seen as edging into behavior. It's not all-or-none. --Tryptofish (talk) 21:00, 26 January 2013 (UTC)

Bot

At an RfC elsewhere, I posted the following view, except that I modified it for presentation here so that the editor isn't mentioned by name.

From the editor's edit summaries, it appears that the editor has been deleting material according to the criteria that the material is unsourced and has been tagged for a long time. This seems like a task that can be done by a bot. So I went over to the Village Pump to see how editors felt about having such a bot for the case of unreferenced sections that have been tagged for more than a year. They didn't like it. One administrator over there wrote, "No, a person would need to look at each section to see if it should be removed, not just removed without any review."[7]

If the editor has a reason to delete unsourced material other than it has been tagged for a long time, I would suggest that the reason should be indicated in the edit summary. If there is no other reason than unsourced and tagged for a long time, I would suggest that the editor refrain from deleting the material.

Note the second sentence of the following excerpt from the section Burden of evidence of WP:Verifiability.

"Any material lacking a reliable source directly supporting it may be removed. Whether and how quickly this should happen depends on the material and the overall state of the article."

--Bob K31416 (talk) 15:08, 24 January 2013 (UTC)

FWIW, if material's been tagged for a long time and an editor has concerns about it, I believe the material probably should be removed from the article, though it should be handled on a case-by-case basis and assuming we're talking about more than a sentence or two I would generally advocate moving it to the article's Talk page over a simple deletion. I definitely don't think a bot should make those kinds of decisions though, as these situations tend to be unpredictably controversial enough without that kind of help. Doniago (talk) 15:56, 24 January 2013 (UTC)

Re "FWIW, if material's been tagged for a long time and an editor has concerns about it, I believe the material probably should be removed from the article..." — Did you mean concerns about issues in addition to it being unsourced and tagged for a long time? --Bob K31416 (talk) 16:20, 24 January 2013 (UTC)

If we assume that the information was originally tagged because an editor believed the material may not be verifiable, and it then sat around for a long time (let's say three months or more, different editors have different ideas about that), and nothing was done about it? Then no, I believe moving the information to the Talk page is justifiable at at that point regardless of whether there are concerns beyond verifiability. I don't believe unsourced information should be allowed to remain in articles indefinitely, though as I said, how it should be addressed should be handled on a case-by-case basis and different editors tend to have different feelings on what the best approach to resolution ultimately is. Doniago (talk) 16:46, 24 January 2013 (UTC)

Bob, you are conflating two separate issues... The issue of whether an edit summary should be required (or not) has no baring on the issue of whether bots should be allowed to automatically remove material. After all, a bot can be instructed to leave an very detailed edit summary. Imagine a removal accompanied by the following edit summary: "This material has been tagged as needing a citation, but has been left unsourced for many months. It is thus assumed to be unverifiable. If this assumption is incorrect, return the material with a citation." That summary could be added by a human or a bot. If a human left it, it would be an acceptable edit summary. So why would it not be an acceptable summary if left by a bot? The difference is that the human makes a judgement... and could have chosen not to remove. A bot can not make that judgement and would always remove. On the other hand... the bot would always leave the edit summary, while a human might hit "save page" without remembering to do so. Blueboar (talk) 16:53, 24 January 2013 (UTC)

I don't know whether it would be possible for a bot to locate sections that have been tagged "long-term", scan them for a lack of citations, and move the info to the article's Talk page while leaving a reasonable edit summary, but it's an interesting question. I find myself uneasy with the notion of it being automated, but I think that's only because my own experiences with dealing with unsourced material lead me to believe that editors would start screaming over the "deletion" of information. Doniago (talk) 17:22, 24 January 2013 (UTC)

I certainly would object to a bot removing tagged material... a human can choose to remove or not... a bot would always remove, even when it is inappropriate. Blueboar (talk) 17:38, 24 January 2013 (UTC)

There's also a possible practical issue: you say a bot could easily remove the unsourced material. And, without knowing too much about programming, I expect that it could with a high degree of accuracy. But that section doesn't exist in isolation: if a later sentence refers back to it, or some such, the article then wouldn't make sense. Of course, that and the issue that the bot always removes could be remedied by patrolling the bot's edits, but then there's no point in having the bot remove it - it would be as easy for the patroller to do it. But identifying these sections could certainly be done by a bot, so how about instead of having the bot remove the information, it starts a talk page section, or lists the articles on a 'long-term unsourced material' page? CarrieVS (talk) 17:49, 24 January 2013 (UTC)

Category:Articles with unsourced statements? Doniago (talk) 17:58, 24 January 2013 (UTC)

Let me clarify a few things. I'm not proposing a bot but using it to make the point that editors shouldn't behave like bots and delete material only because it is unsourced and has been tagged for a long time. The editor's edit summaries were mentioned to discern the reason for the editor's removal of material. --Bob K31416 (talk) 18:07, 24 January 2013 (UTC)

Re: an editor deleting material "only because it is unsourced and has been tagged for a long time." If they think the removal is in the best interest of the article or section, I have no problem with it. The editor is not behaving like a bot, because they are making a choice (a choice that they might make differently in a different article). Blueboar (talk) 18:14, 24 January 2013 (UTC)

Then they are using criteria other than just that it is unsourced and has been tagged for a long time, which is all that I'm saying they should. --Bob K31416 (talk) 18:27, 24 January 2013 (UTC)

True. But that is the case with every removal. No one removes material that they think should stay in an article. They only remove material when they believe it should be removed... when they feel that removing it is in the best interest of the article. Blueboar (talk) 19:30, 24 January 2013 (UTC)

(edit conflict)Re "But that is the case with every removal." — That's quite a sweeping statement. It's not clear why you can't conceive of an editor who deletes unsourced material only because it has been tagged for a long time because the editor has chosen that way to contribute to Wikipedia, i.e. to rid Wikipedia of chronically unsourced material. --Bob K31416 (talk) 20:54, 24 January 2013 (UTC)

I think this is one of those areas where it is OK to do something to any particular article, but at the same time not OK to do it to 1,000 articles. If an editor wants to remove content from one article, that's fine, but it they started doing it in a bot-like way that suggests they are not considering each article separately, that would be bad. This can be hard for people who wish we had a firm rule - "can I remove it, or can't I" - that they could apply without further thought. But that's not how Wikipedia works, and in particular it's not how we handle WP:V. Editors can removed material even if it is sourced... and add it even if it is unsourced. Of course they need to be ready to take responsibility for their edit afterwards. — Carl (CBM · talk) 20:43, 24 January 2013 (UTC)

The problem is that we have no way to know whether someone is acting in "a bot-like way" or "carefully considering each removal"... you might carefully review 500 articles with long standing tags... at 400 of them you might decide to not delete the tagged information (and only decide to delete things at 100 of them).

Now we get to the problem... how do I know what you did? I could check your edit history... but your edit history does not reflect the 400 times you did NOT make a deletion. All it reflects are the edits you DID make... those 100 articles where you DID make a deletion. Should I assume you went on a deletion "spree", acting in a bot-like way? No... If I did that, I would be assuming bad faith based on flawed, incomplete evidence... what I should do is assume good faith. Blueboar (talk) 21:51, 24 January 2013 (UTC)

We could look at the number and length of time between the edits, among other things, to get a sense of whether it is feasible for an editor to review the material as it is being removed. My point, though, is that there are some users who falsely extrapolate from "you can do it to one article" to "you can do it to every article". Things are rarely that black and white. — Carl (CBM · talk) 23:04, 24 January 2013 (UTC)

There are some pretty good indicators that someone is behaving like a bot, and not giving any level of consideration to the content with which they're working. For example if the timestamps of their edits indicate that there is very little space between edits, then we can say that they're acting like a bot. In the case to which Bob is referring there is less than a minute gap between the tagging one article, then making two edits to remove around 2200 characters from another article. Another example was 5 edits to 5 separate and unlinked articles removing content with the edit summary "d uncited" or "d per tag" each time in the space of approximately 4 minutes. ˜danjel [ talk | contribs ] 23:17, 24 January 2013 (UTC)

I would question that logic. It shows the editor worked on a lot at once, but not necessarily that (s)he just opened each page and summarily deleted the stuff. They might have, but this isn't the only place I've seen it suggested to be conclusive, and I disagree. For instance, I disambiguate links. I usually have anywhere from three to a dozen or more tabs open; in each case I find the link, figure out what needs to be done, and open the edit window. While it loads I do the same with all the rest, and then edit and save them all in quick succession, leaving me with a dozen edits within a couple of minutes. Often it did only take me a moment to work it out, but sometimes it needs quite a bit of research and it looks just the same. CarrieVS (talk) 16:34, 25 January 2013 (UTC)

In your personal case of working with many tabs and implementing all the edits at about the same time, that activity would be supported by looking at your contributions just before you made the edits in the tabs. There wouldn't have been any contributions while you were working on the tabs. In the multi-article deletion example I gave below, I checked the editor's activity just before the edits and it was filled with activity on other articles. Regards, --Bob K31416 (talk) 17:06, 25 January 2013 (UTC)

Ok. If it's not just that it's a lot of edits all in one go, it does sound pretty well conclusive. CarrieVS (talk) 17:15, 25 January 2013 (UTC)

Blueboar, You seem to be ignoring the edit summaries of the deleting editor, e.g. "d per tag". From what you've written so far, it seems that you agree that deleting unsourced material only because it has been tagged for a long time is wrong and bot-like. But let's see. Suppose when asked, the editor says that the only reasons for deleting the material was that it was unsourced and it was tagged for a long time, and that it is the right thing to do. Would you agree with the editor? --Bob K31416 (talk) 01:55, 25 January 2013 (UTC)

Actually, I was assuming that the editor did not leave any edit summary. "d per tag" would certainly explain the edit and would be more than acceptable to me. Sure, I might have reached a different conclusion had I been the one deciding whether to remove or leave it tagged, but I would understand why the editor made the choice he did.

Sorry if I have not made my view clear... I am acknowledging that others think that deleting unsourced material "only because it has been tagged for a long time" is wrong and bot-like... however, my view is that it is not "wrong" to remove material "only because it has been tagged for a long time". It may not always be the right thing to do, but it is not wrong. It is simply a binary choice we can make. Neither right nor wrong. And while such actions can be bot-like, they are not necessarily bot-like... more importantly, there is no way for anyone except the deleting editor himself/herself to know whether the actions were bot-like or not. We can not read other people's minds. We should assume good faith... and should not ascribe bad faith motivations to other people's edits without very, very strong evidence of bad faith. Blueboar (talk) 04:19, 25 January 2013 (UTC)

I'm assuming that the editor is acting in good faith, but rather is misinterpreting this policy. My interpretation of the Burden section regarding this case is that material shouldn't be deleted just because it is unsourced and tagged for a long time. I'm using these two sentences,

"Any material lacking a reliable source directly supporting it may be removed. Whether and how quickly this should happen depends on the material and the overall state of the article."

The second sentence says that whether the material should be deleted depends on the material. With an edit summary of "d per tag", it looks like the editor isn't considering the material that is being deleted but is deleting it just because there is a tag. BTW, if the editor did make a good faith effort to consider the credibility of the material or consider whether the material appeared in any reliable source, then the editor could modify the edit summary to read something like "d questionable material per tag", or "d material that may not be in any reliable source, per tag", and thus effectively state compliance with the above excerpt from Burden.

What is your interpretation of this excerpt from Burden regarding this case? --Bob K31416 (talk) 13:31, 25 January 2013 (UTC)

Has anyone approached the editor in question regarding these concerns? As has been discussed before, we can't mandate what editors leave in the edit summary (or that they leave one at all), but we could ask them to leave a more specific summary, if that's really a concern. Personally, I read it as, "The section's been tagged for needing sources for a long time, and nothing was done about it, and I'm concerned that the material isn't verifiable."

A related question would be whether they're deleting the material or moving it to the Talk page. If the latter, then I suspect I wouldn't have any problems with what they're doing. If the former, then again, it couldn't hurt to suggest they move the material rather than deleting it.

In any case, taking that excerpt at face value, the information could have been deleted without even having been tagged. If it was tagged long-term then I find the information especially likely to be suitable for deletion/moving to the Talk page.

As has been brought up numerous times before, in many if not all cases, the best (or at least least aggravating) approach to concerns about such deletions is to provide refs for the unsourced material and move on. Doniago (talk) 14:27, 25 January 2013 (UTC)

In the quote box at the beginning of this section, I wrote, "If the editor has a reason to delete unsourced material other than it has been tagged for a long time, I would suggest that the reason should be indicated in the edit summary." --Bob K31416 (talk) 15:23, 25 January 2013 (UTC)

In regards to the question of whether anyone has approached the editor concerned, the answer is "yes". On the talk page of one of the affected articles (because the source was very obvious), on the user's own talkpage, twice, and in an RFC/U. Each time, the response has been approximately the same, he's said that unsourced information can be removed per WP:V, and little more (he has extreme tunnel vision regarding what WP:V says, and doesn't read when any other part of V, or any other policy, is raised). In regards to your second point, the information is not being moved to the talk page, nor is it actually being raised at all on the talkpage.

My concern is that the information is plainly easily verifiable. In one instance, the information deleted related to the main campus of a major Australian university. In another, it was about the sporting achievements of a school that has been the subject of media interest. In another two, the information could have been sourced using one of the other sources already given in the article. In the latest instance, the information deleted was easily source-able, and was the information that conferred notability on the subject of the article. No reasonable person would consider the information deleted to be unverifiable. For these articles, I was able to fix the problem. However...

In regards to the provision of references, yes, it would seem reasonable for someone else to provide the detail. However, the range and speed of the removal makes that difficult. There is currently 28 29 separate instances listed at the RFC/U, and counting, and I haven't been paying that close attention (up to when I initiated the RFC/U, I only noticed when the editor hit pages on my watchlist). ˜danjel [ talk | contribs ] 14:39, 25 January 2013 (UTC)

I'm hesitant to comment further on the matter without seeing the specific comments involved...in part because my experience is that editors who feel information should not have been removed from an article frequently express their concerns in less than diplomatic terms...I'm reminded of a saying about honey vs. vinegar.

Since you've said that the editor isn't moving the info to the Talk page, I think it would be a reasonable compromise to ask if they would be willing to do that rather than outright removing it. It's not that difficult or time-consuming, and if there's a large volume of information it's easy enough to use a collapsible box. I am of course assuming that that option hasn't been raised. I do acknowlege that, based on your description of events, it sounds like the editor themselves may not be handling this as well as they could.

I don't think it's really kosher to say that information is "plainly easily verifiable". Different WP editors may have differing levels of access to references, and the fact that the information did go unsourced for however long it did, to me, is an implication that the information may not have been easy to verify. Or no editors cared to do the legwork.

Without seeing the editor's specific changes, I'm not inclined to consider them problematic (which doesn't mean they're not, obviously), but another possibility in terms of a compromise with them might be to suggest that they limit the frequency of their large-scale deletions of unsourced material. You listed 28, but was it 28 in two days? Two months? Two years? I'm not sure how frequently I make changes on that scale myself, but if an editor reasonably approached me with concerns about such, I'd like to think we could determine some sort of reasonable rate at which to make such changes...assuming there are concerns that not all of the changes are reasonable in any case.

Obviously, this is just my opinion on the matter. I'm well aware that other editors have different feelings about this sort of thing. Doniago (talk) 14:59, 25 January 2013 (UTC)

I'm not willing to post the link to the RFC/U for fear of it being considered WP:CANVAS-ing. By "easily verifiable", I mean that a simple google search would provide multiple sources for particularly non-contentious information (such as, for example, the location of a major university's main campus). At the outset (when he first hit my watchlist), I asked "are you challenging it because you genuinely believe that there is a question as to whether it is true or not?" He responded "yes", that he believed that a long established Australian school, working in the English school tradition, might buck the trend against the other several thousand other Australian schools and not have school houses.

The editor works in spurts. The latest was 12 in a single day, in amongst a large number of other edits of various types (but primarily tagging). A couple of those edits, where information was removed, were made within the same minute, and 5 more were made in a 4 minute period (as I've stated above). ˜danjel [ talk | contribs ] 15:17, 25 January 2013 (UTC)

Fair enough. If it hasn't already been done I might recommend approaching the editor with the compromises I mentioned up above. An effort should be made to determine whether they're willing to adapt or refuse to consider options that might be more productive. Doniago (talk) 15:34, 25 January 2013 (UTC)

(ec) I almost entirely agree with Blueboar on this issue, except that I do believe that it is possible to detect patterns in which mass-deletion editors are not engaging in the best practice of trying to find sources for unsourced material before deleting it. (Specifically, if a editor is making mass deletions and isn't also sometimes adding sources rather than deleting the material then it's pretty clear that the editor is acting like a bot.) The sanctionable behavior there is not violation of this policy, per se, but a generalized failure to edit in the best interests of the encyclopedia. What I object strenuously to is this policy being used to attempt to sanction or even criticize an editor who in a single instance, or even occasionally, deletes unsourced material (tagged or untagged) without stating a reason for doing so or offering a reason such as "d per tag". I object even more strenuously to it when the material that has been deleted in that single instance or occasionally has, indeed, been tagged as unsourced for awhile and/or when the editor exercises some care by copying the material to the talk page. Regards, TransporterMan (TALK) 14:48, 25 January 2013 (UTC)

(edit conflict)I agree with you that the mass-deletion editors with bot-like behavior are the concern here, and I thought your other points were good too. --Bob K31416 (talk) 15:03, 25 January 2013 (UTC)

So does 29 instances, including at least 6 that were easily source-able with the simplest of source searches (either googling, or looking at other sources already in the article) count as occasional, or is that crossing a line? ˜danjel [ talk | contribs ] 14:59, 25 January 2013 (UTC)

(ec) A quick side comment to Danjel, regarding the 6 that you say were easily source-able... have you stepped up to "Fix the Problem" by returning the information with one of those easily found sources? If not, you are part of the problem and not its solution. Blueboar (talk) 15:24, 25 January 2013 (UTC)

Yes, I have, as I said above "for these articles, I was able to fix the problem." They were (mostly) within my realm of interest here on wikipedia (school or academic institution articles, hence how I came into the situation, because articles on my watchlist were hit) or were blinkingly obvious (i.e., an acceptable source was elsewhere in the article already). I would not be so critical if these problems were not so easily fixed. ˜danjel [ talk | contribs ] 15:26, 25 January 2013 (UTC)

Thanks for doing that. Second question... would you have fixed the problem if the information had not been deleted? Blueboar (talk) 15:43, 25 January 2013 (UTC)

As I noted up above, I think the frequency of the edits is pertinent. Were these 29 instances in 5 minutes, or six months? Doniago (talk) 15:11, 25 January 2013 (UTC)

@Danjel: I don't think that the number or the apparent ease of sourcing necessarily indicates, alone, that they were improper, but could raise a suspicion. To say it differently: the number or apparent ease of sourcing could provide a starting point into a community investigation but not an ending point. Regards, TransporterMan (TALK) 15:33, 25 January 2013 (UTC)

And that's the course that I've run. That being said, the difference in opinions expressed between here and there is really curious. ˜danjel [ talk | contribs ] 15:36, 25 January 2013 (UTC)

OK... please take a look at my last 5 edits (where I removed information with long standing tags)... did I just act in a bot-like way, or do anything "wrong"? Blueboar (talk) 15:40, 25 January 2013 (UTC)

In my opinion, there is a massive qualitative difference between your edits and the edits about which I'm talking. I'll list your diff's and explain why for each. Keep in mind, I'm not going to the point of accessing academic search engines or anything much more than google and news.google.

1. diff - a source to support the information removed would not be easily found. In fact, the information is possibly somewhat contentious.
2. diff - seems WP:OR to me, therefore there would be no source, or at least nothing easily found; I also can't access the reference preceding the information you removed.
3. diff - again issues with WP:OR, obvious from the use of the word "most...", also information that's difficult to find.
4. diff - I actually thought this one might be your downfall, but I can't seem to find anything. I spent about 10 minutes looking, and actually did bother to take a look through some more complicated sourcing options too (despite what I said above).
5. diff - there are a few sources for this, such as [8], and [9], but I'd be a little concerned about the reliability of those sources, because I'm not sure if their information came from wikipedia, and would probably raise them over at RSN before re-adding them (what do you think?). ˜danjel [ talk | contribs ] 15:54, 25 January 2013 (UTC)

I am glad you agree with my decision to remove in these cases... but that wasn't my question. My question was about my behavior... did I act in a bot-like way? Look at my edit summary... look at the frequency of my edits. Would you assume that I acted in a bot-like way? Blueboar (talk) 16:07, 25 January 2013 (UTC)

That you have only hit content which is genuinely difficult shows that you are not "indiscriminate", as one would expect a bot to be, as the editor at hand has been (as in the examples laid out above). ˜danjel [ talk | contribs ] 16:15, 25 January 2013 (UTC)

OK, let me share my the impression I have formed from your comments so far (with the understanding that impressions can be inaccurate)... It appears that you are not really upset about editors acting in a bot-like way and removing "material with long standing tags"... you are upset about people removing long-tagged material that you don't think should be removed. In other words, it seems your concern is over the nature of the material being removed, and not the behavior of the editor doing the removal. Would you agree with this? Blueboar (talk) 16:47, 25 January 2013 (UTC)

Indiscriminate deletion is definitely a behavioural concern, so I wouldn't agree with that summary. If, as with you, the material was genuinely hard to source for some reason or another, then such deletions would not be a problem. ˜danjel [ talk | contribs ] 16:03, 26 January 2013 (UTC)

(edit conflict)Doniago, Here's an example.

07:20, 7 December 2012 (diff | hist) . . (-1,455)‎ . . Zayed University ‎ (→‎Academics: d uncited per tag)

07:18, 7 December 2012 (diff | hist) . . (+35)‎ . . Quaid-i-Azam University ‎ (Added {{ref improve}} tag to article (TW))

07:18, 7 December 2012 (diff | hist) . . (-1,217)‎ . . University of the Punjab ‎ (→‎Academics: d uncited per tag)

07:16, 7 December 2012 (diff | hist) . . (-3,029)‎ . . Allama Iqbal Open University ‎ (→‎History: d uncited per tag)

07:14, 7 December 2012 (diff | hist) . . (-2,059)‎ . . Yarmouk University ‎ (→‎History: d uncited per tag)

07:12, 7 December 2012 (diff | hist) . . (-2,096)‎ . . Sherborne School ‎ (→‎Pursuits and societies: d uncited per tag)

--Bob K31416 (talk) 15:48, 25 January 2013 (UTC)

(edit conflict):@Danjel: It merely appears to me that the investigation is not supporting your position. (I have no opinion whether or not it should or should not, having not looked at the diffs at the RFC/U, but just observing what others have said there.) But the opinions being expressed at the RFC/U are not unrepresented here; they're very much in line with those expressed here by me, Blueboar, and Doniago, and probably a couple of others. Moreover, the position I took above is, I am the first to acknowledge, fairly theoretical and I have elsewhere in this discussion also said that it is also my opinion that if we are going to sanction people for engaging in mass deletions in a way which is not in keeping with best practices that we would be well-advised to clearly say so here at V so as to remove any doubt that such behavior is sanctionable. Until we do so, editors can in good faith argue that there are no existing grounds on which to sanction an editor for that behavior, and that is I think exactly what is being argued at the RFC/U. Regards, TransporterMan (TALK) 15:57, 25 January 2013 (UTC)

Sorry about accidentally deleting your post, had a confusing edit conflict page.

The investigation at the moment is essentially arisen as a result of misdirection. Most of the commenters are arguing from the perspective that contentious or difficult to source information should be able to be deleted without problem, per what the editor in question said in his response. But that's not the question. It is easily source-able information where I find the problem. The first group of diffs show that.

I think that heading off those actions should be done, and highlighting that in flashing red letters will end that RFC/U. ˜danjel [ talk | contribs ] 16:03, 25 January 2013 (UTC)

But how does one define "easily" sourceable?

What I find most troubling is that we spend hours and hours debating (in sometimes bruising discussions) whether or not information should have been removed and what the policies regarding removal should be when in many cases we could be turning up sources and reinserting the information and not even needing to discuss the removals. It sometimes makes me wish the policy was boiled down to - "Anyone may add unsourced information, but anyone else may remove unsourced information. At that point, anyone is welcome to re-add the information if a reliable source is provided." Doniago (talk) 16:12, 25 January 2013 (UTC)

Well, in these cases, it was a matter of highlighting a word, or typing it into the google box on the top right of my screen, clicking "search", and then looking at the top one or two non-Wikipedia (or copied from WP) results/

I'd actually prefer for all the policies to be boiled down to "don't be a bloody tool, you asshole" (that is almost what the school rules are at most of the schools where I work, more or less), which could simultaneously deal with every aspect of WP:Policy, but... Maybe except WP:CIVIL. ˜danjel [ talk | contribs ] 16:19, 25 January 2013 (UTC)

One of the down sides to being the "Encyclopedia that anyone can edit" is that "anyone" includes bloody tools and assholes. Blueboar (talk) 16:47, 25 January 2013 (UTC)

Bot — opinion summary

I'm not sure how to summarize the opinions here, so I'll ask the editors here a question to clarify for myself one aspect of their opinions which I think is the main issue. Do most editors think that it is OK for an editor to make mass deletions of unsourced material tagged for a long time, with edit summaries that only say "delete per tag"? Thanks. --Bob K31416 (talk) 13:06, 26 January 2013 (UTC)

• The problem is that you are phrasing this as an "always" proposition. Sure, there may be occasional situations where a specific mass removal might be problematic... but more often than not, mass removals are OK. In fact, usually such removals can be beneficial. The removal will provide motivation for other editors to try to find sources that will allow a return of the deleted material (with a citation), thus improving the article. If it turns out that the material is (in fact) verifiable, the removal will be only temporary. If it turns out that the material is not verifiable, it was probably OR and should have been removed long before. Blueboar (talk) 14:04, 26 January 2013 (UTC)

Thanks. Would you also like to comment on the complete question, which includes the part, "with edit summaries that only say 'delete per tag' ". --Bob K31416 (talk) 14:11, 26 January 2013 (UTC)

I don't have a problem with it. Sure, a more extensive edit summary might have been more informative and helpful... but not everyone is good at writing informative and helpful edit summaries. "delete per tag" at least informs other editors that there was a tag, and gives them some idea as to why the material was removed. And given that we are not actually required to leave any edit summary, I find a less than informative edit summary better than no edit summary at all. If/When an edit summary is not particularly informative or helpful, other editors can always ask for clarification. Blueboar (talk) 16:27, 26 January 2013 (UTC)

Comment: This whole thread started with a case where asking for clarification did not lead to any clarification other than 'there was a tag'.

About what's beneficial/detrimental for Wikipedia: the challenge is to motivate editors to walk a fine line avoiding different detrimental effects:

1. Any questionable content undermines the credibility of Wikipedia. Unsourced material is to some extent questionable. This effect is somewhat reduced for material that is tagged as unsourced, since readers are made aware that the material is not yet verified.
2. Depending of the style and attitude of its performer, a deletion campaign, even if fully backed up by policy, can drive away valuable editors, which does less immediate, but sustained damage.

I'm sorry I don't have a good suggestion for defining this line, but I think this policy should try to minimize both problems.— HHHIPPO 21:44, 26 January 2013 (UTC)

And what about the editors who are, in good faith, simply trying to clean up article by removing long standing unsourced information? We should be concerned about driving away these valuable editors. Blueboar (talk) 02:29, 27 January 2013 (UTC)

I agree. We would like to have both types of editors, and we would like them to work together. They should both feel welcome and they should both be aware of the concerns of the other. That's why I think that any extreme inclusionist or deletionist strategy is not helpful here, we need to find a reasonable middle ground. Explicitly mentioning the different opinions in the footnote is probably a good idea.— HHHIPPO 10:09, 27 January 2013 (UTC)

The removal is a bad idea. But that does not mean making a policy against it which would be problematic. Removals should be one by one and where for each there is a question or concern about the material (beyond just lack of a cite.) I don't see bot in the question, but we need to cut back by about 10% on what bots do, not expand. Having human editors having to spend the limited and valuable time dealing with bot errors and misfires is an absolute travesty. North8000 (talk)

The bot wasn't a suggestion, but only used for discussing the issue. It seems clear to me that no one discussing this subject, including myself, wants a bot for deleting unsourced material tagged for a long time. --Bob K31416 (talk) 14:35, 26 January 2013 (UTC)

I think that part of the reason that it's hard to generate a simple summary of opinions is that the question deals with a complex issue. A lot of it has to do with the difference between "disruptive" and "bad editing style". When I say "disruptive", I mean it in the Wikipedia usage of the word, meaning that it's likely to go to dispute resolution and result in sanctions against the disruptor (as opposed to the general dictionary meaning of merely disrupting something). We have lots of editors whose style of editing is less than ideal, but we don't block or ban them. An edit summary of "delete per tag" would be a lousy reason to block somebody. But a large number of seemingly bot-like edit summaries of "delete per tag" would be an excellent reason to communicate to that editor that they would do better to provide individual edit summaries for each edit, in which they communicate specifically why they think the material fails WP:V. --Tryptofish (talk) 21:12, 26 January 2013 (UTC)

There's something else that needs to be made clear. Although some editors see the so-called bot-like pattern of deleting as entirely a good thing, many other editors find it annoying. Of those who find it annoying are doubtless some who are guilty of WP:OWN, but there are also a great many more who are reasonable, thoughtful, good-faith editors who are simply concerned that encyclopedic material is being lost for no good reason, only a bot-like, rigid interpretation of WP:BURDEN. --Tryptofish (talk) 21:30, 26 January 2013 (UTC)

That it can be annoying I fully agree with. However, there is no Wikipedia:Do Not Do Annoying Things policy. Nor should there be. Blueboar (talk) 23:23, 26 January 2013 (UTC)

Which goes directly to the distinction that I drew between "disruptive" and "bad editing style". Just because we shouldn't block editors for being annoying, doesn't mean that we should encourage annoying things, nor treat them as something about which one may not comment. There is also no Wikipedia:Never provide other editors with guidance on how to be better editors policy. --Tryptofish (talk) 23:30, 26 January 2013 (UTC)

See above. Annoying editing can be detrimental to the project as a whole. While there's many good reasons for not generally forbidding annoying behavior, the overall goal should be to minimize the sum of all detrimental effects, and in that respect annoyance is not completely irrelevant.— HHHIPPO 23:50, 26 January 2013 (UTC)

Wikipedia:Requests for comment/Epeefleche

OK, it seems to me that there's a mix of people thinking one way or the other. If anything, I'm on the extreme end of the continuum of the posters above. Therefore, I'm not so concerned that posting here might be viewed as canvassing. Any perspectives from any parties who have commented above (or otherwise), would be welcomed, whether they support my position or not. ˜danjel [ talk | contribs ] 08:44, 27 January 2013 (UTC)

Only non-English sources in an article

WP:NONENG says that English-language sources are preferred, but does not indicate whether there's any issues with an article having only non-English sources. Could there be an argument that if an article only has (reliable, significant) sources in a foreign language, that it has a notability concern from the point of view of English Wikipedia, possibly more so for non-Latin alphabet sources? I did look at the thread above but it seemed to be dealing more with the details of translation. Eldumpo (talk) 23:19, 2 February 2013 (UTC)

Not really, we are an internal encyclopedia in English and there many notable subjects for which no English sources might exist. Note that notability here is international as well, i.e. it is not a notability in the English speaking world. In short articles with no English sources are possible and lack of English sources doesn't necessarily mean a lack of notability.--Kmhkmh (talk) 01:11, 3 February 2013 (UTC)

If a topic does not have any English language sources it suggests that the English-speaking world has not deemed the subject to be important enough to discuss, so why should it subsequently be allowed into the English Wiki? I could understand it if there was only one Wikipedia version, which needed to include articles with a wider appeal, but there is a large number of foreign-language Wikis so a topic notable in a foreign language should always have somewhere to go. Eldumpo (talk) 08:51, 3 February 2013 (UTC)

Because as I said WP is an international encyclopedia in English and not (just) an encyclopedia for the English speaking world/countries. One difference between the other language wikis and en.wp is an that English is a lingua franca of sorts, hence we have readers and editors, who do not originate from the English speaking world. Independent of that at least some other language Wikis (I can speak for all, but the German one does for instance) take this international approach to notability as well and do of course cover topics for which only foreign languages sources (here English in particular) exist.--Kmhkmh (talk) 15:24, 3 February 2013 (UTC)

That's true, there is always somewhere else for the article to go, but the purpose of having many languages is meant to be a service to our readers. It would be a disservice to our readers who want to learn about deeply and specifically foreign topics to force them to learn the relevant language. Someguy1221 (talk) 09:01, 3 February 2013 (UTC)

Absolutely not. The English language Wikipedia has an automatic systemic bias towards the Anglosphere, but that is considered a bad thing that needs to be fought, and definitely not something that needs to be enforced.

That does not mean that an article is automatically notable for the English language Wikipedia just because it is notable for the Wikipedia in another language. Here are two ways in which this may not be the case:

1. Other notability rules. For example, the German language Wikipedia is very well developed and has its own notability rules, sometimes very different from ours. For example, an organisation is notable or not based on the number of members. An organisation may not be notable there even after the New York Times and the biggest German newspapers have published several in-depth articles about it. On the other hand, even an organisation with a million members may not be notable here if only a few news articles mention it in passing.
2. Strong systemic bias of the other Wikipedia. If the language has very few speakers, they will have a very strong systemic bias towards topics of purely local interest for the area where the language is spoken. E.g. they may want to include biographies on all mayors in all cities and villages where the language is spoken. We obviously don't do that because we don't do it for places in the Anglosphere, either. (For most there will be absolutely no sources. For a small language community this is not a problem because everybody will know someone who lives there and can simply ask about that person.)

We just apply our notability rules and judge the available sources exactly the same way as if they were written in English. E.g. a non-English newspaper is judged according to its journalistic reputation and its number of readers. If it's the leading, or even the only, newspaper in that language, then that adds a bit to the journalistic reputation but is otherwise irrelevant. Hans Adler 09:17, 3 February 2013 (UTC)

I agree with Hans and this is a long standing consensus. If there are notability issues, there is a notability policy.--Andrew Lancaster (talk) 11:21, 3 February 2013 (UTC)

This question has nothing to do with verifiability and doesn't belong on this page, but presumably at Wikipedia talk:Notability. --Hegvald (talk) 21:40, 3 February 2013 (UTC)

Self-published or questionable sources as sources on themselves — universities

Re WP:ABOUTSELF item 5 — What does "primarily" mean? Is there anywhere on Wikipedia where one can get guidance for the case of universities? Thanks. --Bob K31416 (talk) 16:35, 7 February 2013 (UTC)

Institutional publications are often an RS source about themselves and often other things. Strictly speaking aboutself mostly is aimed at individuals about individuals themselves (because of an evident lack of oversight), and not individuals about institutions (published by institutions); museums and foundations and governments (and corporations) also fall in this category. But, as with anything, there is nuance there so the particular case (eg. too much puffery) would go to RS Noticebord or NPOV Noticeboard for individualized assessment in cases of disagreement. Alanscottwalker (talk) 16:46, 7 February 2013 (UTC)

That being the case re individuals, would the case of an institution's website, for example a university, be more credible than an individual's website and should something along those lines be mentioned in WP:ABOUTSELF or elsewhere in WP:V? --Bob K31416 (talk) 17:29, 7 February 2013 (UTC)

Don't know. I've only been involved in one "dispute" where an editor said a museum website was not a reliable source for facts and opinions because the museum published them, but no one else agreed with that editor. Alanscottwalker (talk) 17:35, 7 February 2013 (UTC)

Actually, I'm presently involved in a difference of opinion regarding an article that was tagged with {{third-party}}, a tag which essentially says for that case of an article about a university, that there are too many citations to the university's website and that prevents the article about the university from being verifiable and neutral. There is a consensus of 3-1 in favor of the tag. --Bob K31416 (talk) 17:47, 7 February 2013 (UTC)

Ask which particular statements are in dispute. Alanscottwalker (talk) 17:55, 7 February 2013 (UTC)

That's what I did and the following response seemed to sum up the opinion over there, "I'm skeptical of those claims. Not because I have any reason to doubt them, but because the only reason I have to believe them is what is told to me by people who make money selling the university." --Bob K31416 (talk) 18:12, 7 February 2013 (UTC)

[ I notified the editors over there of this discussion.[10] --Bob K31416 (talk) 18:25, 7 February 2013 (UTC) ]

So, remove "those claims" or take "those claims" to RS/N or DR/N Alanscottwalker (talk) 18:53, 7 February 2013 (UTC)

Maybe you should look at the article: Al-Ahliyya Amman University. You can judge for yourself if there is a problem with verifiability and neutrality as the tag says. Also, note that the article seems to comply with the guideline WP:College and university guidelines, and the relevant section in it Reliable sources, for example with the majority (10-7) of the citations to sources other than the university website. Regards, --Bob K31416 (talk) 00:51, 8 February 2013 (UTC)

I don't have an opinion on that article, but those various editors do. If you have a content dispute like that you don't have much choice but to accept that for now, or seek input at relevant content Projects, or the DR notice boards, if it is important enough to do so. Alanscottwalker (talk) 01:41, 8 February 2013 (UTC)

Actually, I wasn't asking you for an opinion on the article, but to judge for yourself, for your own edification, which you can still do quietly. Thanks. --Bob K31416 (talk) 01:53, 8 February 2013 (UTC)

• University Portal: [11]
• Notability/universities: [12]

(olive (talk) 02:48, 8 February 2013 (UTC))

Thanks. --Bob K31416 (talk) 18:38, 7 February 2013 (UTC)

Protection

I'm not sure I understand what the protection level is on this article. It seems to be semi protected but what is the "block all non admins" protection mean and is that still in effect. Given the amount of time non admins have spent on this article, I can't imagine a protection that would now block them, but I may not be understanding this at all.(olive (talk) 17:26, 7 February 2013 (UTC))

• The current level of protection is WP:SEMI. The full protection has expired.—S Marshall T/C 17:54, 7 February 2013 (UTC)

Thanks for the reply. I thought that was the case. I don't think I've ever seen protection that applied to editors but excluded admins. Uninvolved admins yes, but admins in general, no. (olive (talk) 20:07, 7 February 2013 (UTC))

phrasing of the non-English sources rules

At a featured article nomination, I brought up the issue of WP:NONENG. However, it may have been based on a bad premise.

It seems to have been caused by this edit:

• Revision as of 14:16, 22 June 2011 Blueboar (→‎Non-English sources: I think this is what was intended)

...which in turn seems to have come as a result of:

• Wikipedia talk:Verifiability/Archive 46#Should WP:V say that non-English text and translation should be provided in a footnote?
• Wikipedia talk:Verifiability/Archive 49#Non-English sources for new material

The text moved from requiring both original and translation, to requiring... something. I thought we still require the original by default, and sometimes also the translation. Others disagree. Which is it, then? --Joy [shallot] (talk) 23:14, 25 January 2013 (UTC)

• I'm pleased to say that multilingual editors are no longer asked to type out long passages from foreign-language sources for the benefit of monoglots. You should never have had to type out your sources in order to be able to use them, and copyright concerns are only part of the reason for that. You merely have to cite your sources clearly and accurately. Editors concerned about the verifiability of a foreign-language passage are advised to contact an editor who speaks the language and can verify it for them. However, equally it is good practice and evidence of good faith, where you're asked to respond to verifiability concerns, to type out passages from your source which are short enough to be fair use but long enough to give the whole fact being verified in context.—S Marshall T/C 23:35, 25 January 2013 (UTC)
  • That's my understanding too, Joy. - Dank (push to talk) 23:49, 25 January 2013 (UTC)

So why is this phrased in such a roundabout fashion? Why can't we spell it out?

Also, is the translation what is meant by a courtesy? Or is the relevant portion of the original also just a courtesy? I'd go with the former. For example, in a long text, particularly one without page numbers, we want to see which exact part of the original is meant, but translation can be optional.

"When citing a non-English source for information, you do not have to provide the original text or a translation, unless a question should arise as to whether the non-English original actually supports the information. In that case, relevant portions of the original (short enough for fair use) should be given in a footnote, and a translation should be provided as a courtesy."

--Joy [shallot] (talk) 10:23, 26 January 2013 (UTC)

FYI.... I was actually the editor who first added the old requirement to provide the original text and a translation. The issue we were dealing with at the time was "flawed translations by Wikipedians" ... the idea behind providing the original (non-English) text in the footnote was to allow other editors to check that the original had been translated correctly. However, since I added that instruction, I have come to the realization that this is only necessary when someone actually questions the translation. There is certainly nothing wrong with providing a footnote with the original non-English text as a courtesy... the change is an acknowledgement that it is not always necessary to do so. Blueboar (talk) 17:06, 26 January 2013 (UTC)

So how about this to replace the two list items (why the list anyway?):

A quotation from a source in a different language needs to include the original text and an English translation, either in the body of the article or in a footnote.

If a question should arise as to whether a non-English source actually supports a reference made to it, such a citation should include the relevant portions of the original and a translation, in a footnote.

<ref name=Courtesy /> seems to be from another context and it largely duplicates the above text. I'm not sure it's useful here. --Joy [shallot] (talk) 12:18, 27 January 2013 (UTC)

• Absolutely not. We won't make editors type out passages from their sources before they use them, whether or not they're in English.—S Marshall T/C 13:34, 27 January 2013 (UTC)

• The bullet list is to emphasize the the two different uses of sources: (1) for a direct quotation, (2) for supporting the meaning (but not the phrasing) of a statement. The two have (and should have) different requirements on how much of the original should be provided. I agree that the footnote has large overlap with the main text though, and could be merged in there. — HHHIPPO 14:42, 27 January 2013 (UTC)

Okay, so my attempt to clarify was instantly reverted. S Marshall, what exactly is the problem with removing the confusing "is not always necessary to provide a translation" clause? I actually agree with the consensus reading of the policy. I had my doubts previously because I see some merit in doing it differently, but if everyone is happy to let me do less work when referencing non-English sources, I've no reason not to be happy with that, too. So the apparent consensus is that it is not necessary more often than not. Given the talk of courtesy, one could even argue that it's never necessary. So why should we keep the implication that it is ever always necessary in the policy text? What is the purpose of such a convoluted statement, one that may lead others to misinterpret the intent? --Joy [shallot] (talk) 23:40, 6 February 2013 (UTC)

• The concern I have is removing "it is not always necessary to provide a translation" without replacing it with something clearer. I'd like to keep something in the policy that specifically says I shouldn't normally have to rush around translating my sources into English on demand for the benefit of suspicious monoglots. (What I really want it to say is, if you're concerned about a source but you don't speak the language, go and ask an editor who does. That's a more substantive change than we have contemplated.)—S Marshall T/C 23:53, 6 February 2013 (UTC)

So how about the wording I proposed earlier, you do not have to provide the original text or a translation, unless ...? --Joy [shallot] (talk) 07:32, 7 February 2013 (UTC)

I'm against making a stronger requirement there, which is what I believe that your reverted change did. Sincerely, North8000 (talk) 12:39, 7 February 2013 (UTC)

In fairness, North, I think Joy does have a point: it's not clear. Maybe we can rephrase it completely?

If you want to check a source in a language you do not speak fluently, ask an editor who does speak the language for help. A relevant wikiproject may be able to help you locate one. If you do speak the language, consider whether a brief fair-use passage from the source text can be quoted, and an English translation provided, in a footnote.

I'm concerned that WP:V places a heavy burden on editors who speak certain languages and I don't want them to have to spend long periods of time dealing with sourcing disputes. By its very nature, WP:V creates a tyranny of ignorance. What I mean is, for example, WP:BURDEN means that those who don't understand something are entitled to demand large amounts of work from those who do. That's unavoidable, a necessary consequence of having a verifiable encyclopaedia and much better than the alternative, but we need to recognise that it also means editors who have rare and valuable expertise on the English Wikipedia, speaking an unusual language, can be asked to spend their volunteering time dealing with wrangles between monoglot American teenagers instead of translating important articles. We need to make sure the burden on those people is sensible and proportional.—S Marshall T/C 12:44, 7 February 2013 (UTC)

My main thought about making the requirement stronger revolves around the the person who uses wp:ver to selectively knock out material to pursue a POV or pissing war. I think that any change that clarifies without increasing that problem would be good. North8000 (talk) 13:06, 7 February 2013 (UTC)

I totally support this new formulation. It's what the policy should have said all along because that's how it is applied. Hans Adler 15:21, 7 February 2013 (UTC)

The fact that you thought I was making the requirement stricter than it is, is a testament to the confusing nature of this text - I was thinking that I was making it more lax!

Another attempt at a minimal change to the text - replace the 2nd list item (When citing a non-English source [...]) with a normal paragraph saying:

When addressing accuracy disputes involving citations to non-English sources, a translation of the relevant portions of the original may be provided in a footnote, as a courtesy.<ref name="Courtesy"/>

That way, we completely drop any inference that it would ever be strictly necessary to copy and translate by default when citing (not quoting), and even in case of accuracy disputes, it's still a may (not should nor must). --Joy [shallot] (talk) 11:16, 8 February 2013 (UTC)

• I have no objection to that.—S Marshall T/C 01:37, 9 February 2013 (UTC)

Where is guideline on personal knowledge?

I encountered an editor who wanted to exclude a properly sourced sentence from an article because he believed the sentence to be false (based on his personal knowledge). It all worked out in the end; but I wanted to point the editor to a WP policy or guideline that says "An editor cannot exclude material (supported by a reliable source) from an article because their own personal knowledge or experience suggests that the material is false". I'm sure I've seen such a guideline in the past, but I cannot find it. Does anyone know where such a guideline is? WP:V says:

[WP's] content is determined by previously published information rather than the beliefs or experiences of its editors. Even if you're sure something is true, it must be verifiable before you can add it.

but that is the reverse of the situation I encountered. --Noleander (talk) 11:46, 8 February 2013 (UTC)

What you are basically talking about is a policy that mandates inclusion of a particular piece of material, or can be used to unilaterally force inclusion of a particular piece of material. Thankfully and rightly it does not exist. Verifiability is a requirement for inclusion, not a mandate for inclusion. Per wp:ver, personal knowledge carries zero weight for inclusion. And per common practice, personal knowledge carries little weight for exclusion, more and better sources carry the day. North8000 (talk) 11:53, 8 February 2013 (UTC)

Perhaps I did not explain myself properly. The article in question had a sentence "Object A is B". The sentence was sourced to a reliable source. Editor E, on the talk page said (I'm paraphrasing) "I know that A is not B, so I want to remove that sentence". E had no sources backing up his belief. I wanted to tell E that he could not remove the sentence, because the sentence had a decent source, and that E's own personal knowledge was not sufficient to override the source. Are you saying E was correct to remove the sentence? Or, if E would be wrong to remove the sentence, what WP guideline would you point to to justify that position? --Noleander (talk) 12:00, 8 February 2013 (UTC)

(edit conflict)North: you say "And per common practice, personal knowledge carries little weight for exclusion ..." - that is what I'm asking about: isnt that written down anywhere? Even in a footnote or essay? --Noleander (talk) 12:02, 8 February 2013 (UTC)

I'm saying that (depending on the particulars) E will probably lose that argument because they have no policy basis supporting them, and only their personal opinion to try to influence the discussion. But to take a slightly different example, if an unreliable wp:rs makes what editors agree is clearly an error, there is nothing in police that pe se can allow one editor to force inclusion. North8000 (talk) 12:11, 8 February 2013 (UTC)

Yes, your "different example" makes sense, and the RS policy covers that. But back to the situation I encountered. Is there any objection to adding something like

'"An editor cannot exclude material (supported by a reliable source) from an article solely because their own personal knowledge or experience suggests that the material is false"'

to some WP guideline or essay? This must come up somewhat frequently, and it would nice to have it in writing somewhere. --Noleander (talk) 12:18, 8 February 2013 (UTC)

(edit conflict)That wouldn't work. Although the editor's experience is not authoritative, it is part of the process of deciding whether material should be included. That policy change allows the situation where an editor questions the material because of personal experience, another editor believes the editor's experience and agrees that the material is questionable, but quotes the policy that you are suggesting and says that they have to put it in because of policy. It should be left to the consensus of the involved editors to determine whether or not the material should be included based on its merits. --Bob K31416 (talk) 13:01, 8 February 2013 (UTC)

• I wouldn't want that to be policy, personally, at least not in that form. Things that Wikipedia defines as "reliable sources" can and do make mistakes. I wouldn't want WP:V to compel us to preserve every error ever made by a hurried newspaper journalist or every typo that slipped past the copy-editor. It's right that in some circumstances, Wikipedians can form a consensus to exclude a source because it's unreliable, and it's also right that an error in the source affects its reliability.—S Marshall T/C 12:52, 8 February 2013 (UTC)

I think that your proposal covers a case so specialized (a single editor excluding material solely based on personal opinion) that it is both a red herring and too specialized. And at the same time a camel's nose under the tent to try to unilaterally force in material. North8000 (talk) 13:08, 8 February 2013 (UTC)

Hmmm... we do have a policy against adding information based on personal knowledge... it's called WP:No original research. I think it can apply to removing information as well. Blueboar (talk) 13:28, 8 February 2013 (UTC)

Finally, a constructive response :-) yes, I considered referring the editor to OR, but OR describes the "cannot include" situation; OR unfortunately does not mention the opposite "cannot exclude" situation. This talk page proposal is asking if WP:OR (or any other guideline) should be expanded to include that "dont exclude" guideline. --Noleander (talk) 14:52, 8 February 2013 (UTC)

I have a big objection to the proposed addition. It says nothing about how well-sourced the information is. If someone puts in some nonsense such as "AD means After Death" in a calendar context, and provides no reliable source, I'm just going to revert because I know it's false. I'm not going to bother providing a source, although I could. Even if it's something I know but would have trouble finding a source for, I'm still going to revert if the editor who inserted the claim doesn't have a reliable source. I also agree with the objections others have raised. Jc3s5h (talk) 14:27, 8 February 2013 (UTC)

I think you have mis-understood the proposal. Your example has no source for the material, but the proposal discusses a situation where the material is supported by a reliable source. --Noleander (talk) 14:49, 8 February 2013 (UTC)

Perhaps the scenario in question relates to the "verifiability, not truth" change that was recently made to WP:V (btw, I was a proponent of the removal of that catchphrase). This scenario described at top could have been remedied by pointing the editor to the essay Wikipedia:Verifiability, not truth, which states "editors may not ... remove sources' views from articles simply because they disagree with them." ... that is precisely the dictum I wanted to show to the offending editor. Just out of curiosity, was that dictum removed from WP:V during the recent overhaul and moved into that essay? --Noleander (talk) 14:49, 8 February 2013 (UTC)

I looked in an older version of WP:V, and I did not find the line "editors may not ... remove sources' views from articles simply because they disagree with them." So, apparently that proclamation was in the Wikipedia:Verifiability, not truth essay, but never in WP:V. In any case, I'm content now; that essay contains the guidance that is appropriate for this scenario, so there is no need to add to some other guideline. Thanks to all for the input. --Noleander (talk) 14:59, 8 February 2013 (UTC)

That may not work because it is referring to "view", i.e. opinion etc, not an alleged fact. It depends on the specifics of your case, i.e. whether it is a view or an alleged fact. --Bob K31416 (talk) 15:13, 8 February 2013 (UTC)

I read that essay as "in WP every assertion is treated equally: WP doesnt care if it is a fact or true or false or a view or an opinion. The only thing that matter is whether a reliable source makes the statement". Are you saying if an article has a fact F, supported by a reliable source, and an editor believes that fact to be false, the editor can remove the fact from the article based solely on their belief? That seems contrary to the V and OR policies. --Noleander (talk) 15:26, 8 February 2013 (UTC)

I'm just giving you my interpretation of the excerpt you presented from the essay. Perhaps it would help clarify my last message if I gave two hypothetical examples, one where the excerpt doesn't apply, and one where it does.

For example, suppose an article says, "There is a signal at the corner of Elm St. and Maple Ave.[1]" An editor says he lives on that corner and there is no signal. This is the case of an alleged fact and that excerpt from the essay doesn't apply.

But if the article says, "There should be a signal at the corner of ElmSt. and Maple Ave.[1]", and the editor says that there shouldn't be, then this is the case of a view and the excerpt from the essay applies.

Which example corresponds to your case? --Bob K31416 (talk) 15:34, 8 February 2013 (UTC)

Challenging sourced material takes effort. You should not simply remove sourced material because you thinks it is "incorrect". You need to go to the talk page and explain why you think it should be removed, and gain a consensus to remove it (ie you need to convince other editors to remove it).

Now, we are allowed to mention personal knowledge/experience in a talk page discussion. On the talk page, you can say "I know from my personal experience that XYZ is wrong"... and others can examine whether that personal experience is enough to merit removal, or not. If your explanation convinces other editors, then the group can reach a consensus to remove. The key is that removing sourced information is a group (consensus) decision, and not something that should be done by any single editor acting alone. Blueboar (talk) 15:43, 8 February 2013 (UTC)

That seems reasonable, except possibly in the case of WP:BLP. --Bob K31416 (talk) 15:49, 8 February 2013 (UTC)

I agree with Blueboar's last summary (for the (common) case that it covers). And it is the current reality. North8000 (talk) 15:59, 8 February 2013 (UTC)

It takes more than a discussion on the talk page. Consider the "traffic signal" example given by BobK31416: If an article contains a statement (backed by a reliable source) that says there is a traffic signal; and an editor on the talk page says "I live near there, and there is not a signal .. so Im gonna remove that sentence". The other editors should reply with: "You cannot do that: you need to provide some proof, such as a photo, or diagram from the city planning department, or a newspaper article that says the signal was removed". We don't take editor's unsupported personal experiences to trump published sources. If we do, that is news to me. --Noleander (talk) 16:13, 8 February 2013 (UTC)

One has to do a logical dissection on these; other wise we are conflating multiple questions. Certainly, in a discussion, that editor has no basis to prevail over the others and remove it. But the group of editors may decide to remove it based on such input. North8000 (talk) 16:25, 8 February 2013 (UTC)

I agree. But what are the WP policies involved? The "an editor's unsupported assertion doesnt trump a RS" rule is sort of hidden within WP:V and WP:OR (and more explicitly stated in the Wikipedia:Verifiability, not truth essay); and the "group of editors on talk page can decide that an editor's knowledge trumps a source" rule is WP:IAR, specifically Wikipedia:Common sense. Anyway, like I said, I'm happy: the VerifNotTruth essay contains what I was looking for. --Noleander (talk) 17:08, 8 February 2013 (UTC)

This is inherent in the process. There are trillions of things to do and don't happen, we don't define all of those in policy.North8000 (talk) 17:14, 8 February 2013 (UTC)

It's an interesting question. My first instinct was to think of WP:NOR, as discussed above, but I guess it gets complicated in that we don't want to include material based on OR, while the question here is about exclusion, and we do allow for editorial judgments in discussions about what is or isn't "encyclopedic". I think the question makes an assumption that in this special case the challenging editor will never be able to find a reliable source that says the material should be removed. That's unusual. Most of the time, when someone has good-faith personal knowledge that one source is incorrect, there's a second source (or more) to back that up. Then, editors can reasonably discuss the relative reliability of the various sources, and that's that. Here, there's one source that says something is true, no source to contradict it, and an editor who doubts the source. I'd say that editors should evaluate whether the single source appears reliable for information other than the information in question, and, if it is otherwise reliable, whether there is a clear reason other than the one editor's statement to call into question the reliability of the source for this information. If there seem to be reasons to doubt the source's reliability, then editors might agree to AGF the editors hunch, on the grounds that there is only a single source, the source has found to be unreliable, and it's better the leave the information out if it cannot be reliably sourced. But if the only reason to question the source's reliability is what one editor says, then it's one editor versus a reliable source. And the editor loses, and the reliable source wins. So I guess my answer to Noleander's question is to (1) ask whether the editor can possibly find other sourcing, anything at all, and (2) if not, to evaluate the existing source with respect to WP:RS. The challenging editor must either provide a more reliable source, or demonstrate unreliability of the existing source. --Tryptofish (talk) 17:24, 8 February 2013 (UTC)

For those that are interested: the situation that gave rise to this policy question was in the article Ipse dixit, which is a kind of logical/rhetorical statement. Editor A wanted to add an example of ipse dixit into the article, a quote from Lewis Carroll. Editor A had a source S that said Carroll's quote was an example of ipse dixit. Editor B said "that quote is not an example of ipse dixit, and source S is wrong", so B removed the material from the article. The question was: could B remove the material from the article? I maintained that B needed to find a source that said S was wrong, because we could not rely on B's personal knowledge to trump S. --Noleander (talk) 17:38, 8 February 2013 (UTC)

I agree with you, and I'd argue that this analysis comes back to WP:RS. --Tryptofish (talk) 17:42, 8 February 2013 (UTC)

Again... One single editor should not say "I think this source is wrong" and remove the material... but if multiple editors agree that "yeah... this source did indeed get it wrong" (especially if no one disagrees with that assessment) then the group can remove the material. If you need a policy/guideline rational... consensus can determine that the cited source is unreliable on this one specific fact. We do this with questionable news reports all the time - saying that while the news organization in question is usually reliable, in X specific report they got their facts wrong. Blueboar (talk) 18:05, 8 February 2013 (UTC)

Yes, but my question was not how multiple editors can agree that the source is unreliable. My question was on the first part of your statement: where is the WP policy that says "One single editor should not say "I think this source is wrong" and remove the material." That rule is not explicity stated in WP:V or in WP:OR. But it is - more or less - stated in the VerifNotTruth essay. --Noleander (talk) 18:12, 8 February 2013 (UTC)

Maybe WP:CONSENSUS? After all, if the removing editor had removed it and (like the proverbial tree falling in the woods) no one else cared, or noticed, then it might be hard to say that the removal was really wrong. Here, they removed it, and you (and maybe others, maybe not) objected to the removal. --Tryptofish (talk) 19:17, 8 February 2013 (UTC)

Yeah, that is kind of what happened. Editor A inserted it; editor B removed it; a request was put on WP:3O asking for help, and that is where I got involved. It all worked out in the end. --Noleander (talk) 19:25, 8 February 2013 (UTC)

Noleander, Re " I maintained that B needed to find a source that said S was wrong, because we could not rely on B's personal knowledge to trump S." — From how you described the case, it looks even worse than "personal knowledge", viz. it was only the editor's personal opinion, i.e. view. Thus the excerpt that you previously quoted from the essay applies. --Bob K31416 (talk) 20:49, 8 February 2013 (UTC)

I was going to say the same thing. It was personal opinion not personal knowledge. Not only that, but the question itself was a matter of opinion (rather than objective fact).North8000 (talk) 21:07, 8 February 2013 (UTC)

WP policies do not make the distinction between fact & opinion that is being suggested by Bob & North. One person's opinion is another person's fact. There is no WP policy that distinguishes between the two. The policies only ask: Does a reliable source make the assertion? Whether the assertion is a fact or opinion is not a consideration. --Noleander (talk) 22:18, 8 February 2013 (UTC)

Well... we do distinguish between fact and opinion (even if not in the way Bob and North are talking about). For example, take the statement "Obama is a socialist". There are certainly sources that can be used to support this statement, so it could be presented as a cited statement of fact. However, doing so is almost guaranteed to be challenged. On the other hand, if we phrase the statement as an opinion ("According to X, Obama is a socialist"... "Y has said that Obama is a socialist" etc.) and cite that opinion, and it is much more acceptable and much less likely to be challenged.

So, let's take a situation where someone does include the cited statement "Obama is a socialist" in an article... and let us assume someone else objects to that statement and wants to remove it. This is where WP:NPOV and WP:PRESERVE come in... Both of these policies indicate that, instead of removing the statement outright, a better reaction is to rewrite it - turning it into a more acceptable cited statement of opinion. Blueboar (talk) 22:38, 8 February 2013 (UTC)

Sure, but that is just WP:ATTRIBUTEPOV. Identifying the source of an assertion is generally a good thing to do, but does not impact whether or not an assertion can be removed from an article. I was addressing the suggestion that an editor can remove sourced "views" but cannot remove sourced "facts". That is not consistent with WP policies. "The sky is blue" "Madam Curie discovered radium" "Columbus was the first european to step foot in the Americas" "The second amendment permits gun ownership" ... every single assertion in WP can be viewed either as fact or opinion. There is no WP policy that permits editors to remove a sourced opinion, but excepts sourced facts. Anyway, my original query was answered, and I'm taking off for the weekend ... thanks for all the mentally stimulating discussion! --Noleander (talk) 22:53, 8 February 2013 (UTC)

Noleander, Blueboar, Re "WP policies do not make the distinction between fact & opinion that is being suggested by Bob & North." — I didn't express any opinion regarding that and Wikipedia policies. My comments came out of the discussion we had regarding the excerpt Noleander quoted from the essay. --Bob K31416 (talk) 23:07, 8 February 2013 (UTC)

Someone misunderstood or mis-stated what I said. I never said that policies made a distinction between fact and opinion. I in essence said that editors are usually free to decide to leave out material for nearly any reason (after all, they "leave out" 99.999999999999% of all of the material in the world from any given article) However, one editor is unlikely to persuade them to leave out sourced material, and "unlikely" becomes wp:snow if his reason is a matter of opinion. North8000 (talk) 23:28, 8 February 2013 (UTC)

Irrelevant or Trivial information can be omitted, even if Verifiable

This has come up at VP again... I think it would be helpful to include a statement that effectively says: While information must be verifiable in order to be included in an article, this does not mean that all verifiable information must be included in an article. Verifiable information that is deemed irrelevant or trivial to the topic of the article may be omitted. Thoughts? Blueboar (talk) 22:53, 8 February 2013 (UTC)

Yes, that could be useful. Much if it is pretty well covered in Wikipedia:What Wikipedia is not. Maybe include your sentence and link to that? --Noleander (talk) 22:56, 8 February 2013 (UTC)

Yes, that is an excellent idea. A logician would say that a part of that is redundant (there is nothing that says that it must be included) but the introducing the idea of considering relevancy is touching on a very important point that is not covered in policies / guidelines. Personally I think it would be good for it to evolve to the point that degree of relevancy is one of the factors weighed in wp:undue as well. North8000 (talk) 23:01, 8 February 2013 (UTC)

• I agree that we should say this. I also agree with North8000 that it belongs in WP:NOR (ideally WP:DUE) rather than WP:V.—S Marshall T/C 01:40, 9 February 2013 (UTC)

Do you mean wp:npov? North8000 (talk) 02:04, 9 February 2013 (UTC)

I agree that what I am talking about is a compliment to UNDUE. But I think the emphasis is a bit different... UNDUE focuses on omitting opinions (because they are fringe), while my new statement would relate to omitting facts (because they are irrelevant or trivial to the topic, not necessarily because they are fringe). Blueboar (talk) 04:01, 9 February 2013 (UTC)

I'm not sure that the distinction between facts and opinions is really part of undue. At one stage it says "An article should not give undue weight to any aspects of the subject but should strive to treat each aspect with a weight appropriate to its significance to the subject." That sounds more like facts, and certainly I've argued for removal of trivial but negative information from BLPs on those grounds. Perhaps the issue isn't really here, so much as in tightening undue a bit more? - Bilby (talk) 06:04, 9 February 2013 (UTC)

This (taking degree of relevancy into consideration) addresses one of the 2 big deficiencies that makes the core of wp:undue (coverage based on amount of coverage in sources) mis-firing and ineffective. Wikipedia:Strategic issues with core policies#Adding relevancy guidance would substantially increase Wp:npov's effectiveness North8000 (talk) 14:33, 9 February 2013 (UTC)

• Yes. Do this. --Jayron32 05:45, 10 February 2013 (UTC)

OK... I don't see any strong objections (yet)... so let's move to the next step... there are two sub-issues to discuss before we add it:

1) Language - Is this acceptable language? (any suggested tweaks?) ....

• While information must be verifiable in order to be included in an article, this does not mean that all verifiable information must be included in an article. Verifiable information that is deemed irrelevant or trivial to the topic of the article may be omitted. Whether specific information is irrelevant or trivial to the topic (or not) should be determined by consensus.

2) Placement - where would be the best place to put this (under an existing header, or in a new section?) Blueboar (talk) 13:46, 10 February 2013 (UTC)

1) I'm fine with that language. (Maybe link "consensus" to WP:Consensus?) I'd suggest including a "see also" hatnote pointing to WP:UNDUE and WP:PAGEDECIDE.

2) I'd make it its own sub-section of WP:V#Other issues. --Tryptofish (talk) 21:23, 10 February 2013 (UTC)

we have precedent at WP:N with the "This is not a guarantee that a topic will necessarily be handled as a separate, stand-alone page. Editors may use their discretion to merge or group two or more related topics into a single article.". -- TRPoD aka The Red Pen of Doom 21:39, 10 February 2013 (UTC)

Yes, very true even though it is a slightly different topic. The common point is that the fact that an inclusion is allowed ( =not prohibited) by policy is not an argument to force inclusion of that material. North8000 (talk) 22:07, 10 February 2013 (UTC)

Note that the suggested changes here go into some of the specifics of how to write a good article, rather than the verifiability policy. BTW, at the beginning of this section, Blueboar wrote, "This has come up at VP again..." Could Blueboar provide a link to that discussion? --Bob K31416 (talk) 12:52, 11 February 2013 (UTC)

It came up in this discussion. Blueboar (talk) 19:16, 11 February 2013 (UTC)

OK... I have added it. Feel free to tweak as you think might be needed. Blueboar (talk) 21:20, 11 February 2013 (UTC)

Cool! North8000 (talk) 21:40, 11 February 2013 (UTC)

+1. I tweaked it as I mentioned above. --Tryptofish (talk) 22:35, 11 February 2013 (UTC)

I think the change is an improvement, but the words "irrelevant" and "trivial" are a bit too weak. I think a fact should be omitted because it isn't important enough to include in an article of the length and scope of the article under consideration. So George Washington's birthday is neither irrelevant nor trivial to the topic of the United States, but isn't important enough to include in the article "United States". Jc3s5h (talk) 22:34, 11 February 2013 (UTC)

Yes, I could support "irrelevant or trivial" --> "insufficiently important". --Tryptofish (talk) 22:38, 11 February 2013 (UTC)

Blueboar, First off, I just looked at that discussion and I don't see anywhere in it the argument that since the item is verifiable, it must be included. The editor who wanted to keep it in, essentially gave the reason that it was useful info. Those opposed said it wasn't. The argument to keep it in didn't involve verifiability. Anyhow, even though that discussion isn't relevant, we can still work on the change.

"Irrelevant" and "trivial" are just two possible reasons for excluding sourced material, so it may be better to make it clear that they are examples as in the following version. This would mean modifying the last sentence too, as in the following version.

"While information must be verifiable in order to be included in an article, this does not mean that all verifiable information must be included in an article. For example, verifiable information that is deemed irrelevant or trivial to the topic of the article may be omitted, but if there is an objection to removing verifiable information, then its removal would require a consensus."

It seems that there is more of a problem with biased editors trying to remove sourced material, than editors trying to put in irrelevant or trivial info with the justification that it is verifiable. Again, note that in the VP example that Blueboar gave, the editor wasn't using verifiability as a justification for putting in the material. In the example, it looked like editors may have been trying to keep the material out because it was revealing the secret motto of the organization. If the motto wasn't secret, it would probably have been accepted, for example Semper Fidelis. --Bob K31416 (talk) 22:55, 11 February 2013 (UTC) --Bob K31416 (talk) 22:43, 11 February 2013 (UTC)

Well, it seemed like a relevant discussion to me. In any case, there seems to be consensus to add something like this, whether the discussion on VP was directly related or not. If you object, we can discuss further. Blueboar (talk) 22:50, 11 February 2013 (UTC)

(edit conflict) I support the addition, even if the discussion that gave rise to the idea wasn't a perfect case study. What the addition gets at strikes me as a reasonable and appropriate addition to the page; the previous discussion is an "n" of 1. And I think that Bob's suggested last sentence ruins the meaning of it, by treating it as a more complex dispute than what the rest of us were thinking about. --Tryptofish (talk) 22:53, 11 February 2013 (UTC)

Here's my point again so you can address it more directly, ""Irrelevant" and "trivial" are just two possible reasons for excluding sourced material, so it may be better to make it clear that they are examples..." Not clear why you want to focus on just these two reasons. --Bob K31416 (talk) 23:04, 11 February 2013 (UTC)

I don't. Please see what I said to Jc3s5h, above, just before you reverted the section. --Tryptofish (talk) 23:08, 11 February 2013 (UTC)

I saw it. That's a matter of using "or" instead of "and", which isn't relevant to the point that I mentioned and then repeated in my last message. I can't force you to discuss this with me, but I don't think policy should be developed in that way. Well, if you don't want to address the "focus" point for some reason, here's the other point: As I mentioned, if there is an objection, there should be consensus for removal of sourced material. We wouldn't want to have people unilaterally removing reliably sourced material above an objection, just because they don't like it, and using the excuse that it is trivial and irrelevant, which is a matter of judgement. --Bob K31416 (talk) 23:18, 11 February 2013 (UTC)

I support the addition even though a logician purist would mistakenly say that it is unnecessary, and that it incorrectly implies that those are the only two reasons for leaving verifiable material out. North8000 (talk) 00:21, 12 February 2013 (UTC)

@Bob...Re: We wouldn't want to have people unilaterally removing reliably sourced material above an objection... that is why I added the line about consensus. Is that not enough? Blueboar (talk) 13:07, 12 February 2013 (UTC)

I noticed that. Here's what you have, "Whether specific information is irrelevant or trivial to the topic (or not) should be determined by consensus." It doesn't indicate who has the responsibility to get consensus, which can lead to edit wars. What I have is, "but if there is an objection to removing verifiable information, then its removal would require a consensus." --Bob K31416 (talk) 15:49, 12 February 2013 (UTC)

(ec)Yes... that's what "determined by consensus" means. We assume that an edit has consensus until someone reverts or states an objection. As soon as someone does revert or otherwise object, then everyone is supposed to go to the article talk page to discuss it. That discussion is how a consensus is formed. There is no rule on who has to start the discussion... only that the discussion should take place. Everyone working on the article has a "responsibility to get consensus"... the remover, the objector, and everyone else working on the article. Blueboar (talk) 16:58, 12 February 2013 (UTC)

If we go much farther than a hint at the concept, then I think we're going to move into a quagmire. One way to say it that avoids the quagmire is "Verifiability is a requirement for inclusion, not a force for inclusion." North8000 (talk) 16:51, 12 February 2013 (UTC)

I know what you mean, but I am not sure others will. Keep it simple... there are two concepts being presented: 1) it is OK to remove verifiable information that you think is trivial or irrelevant to the topic, and 2) any disputes are settled through discussion and consensus. There is really no need to over think this. Blueboar (talk) 16:58, 12 February 2013 (UTC)

My points are valid and my version is better, but I don't see any progress being made in their acceptance with the editors so far involved. So I don't plan to continue if there isn't any support. I'll wait and see. --Bob K31416 (talk) 22:20, 12 February 2013 (UTC)

For convenience of reading, I'm going to reproduce the proposed paragraph here, but with the change from "irrelevant or trivial" put in:

• While information must be verifiable in order to be included in an article, this does not mean that all verifiable information must be included in an article. Verifiable information that is deemed insufficiently important to the topic of the article may be omitted. Whether specific information is sufficiently important to the topic (or not) should be determined by consensus.

Bob seems to feel that I and maybe others aren't listening to him, but actually I just don't really think he is correct about the meaning of the words. If I understand correctly, the concern is that someone could feel empowered by this language to go around removing reliably sourced content, and that the burden of discussion and consensus would fall only on those editors who object to the removal. I don't see that as really being what the language says. But let me suggest this change:

• While information must be verifiable in order to be included in an article, this does not mean that all verifiable information must be included in an article. Verifiable information that is deemed insufficiently important to the topic of the article may be omitted, if there is consensus to do so.

Does that help, by more explicitly putting the consensus at the time of the removal, as opposed to after the fact?

Alternatively, I'm thinking about whether any of this really does anything more than explain what is already consensus. I think that it is just current consensus, but I'm open to persuasion that it's somehow creating something new. WP:V already makes it clear that material has to satisfy WP:V (and by extension WP:RS), and not be ruled out by WP:NOT, to be included. It also discusses how WP:V interacts with other core policies, including WP:UNDUE. I think everything we're talking about here – as being "trivial" or "insignificant" or "insufficiently important" – basically amounts to failing WP:UNDUE. (Maybe also, WP:IINFO comes into it.) Am I missing anything? If I'm not, would a sentence or two related to UNDUE, at WP:V#Neutrality, be an alternative to what we've talked about here? --Tryptofish (talk) 22:49, 12 February 2013 (UTC)

I am laying back on this not due to it's lack of importance of what this is nibbling at (which is immense) but because I know that that your current course of trying to get that deep in (specifying the criteria and process for removal) is a quagmire...you would need to describe the whole fuzzy Wikipedia system on how articles are written and debates are handled. The sound sparse core of it is "Verifiability is a requirement for inclusion, not a force for inclusion." I also supported Blueboar's addition because it hinted at the concept (which would have some influence) without getting too deep in. Sincerely, North8000 (talk) 03:08, 13 February 2013 (UTC)

Then we should restore Blueboar's addition, and someone other than me can try to accommodate Bob. --Tryptofish (talk) 20:14, 13 February 2013 (UTC)

I put it back in with a change which I think addresses Bob's valid concerns. North8000 (talk) 20:55, 13 February 2013 (UTC)

I'd suggest a grammatical tweak: change omit it to should be omitted, since the consensus doesn't omit the material, it just determines whether it is to be omitted. Otherwise, I didn't follow all the discussion but the result looks ok. — HHHIPPO 21:07, 13 February 2013 (UTC)

I wasn't totally happy with the sound myself, but the point of my choice of words was that in the example consensus decided both the determination and the removal...the first decision alone did not prescribe removal. So that made it an extra safe example. But others should feel free to change or revert.....that was just BRD to try to get somewhere Sincerely, North8000 (talk) 21:12, 13 February 2013 (UTC)

Thanks! I made a few relatively minor revisions along those lines. --Tryptofish (talk) 21:38, 13 February 2013 (UTC)

Good point, North8000. I think I found a way to avoid the problem without blowing up the sentence. — HHHIPPO 21:44, 13 February 2013 (UTC)

That's even better! --Tryptofish (talk) 21:49, 13 February 2013 (UTC)

Nice! North8000 (talk) 21:54, 13 February 2013 (UTC)

Works for me too. I have boldly added another example (feel free to revert or rewrite if needed)... to note that it can be OK to omit verifiable information from one article if it would be better presented within the context of some other article. Blueboar (talk) 13:35, 14 February 2013 (UTC)

I changed to "does not improve". Specifying "importance" or "relevance" gives ideas to editors who are looking to delete something because it is contrary to their personal beliefs or agenda. Importance is especially a poor criterion since, for example, an article about an obscure species could mention that a small part of its anatomy has a certain color, which could be considered unimportant by some, yet improve the article. --Bob K31416 (talk) 16:29, 14 February 2013 (UTC)

Verifiability policy for images offered and used to illustrate article content

I don't recall seeing a policy on or discussions about verifiability for images which ostensibly illustrate article content. Is there a policy covering this? Have there been discussions about this? I do see the question come up from time to time on article discussion pages, and I started this talk page section here after seeing Talk:State religion#Map. I mention this particular case just as a single concrete example. I have seen other examples.

The discussion linked above re this exampled case starts off with an observation that the State religion article prose says that Bhutan has a state religion and that the map at File:Map of state religions.svg which is included in the article shows Bhutan having no state religion. Without getting bogged down in the specifics there, looking at the file history for the image I see that a number of changes have been made in the image in the couple of years it has been on Wikipedia, and that no support from reliable sources has been offered either for the map image as originally uploaded or for any of the changes to it. As far as I am aware, Wikipedia has neither policy nor guidelines relating to citing supporting sources for image content or for changes thereto.

I see that that particular image is also used by the Christianity article, and that there is less than perfect agreement between what the image asserts visually there and what that article asserts textually (e.g., re El Salvador, Haiti, the Dominican Republic).

Am I missing something which should be obvious here, or is work needed in the area of Wikipedia policies and guidelines re image content verifiability. Wtmitchell (talk) (earlier Boracay Bill) 07:31, 9 February 2013 (UTC)

Images hosted on WP should by no means be used as sources for contentious statements, due to the open wiki nature (just like we don't use WP as a reference). Images that are representing factual data should be including links to said data in their file: page to assure they can be confirmed, and if it is the case that a fact is being sourced to the image, it should really be sourced to the source on the file: page (the image being an intermediate of displaying that data). --MASEM (t) 07:40, 9 February 2013 (UTC)

I don't understand that. How, specifically, does that apply to the above? How, specifically, does that apply to the image and the articles exampled above? Where, specifically, is that supported in WP policies and guidelines? Consider that images can be, and are, placed on pages by editors of individual articles who may or may not be (often are not) the editors who uploaded and/or modified those images. Also consider that images which have been used in articles may be, and are, modified by image editors independently from and without consideration of the content of the articles on which those images appear. Wtmitchell (talk) (earlier Boracay Bill) 12:35, 9 February 2013 (UTC)

I explored this once. In general straightforward descriptions of what the uploader said that the image is are accepted without sourcing to prove it, if the image is really just an image of something (e.g. not an "image" of something written by the editor). Everything else is subject to wp:ver. And an image is not a source with respect to meeting wp:ver. North8000 (talk) 14:15, 9 February 2013 (UTC)

So, corollaries of that are:

• For an image of a homeless person, a "homeless person" description is generally accepted without sourcing. A caption of "former middle-class person now homeless due to Obama's Socialist policies" would not be accepted without sourcing.North8000 (talk) 14:20, 9 February 2013 (UTC)
• Charts, content etc. made by Wikipedia editors will get removed as OR if challenged.
• Maps, charts made by sources (and uses of them and interpretations from them) are subject to the same rules and restrictions as any source.
• Captions (other than the original straightforward description of what the image is) are subject to wp:ver and all other policies for text.

This is sort of a summary of applicable policies. North8000 (talk) 14:25, 9 February 2013 (UTC)

Careful with your formulations. Quite a few editors would read "Charts made by Wikipedia editors will get removed as OR if challenged" as a free pass to 'challenge' charts without providing any valid reason. E.g. an editor who is in a conflict with me elsewhere might look at my upload history and then 'challenge' this illustration based on a general, unsubstantiated claim of original research. The editor would not require any knowledge of mathematics whatsoever. In practice that would not work, but we are in a continuous struggle against such faulty interpretations of policy. I believe they spread at least in part through abbreviated explanations that some don't understand correctly. Hans Adler 14:37, 9 February 2013 (UTC)

Images on Wikipedia should illustrate information contained in the text of the article. That text that needs to be verifiable. Blueboar (talk) 14:44, 9 February 2013 (UTC)

I was mostly describing the defacto situation. The fact is that any sky-is-blue text or wp:editor-created material can get zapped with no sincere (much less expert) challenge to it. Conversely, why should wp:editor-created material be exempt from wp:ver just because they placed it into an image? North8000 (talk) 14:48, 9 February 2013 (UTC)

Editor-created images can be included because the WP:NOR policy explicitly allows them. Of course the allowance assumes that the editor created image is based on reliable sources... sources that are cited in the text of the article. Blueboar (talk) 15:28, 9 February 2013 (UTC)

General practice is that if there are enough experts at an article (e.g. after notification of WP:WikiProject Sky Colors) and they all agree there is no problem with the text, then "I have no way of knowing the sky's colour because I am blind and not even interested in the sky, so put in sources already or I will take it out" is not a valid challenge. However, disruption can occur in such situations. See Talk:Synchronous motor#Unsourced Material for an example. However, it resulted in a lot of unnecessary concentration on sourcing uncontroversial material. Hans Adler 15:35, 9 February 2013 (UTC)

I want to insert a quick note here, re the example brought up above: the article Synchronous motor and the Talk:Synchronous motor#Unsourced Material page referred to by Hans Adler is not really a good example of what is being discussed here. In that article, there were numerous detailed technical claims that were not even close to "sky is blue" true. I am an electrical engineer, and have had the requisite motor courses to know something of what I speak. While I had nothing to do with the early requests for citations on that material, I did subsequently help work to get the article better sourced, added a few additional {{citation needed}} tags of specific claims, and worked to keep the Talk page discussion civil.

The fact that a mixture of editors in the emergent soup being cooked on that Talk page was disruptive, had nothing to do with the article being well-sourced in the first place, but more to do with editors of quite-different persuasions on sourcing and citations having it out on that particular article Talk page. That article is vastly better sourced today; however, the process to get their was not pretty. Cheers. N2e (talk) 14:50, 15 February 2013 (UTC)

Agree, but that is not limited to images. I've been an advocate of a sort of lightweight "sincerity test" when challenging material. North8000 (talk) 02:00, 10 February 2013 (UTC)

Many images are used in multiple articles. The image I used as an example for this discussion (File:Map of state religions.svg) is a map which visually makes specific assertions about particular countries. It is currently used in two articles and, as I mentioned above, contradicts portions of the content of both the articles where it is currently used. The image has a history of changes which is not keyed in any way to supporting sources cited in either of the two articles which currently use the image, or cited anywhere else. There doesn't appear to be a way for WP editors to verify what supporting sources (if any) the content of an image might be based on. There doesn't appear to be a guideline for how the uploader of an image is expected to indicate what sources support the visual message of an image.

The image description page of the exampled image has a Summary section in it. It could have a References section with a list of footnoted cites, but it does not. I don't think that I have ever seen an image description page with supporting source(s) indicated in the form of footnoted cites, though it would seem useful. Perhaps this policy should indicate that the content expressed visually by images (and expressed by other multimedia content included in articles) is subject to the same verifiability rules as article content. The Wikipedia:Images project page could then mention that, and might include a guideline re the citation of supporting sources on image description pages. WP editors adding images to articles then could browse the image description pages for supporting sources which might be cited in WP articles. Wtmitchell (talk) (earlier Boracay Bill) 02:54, 10 February 2013 (UTC)

While there is some justification for that I'm a bit wary of yet another policy. I'd rather prefer editorial judgment/discretion with common sense based on existing policies/project goals. Obviously some images need strict sourcing for the visual claims the make, but imho that can be handled on individual article basis, i.e. if a picture seems unreliable or makes questionable claims without being properly sourced simply remove it. In the bigger picture however we have large number of totally uncontroversial images that have no such sourcing (technical, mathematical, scientific drawings, most photos, etc.). If a policy requires sourcing images in general and hence those image as well and we have policy literalists removing them systematically from articles, we'll be in a lot of trouble. Such a scenario is also a recipe for driving away authors. Another thing is that strict sourcing is often not even possible, just consider pictures of the real world taken by wp or common contributors. There is no way to source that we simply have to believe that the description is accurate (and maybe perform a sanity check), but there isn't really any sourcing possible besides the first hand account of the contributor.--Kmhkmh (talk) 04:46, 10 February 2013 (UTC)

I'm not suggesting another policy. I am suggesting that Wikipedia's verifiability policy should make it explicit that Wikipedia's policy re verifiability applies to all wikipedia content which can appear in mainspace articles, not just to article prose, and that that policy should take into account the reality that association of pages outside of mainspace with particular articles can be, and often is, weak, and that such pages can be, and often are, associated with multiple articles. It does not make much sense to address verifiability issues for non-mainspace content in the References sections of mainspace articles. It makes more sense to address those issues, to the extent that they need addressing, in the non-mainspace pages themselves. Many non-mainspace pages require no support, or no more than a simple description and statement of sourcing (e.g., File:Washington Monument.JPG, currently with no articles linking to it). some such pages (e.g., File:Map of state religions.svg, the page which I exampled for this discussion, currently with two articles linking to it) provide content which should be supported by the editors uploading and editing those pages. I believe that this need goes unfilled largely because Wikipedia policy in this area is unclear or unstated and because guidelines in this area are unclear or nonexistent.

I've started this discussion with my thinking limited to images. A first step might be the addition of a section to WP:V heading something like "Images" containing something like the following:

Images in the File: namespace which are linked from mainspace articles appear in those articles as part of the rendered content. The content of these images is subject to the same verifiability policy as content appearing directly in the articles linking them. Because Images are uploaded and maintained separately from articles which might link to them, source citations supporting their content should be expressed on the image description page. Where appropriate, such source citations should be restated in articles linking to these images.

That example content is pretty lame but, hopefully, it conveys the idea. It really ought to be generalized to address all non-mainspace pages which can appear to WP readers as part of article content. Wtmitchell (talk) (earlier Boracay Bill) 11:58, 10 February 2013 (UTC)

I agree that some clarification should be added, but this is a difficult/complicated area (possibly to the point of impossibility) and I don't think that that proposal would be workable. Perhaps we should stick to a narrower area, where "images" are actually wp:editor-created statements/information. North8000 (talk) 13:47, 10 February 2013 (UTC)

I've notified editors who have updated the File:Map of state religions.svg image which I used as an example above that this discussion is going on and have invited them to join. Wtmitchell (talk) (earlier Boracay Bill) 17:48, 10 February 2013 (UTC)

• I agree with Wtmitchell that there's a potential lacuna in the rules here. I noticed it recently when someone added a graph to an article I wrote (Forestry in the United Kingdom, if anyone cares). In this case the addition was appropriate and well-sourced, but I recall wondering what I'd do if it hadn't been. I'd like to endorse what Hans Adler says: careful with your formulations. We don't want to introduce a new rule that will inconvenience good faith content-builders without evidence that there's a widespread problem.

  I don't think that our wording should concern the file namespace. It should concern the introduction of images into a specific article, because a picture that's appropriate and NPOV in one article could be a problem in a different one.—S Marshall T/C 18:16, 10 February 2013 (UTC)

A very good point. Blueboar (talk) 18:32, 10 February 2013 (UTC)

I agree that the point re appropriateness of introduction of particular images into a specific article is a point of concern. However, I believe that point is independent of the point I am trying to make here re concern about RS verifiability of assertions made visually in uploaded images. — Preceding unsigned comment added by Wtmitchell (talk • contribs) 02:22, 12 February 2013

For an example of the possible impact of explicit rules on images, see the following:

• Historical German linguistical area.PNG - a linguistic map of early 20th century German-speaking areas
• Current discussion: WP:WikiProject Poland#File:Historical German linguistical area.PNG
• Historisches deutsches Sprachgebiet.PNG - original version of the map (includes Dutch)
• Historisches deutsches Sprachgebiet.PNG - lots of discussion about the map on Commons

The sources of the original map are given on Commons. That is appropriate and, I believe, the only reasonable approach because otherwise we would have to update/correct them wherever the map is used (many languages). As you can see, there is meaningful discussion happening on Commons.

There is or was some misleading information. It appears that the map was pieced together from various pieces of information, some from 1910, some from 1937, and probably other years as well. I think overall the map is most likely to be essentially accurate for the late 1930s. But what does accurate mean? A professionally done map would have a well defined methodology such as including all areas with at least 40% native German speakers or at least 60% German speakers overall, in a specified year.

It appears that this map was created by a German nationalist. Some accusations that the map shows too much territory were misguided, at least for the 1930s. Others may be correct. The map is original research. And yet we are using it. Why? Compared to maps from reliable sources it's similar enough. Most people are not interested in the details, just in the big picture. And the map is important for the articles where it appears.

The ideal solution would be a WikiMedia project concerned with creating maps and other informational graphics. The verifiability experts there would develop methods for taking information from non-free maps without violating copyright. To the extent that this is even possible. But no such thing exists, and we have a choice between using slightly inaccurate maps that are original research and using none at all.

I think the de facto situation with graphics is this: In principle, our accuracy and verifiability standards apply unchanged. But normally they are only applied when someone finds a specific mistake and someone fixes it. Or when there is a choice between two graphics. Normally, incorrect or unverifiable details are not sufficient to remove a graphic completely. That is done only if the overall impression is very wrong.

I am not unhappy with this status quo. I think it's inevitable. What does need more care, though, is how and where we use such graphics, and in particular with what captions. Hans Adler 22:11, 10 February 2013 (UTC)

(edit conflict, responding only to Blueboar) ...yes, good point, although we must acknowledge that what is in the image file (e.g. contents of the image, or the description of it) can be the first domino is a chain that can lead to unsourced assertions being included in the article. North8000 (talk) 22:13, 10 February 2013 (UTC)

In (due to) the fuzzy/ neural net Wikipedia system, the current plan 99% works. I think that that the other 1% relates to controversial wp:editor-created content within the image. So I think that instead of messing with the 99%, we should just deal with the 1%, which is challenged wp:editor-created content/assertions contained withing the image itself. North8000 (talk) 22:21, 10 February 2013 (UTC)

I've boldly added a References section to the image description page at Commons for the image which I exampled to begin this discussion and cited a few supporting sources (not enough of them) there. This may or may not draw some participation here by editors from there. See [13] and [14]. Wtmitchell (talk) (earlier Boracay Bill) 04:52, 14 February 2013 (UTC)

Citations on Wikipedia and discussion at meta:WebCite

There is a discussion at meta:WebCite regarding citations on Wikipedia that would be of interest to those that watchlist this page. For those who don't know, webcitation.org is used to archive newspaper articles and other reliable sources that disappear from the original websites. Wikipedia currently has 182,368 links to this archive site. Regards. 64.40.54.47 (talk) 11:49, 11 February 2013 (UTC)

Note... there are other web archiving sites that can be used to cite dead links. Yes, it would be a massive task to shift all the citations that currently point to webcitation.org to one of these other sites... I am just saying that webcitation.org is not our only option. And, of course, citing dead tree, hard copy sources (if they exist) avoids the whole issue. Blueboar (talk) 15:59, 15 February 2013 (UTC)

Citations and specific figures

Currently, WP:V calls out quotations as the only piece of information requiring inline citations without regard as to whether they are likely to be challenged. From general practice, I think that "specific figures" needs to be included as well. As a rule, we seem to tolerate generalities like "Cotton is one of Georgia's major crops", but the moment someone makes a claim like "Georgia produced 2.466 million 480 pound bales of cotton in 2011", a "citation needed" tag quickly follows. It's certainly the consensus among editors dealing with song articles that specific sales figures need to be supported with a citation, as well as shipment figures. Is there any case where someone could make a specific numeric claim that would not require a citation?—Kww(talk) 22:21, 10 February 2013 (UTC)

An exception that quickly comes to mind is: "There are 50 states in the US." A "cite needed" there would be in the sky-is-blue realm. But I agree with you about the kinds of numbers you are talking about. --Tryptofish (talk) 22:28, 10 February 2013 (UTC)

That seems like a piece of text that would naturally & usually get a source when put in. Aside from that, IMHO I don't see it as a special case requiring different policy treatment (e.g giving it the same special treatment / categorical requirement as quotes.) Sincerely, North8000 (talk) 22:36, 10 February 2013 (UTC)

Does "that" mean "There are 50 states in the US" or "Georgia produced 2.466 million 480 pound bales of cotton in 2011"?—Kww(talk) 22:46, 10 February 2013 (UTC)

Well, I guess it depends on the context. I'm guessing that it would be sourced at United States. But in a subordinate context, maybe in a biography page about someone who traveled through all the states, there could be a sentence referring to the 50 states, where a cite for "50" would be silly. (A cite for the travel would be appropriate, but that would not be a cite for the number itself.) --Tryptofish (talk) 22:41, 10 February 2013 (UTC)

Unsourced sales figures for records are the usual source of problems I encounter, and I've encountered admins that didn't view the constant introduction of unsourced sales figures as disruptive behaviour because a citation wasn't required by WP:V. That was the case that got me thinking. I think I'm on pretty solid ground when I say that general consensus is that such figures do require citations, but wanted to open it up for discussion.—Kww(talk) 22:43, 10 February 2013 (UTC)

Yes, I agree with you about numbers that aren't obvious. (I think whether or not it's disruptive is a function of how "constant" it is.) Sales figures are not numbers that are common knowledge, the way that the number of US states is. Likewise (responding further to North), I think that a mention of the 5 books of the Torah or the 4 canonical Gospels do not fall in the realm of needing an inline cite for every appearance of the number. It seems like there's a distinction to be made, about being common knowledge or not. --Tryptofish (talk) 22:51, 10 February 2013 (UTC)

We may be getting a bit off topic here. Essentially, the question is does this scenario merit special policy wording that has wp:ver treat it differently that other material? North8000 (talk) 23:19, 10 February 2013 (UTC)

Actually, I think it's more "is there a general consensus on this specific area that has developed, and, if so, is it worth mentioning?" I'm not arguing for different treatment, I'm simply observing that in general practice, there's a class of figures that invariably comes under the umbrella of "likely to be challenged."—Kww(talk) 23:44, 10 February 2013 (UTC)

I'm ambivalent. I think that things like "50 states" are not likely to be challenged. On the other hand, I think that non-obvious numbers, such as sales figures or various statistical data about society or about science, are things that pretty much always need to be clearly sourced, to avoid original research. The question I have, and I think this is what North is getting at, is whether there is anything to stop editors from requesting a citation for such numbers, the same way they can request citation for a qualitative statement. Quotations are somewhat special, in that we never want to misquote a source. We never want to give faulty numbers either, but are they a different kind of content than non-numerical content? --Tryptofish (talk) 22:46, 11 February 2013 (UTC)

The issue comes with editors like User:Anthony77600, who constantly update figures without citing sources, or, worse yet, update figures while leaving sources in place that contradict the figures they are adding. I'd love to be able to point them at a crisp policy line that makes it clear to them that what they are doing is unacceptable.—Kww(talk) 17:51, 12 February 2013 (UTC)

Yes, I'm sympathetic to that. I guess one question is whether that editing pattern is really any different than when someone keeps changing other kinds of content – in either case, the changes have to meet WP:V. How can we explain, specifically, what makes the numerical changes more like a misquote than like an incorrect non-numerical fact? And how can we spell out, specifically, what kinds of numbers should or should not be held to an "always-cited" standard. --Tryptofish (talk) 22:21, 12 February 2013 (UTC)

This is a bit off-topic, but I very much disagree with using "it's obvious" as an excuse to not cite. People disagree on what is "obvious" all the time, and such disagreements are usually time consuming debates that frustrate all involved. A statement like "There are 50 states in the US" isn't that "obvious" to someone who lives in the middle of the Amazon forest... and "the sky is blue" isn't obvious to someone who spends their waking hours at night (when the sky is black)... what "exempts" such statement from citation is the fact that the statement is so easily verifiable (there are thousands of sources that can verify them) that no one is likely to challenge them. There is normally no need to include a citation.

Now... sometimes someone comes along and actually does challenge an "easily verifiable" statement. At this point we have a choice... we can either spend hours and hours arguing about whether we need to provide a citation (or not), or we can simply spend two minutes choosing one of those thousands of sources, and slap it into the article. When it comes to responding to unlikely challenges, it is much easier to let the challenger "win" the debate before it gets started (ie slap in a citation) than it is to argue about it. Blueboar (talk) 15:37, 15 February 2013 (UTC)

Actually, that's quite reasonable, and not off-topic. But it probably makes it all the more difficult to craft what was proposed here. (How about someone who goes around deleting things like "50 states"? insert interminable argument here) --Tryptofish (talk) 23:34, 15 February 2013 (UTC)

If someone goes around deleting things like "50 states"... the same choice applies. We can either engage in interminable arguments about how the deletions were (or were not) in accordance with policy, or we can cut through all the bullshit, and simply return the information with a citation. It really is that simple. Blueboar (talk) 13:12, 16 February 2013 (UTC)

Indeed, but we all know what really happens in the real world. (Sorry, I didn't intend to start the "interminable argument" to which I facetiously referred.) --Tryptofish (talk) 17:46, 16 February 2013 (UTC)

Blueboar, Re "We can either engage in interminable arguments about how the deletions were (or were not) in accordance with policy, or we can cut through all the bullshit, and simply return the information with a citation." — Would this apply to such information in 100 articles that is deleted by the editor? --Bob K31416 (talk) 15:21, 17 February 2013 (UTC)

I wonder if you're all not missing the consequences that would occur if a decision to require citations for all statistics was suddenly imposed. We have tens of thousands of articles which, at the moment, do not carry such citations. The present procedure is that, if someone wants to challenge these, they can either replace them with a cited figure, or they can tag that the statistic needs citation. It works reasonably well, and the addition of uncited figures is frequently reverted on the grounds that a citation is required (technically outside the process, but again it works reasonably well).

In the case of quotations we have different rules, and we treat them as requiring a citation at all times. There is a reason why we treat them differently, and it is because misappropriation of quotations to living people can be libellous. They are therefore treated in exactly the same way as we deal with biographies of living people. False or misapplied statistics do not carry the same danger of resulting in the encyclopedia facing legal action.

Whilst I would agree that almost all statistics should be cited, I don't want to see a scenario where we wreck large sections of the encylopedia by encouraging the obsessed to roam across our pages deleting uncited statistics at will. The end result will be chaos as articles are butchered, and other editors find their time taken up almost entirely by repairing the damage done (and finding the citations to do so). Nothing else will get done, and you have to ask yourselves, important though it is to cite statistics, is the replacement of missing citations really our first overriding priority.

It may well make sense to tighten the wording to state that all newly added statistics must be cited, and that existing uncited statistics may be tagged "citation required" even when nobody has challenged the figures. But I would not go beyond that at this stage. Skinsmoke (talk) 08:52, 21 February 2013 (UTC)

Agree with Skinsmoke except I would not not support even adding a "new ones only" idea explored in their last paragraph. North8000 (talk) 10:36, 21 February 2013 (UTC)

cite

The "burden of evidence" section says that one should consider adding a citation needed tag as an interim step. But what if a dispute is about a category or is on a disambiguation page? What is the appropriate interim step then? Pass a Method talk 15:34, 15 February 2013 (UTC)

Discussion on the talk page. Blueboar (talk) 15:38, 15 February 2013 (UTC)

I have added this sentence to deal with this grey area. Pass a Method talk 16:03, 15 February 2013 (UTC)

I moved it from the text to the footnote. I don't object to what it says, but it strikes me as such a specialized and uncommon case that it becomes WP:CREEP to say it in the main text, so I figure moving it to the footnote is a reasonable solution. --Tryptofish (talk) 23:39, 15 February 2013 (UTC)

While I can't agree with the reasoning )I don't see it as creep), I can agree with the footnote. I tend to see this as uncontroversial.--Amadscientist (talk) 02:13, 22 February 2013 (UTC)

The news hoaxes are getting bolder

http://news.yahoo.com/blogs/lookout/baby-naming-contest-hoax-5000-212020244.html Moler pitched the story to dozens of media outlets, and plenty of them—including the Daily News, Daily Mail, Huffington Post, NBC Los Angeles and Yahoo News—picked up the story...

Do we need to adjust our policies to better handle this? Hcobb (talk) 16:03, 10 March 2013 (UTC)

Nope... we handle it fine. Blueboar (talk) 22:22, 10 March 2013 (UTC)

Large-scale removals of unreferenced information

Does WP:V endorse the large-scale removal of challenged information when editors challenge all unreferenced material in the article, regardless of controversial? I ask because while it's in the letter of the policy to allow systemic gutting of articles over several years (like Minor Humans in Shannara, regardless of its notability) or to tag-bomb an article (like this), I believe it is against the spirit, and I'd like to get consensus against the practice. Thanks and regards, Ed ^{[talk] [majestic titan]} 08:40, 20 March 2013 (UTC)

Hello

IMHO, it is not neither against the letter nor the spirit of WP:V to delete all unreferenced material in one article. However, I understand that the reason people just do not do it has to do with WP:CIVIL: People spend a lot of time perfecting articles. Hurting them serves no purpose. In addition, sometimes, leaving unreferenced material for a while longer (not permanently) exposes them to other editors' scrutiny and they ultimately become referenced.

Overall, I think Wikipedians should be kinder to those who remove unreferenced contents and remember: Putting {{Citation needed}} is courtesy extended by the reviewer to the contributor; it is not one of the contributor's right or even privilege.

Best regards,

Codename Lisa (talk) 08:54, 20 March 2013 (UTC)

I generally agree with Lisa. I would argue that if an editor is challenging the removal of unsourced information, then at that point the information becomes "controversial" even if it wasn't previously, because there's now a debate regarding its suitability for inclusion. I believe tagging the article entire in question would have been better than adding numerous CN tags, but arguments can be made either way, and the tagger wasn't obligated to tag the info in any case; they could have jumped directly to removal.

Personally, I feel "best practice" when confronted with unsourced information is as follows:

If it's a BLP and the information could be considered controversial: Revert immediately. If the adding editor can be readily determined and the change was recent, notify them of the reversion with an appropriate advisory regarding BLP policy.

If the information was added recently and the adding editor can be readily determined: revert the addition, noting that it was unsourced. Note the editor's Talk page appropriately, depending on whether they're a regular and whether they've already received advisories about adding unsourced material previously.

If the information wasn't added recently/the adding editor can't be readily determined: attempt to locate and provide sources. If unable to do so, tag the information/the section/the article depending on how bad sourcing issues with the article are. Give it a reasonable amount of time (I normally give it at least two months), then either delete the information or move it to the article's Talk page depending on how much information is involved/the likelihood that it's true/how controversial the deletion seems likely to be.

Hope this helps. As noted, while different editors all have their "best practices", policy places few restrictions on the immediate deletion of unsourced material. Doniago (talk) 14:46, 20 March 2013 (UTC)

Exactly... there is a huge distinction between "you are allowed to do X" and "you should do X". Editors are allowed to delete problematic material (even large sections of material), especially when the material is in question is unsourced... HOWEVER... the fact that you are allowed to do so does not necessarily mean you should do so (That depends on the specifics of article and the situation). Nor does it address the question of how to go about it without pissing someone off. Blueboar (talk) 18:52, 20 March 2013 (UTC)

(ec)Yeah. The unfortunate reality is that there are editors who in some circumstances are protective enough of articles that they won't abide the removal of unsourced material even if every reasonable step is taken to give any interested parties an opportunity to prevent the removal. Even editors who are well-aware of the policies regarding unsourced material on occasion. I'm a bit biased with regards to this, but I'd say that the editors willing to face the fury of a mass of "we don't care whether it's unsourced, leave it in the article" editors aren't necessarily shown the level of appreciation that they deserve for trying to bring the encyclopedia up to standards.

That said, of course there are editors who wield the removal of unsourced material with an iron fist rather than a velvet glove, but it's my opinion that one should be able to reach a compromise with such editors in almost all cases. Tag it, move it to the Talk page, agree to retain the material for an additional X amount of time before removing it...options abound, but the blunt reality in the end is that if no editor can be found who can or will provide reliable sources for the challenged material then it doesn't belong in the article until that changes. Doniago (talk) 19:09, 20 March 2013 (UTC)

@Ed: Removal of content simply for being unsourced is against the letter and spirit of policy. See WP:PRESERVE. But there's nothing wrong with adding {cn} tags to unreferenced content. A Quest For Knowledge (talk) 19:03, 20 March 2013 (UTC)

After reading WP:PRESERVE I think there's enough vagaries in it that it's a bit of a misrepresentation to say that it states that "removal of content simply for being unsourced is against the letter and spirit of policy". If anything, IMO it focuses on alternatives to deletion and possible "better practices" but does not mandate such. Doniago (talk) 19:14, 20 March 2013 (UTC)

(edit conflict)@Ed: If you'll look at the last two archives of this page and read persistently, you'll find exhaustive discussion on this very issue with every possible position and interpretation (and every possible policy modification) discussed in both excruciating detail and excruciating length. With all respect to Quest, it is not against the letter of policy — virtually everything in PRESERVE is advisory, not mandatory, and the spirit of policy is, frankly, in the eye of the beholder. The best practice is, unquestionably, not to remove unsourced or inadequately–sourced material unless you have made a good faith effort to find sources for it, but that's a best practice not a mandate. There's an argument to be made that an editor who continually and demonstrably fails to follow best practices ought to be blocked, but that's a long term process, not a single-instance one. Regards, TransporterMan (TALK) 19:17, 20 March 2013 (UTC)

@Doniago: I don't agree at all. The section named "Problems that may justify removal" lists out the various reasons why content can be deleted. Simply being unsourced is not listed anywhere. A Quest For Knowledge (talk) 19:34, 20 March 2013 (UTC)

As TransporterMan just pointed out, Wikipedia talk:Verifiability/Archive 60 is an exhaustive treatise on just about every conceivable opinion on this matter. (And one that I'm not eager to relive.) I think it's fair to say that different editors have different opinions about large scale deletions and large scale tagging. But the center of gravity, such as it is, is that it's permissible to do this when the material is uncited, but it also risks pissing a lot of people off. It's a lot smarter to pay attention to what it says in Footnote 3 of the Burden section – not precisely mandatory, but a lot more sensible if one wants to avoid ending up on other people's shit lists. And, when being pissed off by that behavior, a smart option to consider is to simply restore the material, but to provide an inline cite of one's own. --Tryptofish (talk) 19:42, 20 March 2013 (UTC)

And when they say "exhaustive", they really, truly mean it.--Amadscientist (talk) 19:54, 20 March 2013 (UTC)

Yes, I'm exhausted just thinking about it. --Tryptofish (talk) 20:08, 20 March 2013 (UTC)

Well, this conversation has been very enlightening. Thank you all. I'd love to see even more opinions, but I'd understand if you would be too tired to provide more. :-) Ed ^{[talk] [majestic titan]} 00:39, 21 March 2013 (UTC)

I am curious about the term you used: "tag-bomb". I suppose a better course would be to just AFD the article. Would that have been better? I looked at the article you demonstrated and the single reference the article contained was a dead link to an archived article with no route that could be found to locate it. I am unsure there is sufficient reasoning to retain such an article if just tagging all the contentious claims is "Tag bombing" I guess just "nuking" it would have been better.--Amadscientist (talk) 00:44, 21 March 2013 (UTC)

I'll be getting around to PRODing it—I'm more concerned that the editor in question here has done that to many articles, notable and not. Ed ^{[talk] [majestic titan]} 08:50, 21 March 2013 (UTC)

Hi. Based on my own experience, people are more likely to opt for deleting a tag-bombed article, than a clean article. Best regards, Codename Lisa (talk) 02:39, 21 March 2013 (UTC)

...which is silly, seeing as we're supposed to determine if they are notable, etc. (not if they don't have enough references), but human psychology is like that, I guess. (that's not aimed at you, by the way, just the "people" you are referring to!) Ed ^{[talk] [majestic titan]} 08:50, 21 March 2013 (UTC)

That's one of the problems I've noticed with the AfD process. Some editors will !vote based on the quality of the article as opposed to the notability of the topic. A Quest For Knowledge (talk) 11:25, 21 March 2013 (UTC)

That's partly because there's a strong correlation between the two. It's virtually impossible to write a decent quality article without building it on sources. And certain kinds of bad writing are giveaways for a lack of notability, for instance lots of personal opinion, weasel words, and puffery. Of course, it's possible to write a junk article on a worthy topic so the correlation isn't perfect, but it's there nonetheless. Reyk _YO! 00:02, 22 March 2013 (UTC)

I concur with Codename Lisa, Doniago, and Tryptofish. In my view, adding specific article inline cleanup templates to an article are helpful in showing what is specifically challenged. Article-level {{refimprove}} tags are ugly, and do not accomplish the same thing; dedicated article-specific editors will often take umbrage at removal of challenged content when the challenge left was article-level and non-specific. Just human nature, I guess.

In the end, we cannot control what motives bring various editors to Wikipedia. WP allows the removal of uncited/unverified material per policy, but policy of course also asks editors to do this in a WP:CIVIL fashion. Wikipedia is an emergent phenomenon in a human complex adaptive system. Since all editors are volunteers, they will each, by virtue of the nature of their interests, work on quite different aspects of what it takes to construct a great online encyclopedia resource. Part of that is bringing articles up to standard where sources are identified, with inline citations, for challenged article statements. The old standards of circa 2005 for Wikipedia sources are no longer the policy standards today. Since the Burden of Evidence is on the editor who wants to add (or retain) an unsourced assertion in a Wikipedia article, it seems simple common courtesy to place some notification on an article before possibly removing the unsourced material at a later time, a removal that policy supports. N2e (talk) 19:54, 22 March 2013 (UTC)

That comment does not seem to acknowledge that systematically finding unsourced articles then proceeding to remove significant content is not helpful to the encyclopedia. The fact that an editor might tag an article to allow other editors to do the work of fixing it does not remove the problem that systematically gutting content is destructive. If an editor focused much of their editing on articles about birds, it would be fine for them to periodically assess such articles for text that appears superfluous or unlikely, then tag it for sourcing, or just remove it as unsourced and unnecessary. However, it would be most unhelpful for an editor to make a habit of visiting random pages where they had no particular understanding of the topic, then ensuring that all unsourced text is removed. Editors need to use judgment: if text appears to be unhelpful for an article, remove it, particularly if unsourced. However, do not use V as an excuse to remove text that appears reasonable and which may be the basis for a useful article when fixed before the deadline. Johnuniq (talk) 22:58, 22 March 2013 (UTC)

@Ed: The next time that this editor deletes a large amount of text simply for being unsourced, can you please notify this talk page (and my talk page, if it's not too much trouble)? Thanks. A Quest For Knowledge (talk) 23:21, 22 March 2013 (UTC)

Johnuniq did a good job of capturing what is also my own personal opinion. However, it seems to me that there are editors who feel really strongly that we cannot make policy against those kinds of behaviors, so the footnote that I pointed to, above, seems likely to be the most that we can currently get consensus for. But I urge editors to point those users who systematically gut content to that footnote. It is probably more workable to pursue dispute resolution against such users on the basis of WP:IDIDNTHEARTHAT after they have been pointed repeatedly to the note, than to try it on the basis of the deletions themselves. --Tryptofish (talk) 23:39, 22 March 2013 (UTC)

Wait, a light bulb just lit up above my noggin. Johnuniq mentioned, quite aptly I think, that WP:There is no deadline is applicable to adding sources in response to cite needed tags. True enough. Would it make sense to add something that says that to this policy page? --Tryptofish (talk) 23:42, 22 March 2013 (UTC)

Unfortunately, judgment is required (and the mantra is not anyone with judgment can edit). There are plenty of cases where one can find dubious text in articles—if confident it is doubtful but unsure, tag it for removal later; if confident it is dubious, remove it. In such a case, BURDEN reasonably ensures that some POV warrior or spam merchant can't simply re-add the text, saying that it will be referenced later. However, that situation leaves a loophole which is sometimes misused by someone who would like to be a helpful wikignome, but imagines that deleting reasonable content is the way to do that. I saw one case in a very specialized article where an extremely misguided editor was tagging and deleting text that was like "the sky is blue", but which did not make sense to the editor because they had essentially zero knowledge concerning the topic. There is no good way to handle such problems, other than wasting time trying to educate the editor concerned. Johnuniq (talk) 00:26, 23 March 2013 (UTC)

Someone pointed to this discussion and suggested that I give my two cents. My recent experience with using the dispute resolution process to get a user who was "systematically finding unsourced articles then proceeding to remove significant content" (for example, the location of the main campus of a major Australian university) was not especially productive. In fact, it was the impetus for my retirement (and I expect that, now that I'm commenting on this discussion, while retired, an ANI thread will shortly be started bleating for my blood again).

The fact that the footnote is posed in modal language ("should", "may", etc.) and the fact that it's a footnote gives license to editors to completely and wilfully ignore them (particularly where they are connected enough to be able to use numbers to protect them from any part of the dispute resolution process). WP:BURDEN is used to gut articles, particularly in pursuit of a particular POV. Regardless and at the expense of any other consideration, it "is broadly correct to remove unsourced content".

But I'm pleased to see this being discussed again. My impression is that a footnote is beyond pointless. It essentially tells people that they can and should ignore it. I won't be participating further in this discussion, just wanted to contribute. ˜danjel [ talk | contribs ] 12:44, 23 March 2013 (UTC)

Well I, for one, would be happy to strengthen the language and make it more prominent. Maybe consensus is evolving, and we should reconsider the language, to make it more forceful. I'd support doing so. (Cue the objections!) --Tryptofish (talk) 22:39, 23 March 2013 (UTC)

Really, Tryp? You want to discuss it again this soon? I'd rather stick my hand in a blender. It's not evolving, it's just that people who have strong feelings about this (like Blueboar and me on the opposite, leave-it-alone side) are the ones who are willing to weigh in whenever this comes up. Best regards, TransporterMan (TALK) 03:30, 24 March 2013 (UTC)

I guess my cue worked! Please don't injure your hand. I was just testing where things stood, and you answered. OK folks, nothing to see here, time to move along. --Tryptofish (talk) 00:30, 25 March 2013 (UTC)

You can just ask me politely on my talk page, and I will be happy to tell you that all unsourced content should generally be deleted upon detection. What I have seen over time is that most of the unsourced content that people get in a tizzy about is actually content that doesn't really belong in the article in the first place. I may be able, with research, to find a source for the linoleum pattern in a fraternity hallway, but it's more reasonable to just delete nonsense like that than it is to chase down a source and then argue that it isn't important enough to include anyway.—Kww(talk) 00:57, 25 March 2013 (UTC)

Did somebody ask you impolitely?? --Tryptofish (talk) 01:00, 25 March 2013 (UTC)

Copyright and translations (again)

Diff. My position is that copy/pasting text from copyrighted sources onto Wikipedia is a copyright violation, irrespective of whether the source is attributed, and I'm curious to know on what basis Amadscientist disagrees.—S Marshall T/C 12:51, 23 March 2013 (UTC)

On the basis of current copyright case law and our policies. Snippets of sources are permitted when properly attributed. Translations are no exception. This is done on talk pages all the time when reference are not available for an editor to verify.--Amadscientist (talk) 13:02, 23 March 2013 (UTC)

It may be a good idea to remind readers to include an inline citation for attribution. It's already at Copyright and plagiarism but it's worth repeating at Non-English sources. Diego (talk) 13:15, 23 March 2013 (UTC)

Agreed.--Amadscientist (talk) 13:17, 23 March 2013 (UTC)

If copying and pasting text was always a copyright violation, we would be unable to include any direct quotes. The added text (although it is not written as well as possible) seems to says that if we are already quoting a foreign language source, we should give an English translation in a footnote. That does not seem to cause any additional copyright problems compared to just quoting the material without a translation. — Carl (CBM · talk) 13:34, 23 March 2013 (UTC)

SMarshal... Just want to double check... is your concern about copyright... or is it more about plagiarism? Blueboar (talk) 14:56, 23 March 2013 (UTC)

My concern is about copyright. Plagiarism is not a crime, it's merely an academic discourtesy. Copyright is a legal matter that we have to take seriously: the imperative not to breach copyright trumps even the principle of verifiability.

As for Amadscientist's argument: copy/pasting properly-attributed snippets is of course perfectly acceptable. The sentence that I removed says: "When quoting a source in a different language, provide the original text and an English translation, either in the body of the article or in a footnote." This is absolutely stupid and cannot be allowed remain in the policy as-is. It is not a request to consider including properly-attributed snippets where appropriate. It's a broad-brush, one-size-fits-all commandment to copy/paste the entire relevant portion of the source onto Wikipedia. Please remove this travesty and return it to the consensus language on translations, which is nuanced, refers to "courtesy", and contains no bold text.—S Marshall T/C 22:17, 23 March 2013 (UTC)

I think that copyright makes a distinction between brief passages, quoted as commentary on the original work, and lengthy reproductions, as editors have sort of been suggesting here. (And, by the way, lengthy reproductions still violate copyright even if they are attributed.) In that case, let me suggest a middle ground: keeping a version of the language, but adding a phrase to it, to indicate that we are talking about an English translation of a brief passage, not something lengthy. --Tryptofish (talk) 22:45, 23 March 2013 (UTC)

That's not necessary because the policy already mentions an English translation of a brief passage. It says "When addressing accuracy disputes involving citations to non-English sources, a translation of the relevant portions of the original may be provided in a footnote, as a courtesy." (This appears directly below the disputed sentence.) Can anyone explain who added this ludicrous new sentence and why?—S Marshall T/C 22:59, 23 March 2013 (UTC)

Doesn't the sentence in question only apply when we have already quoted the foreign language source? The footnote in question is the English summary. The language says nothing about when we merely refer to or cite a foreign language text. — Carl (CBM · talk) 23:07, 23 March 2013 (UTC)

If so, then the disputed sentence means that where we've already quoted the foreign language source, we've still got to provide both the original text and the translation in a footnote. Do you think that was seriously the intention?—S Marshall T/C 23:30, 23 March 2013 (UTC)

• Actually, look, this is where it got mangled: diff. It's a good-faith attempt at cleanup by Joy, following a discussion that's now in WT:V archive 60, that unfortunately seems to have mangled the consensus text. If it's got both limbs, it becomes much clearer in context. Can we restore the version predating that diff, please?—S Marshall T/C 23:36, 23 March 2013 (UTC)

← That version has the same text about quoting non-English sources. In fact that text about putting the ENglish translation in a footnote dates back to Feb 2008 [15]:

"Where editors use their own English translation of a non-English source as a quote in an article, they should provide a direct quotation of the relevant portion of the foreign-language original in a footnote or inline with the article text, so that readers can check the original source and the accuracy of the translation."

As the general acceptability of foreign language sources increased, the language was reversed, and modified so that the translation and original could be in the article text or in a footnote, by March 2011 [16]

"When quoting a source in a different language, provide both the original-language text and an English translation in the text or a footnote. "

That is the same as the text that is being discussed now - so this text is at least one year old, but the general principle to include the "other" version in a footnote is much older No version requires anything to be included twice - the original quote and the translation can each be in the text or in a footnote. — Carl (CBM · talk) 01:52, 24 March 2013 (UTC)

• Which only goes to show that I've been active on this page for far too long.  :( The version I recall didn't have any bold text. Anyway----my position is still that we can't leave that as it currently is. It reads like an instruction to copy/paste large chunks of sources. With its bold text and its bullet point, it's got far too much emphasis for its importance. I strongly prefer the version of this paragraph from this diff (which CBM posted earlier) and would like to return to it.—S Marshall T/C 02:29, 24 March 2013 (UTC)

I would say that quoting large chunks of any source text is wrong... But that has to do with the size of the quote, not what language it is in. Blueboar (talk) 03:52, 24 March 2013 (UTC)

We may be moving a bit past the OP's original intent here which was to remove a part of our policy one verifiability. Since the discussion has progressed it seems clear that consensus is to keep the text (probably with a removal of the bolding) and to clarify with additional prose reminding editors that all quotes must be attributed to a reliable source. I don't believe there is any real issue in the prose about the size of the quote, but that could be mentioned as well. The actual section is a little badly written and has some odd redundancy that should be dealt with. Here is what we currently have:

Non-English sources

Because this is the English Wikipedia, English-language sources are preferred over non-English ones, assuming English sources of equal quality and relevance are available.

• When quoting a source in a different language, provide the original text and an English translation, either in the body of the article or in a footnote.

When addressing accuracy disputes involving citations to non-English sources, a translation of the relevant portions of the original may be provided in a footnote, as a courtesy.

Translations published by reliable sources are preferred over translations by Wikipedians, but translations by Wikipedians are preferred over machine translations. When using a machine translation of source material, editors should be reasonably certain that the translation is accurate and the source is appropriate. When posting original source material, be careful not to violate copyright; see the fair-use guideline.

I propose the following if acceptable:

Non-English sources

Because this is the English Wikipedia, English-language sources are preferred over non-English ones, assuming English sources of equal quality and relevance are available. When quoting a source in a different language you may provide the original text and an English translation, either in the body of the article or in a footnote, provided that it is attributed to a reliable source. All quotations must be limited in length to comply with copyright.

Should a dispute arise involving citations to non-English sources without a translation, as a courtesy, the relevant portions should be provided as a footnote.

Translations published by reliable sources are preferred over translations by Wikipedians, but translations by Wikipedians are preferred over machine translations. When using a machine translation of source material, editors should be reasonably certain that the translation is accurate and the source is appropriate. When posting original source material, be careful not to violate copyright; see the fair-use guideline.

--Amadscientist (talk) 06:10, 24 March 2013 (UTC)

I have boldly edited the text to add the above wording to address the copyright concerns.--Amadscientist (talk) 09:29, 24 March 2013 (UTC)

I have removed a "you may" from the live text. This was hammered out over a long period of time, and as I documented above as long ago as 2008 we had a firmer requirement than just "you may". We are talking about a situation where there is already a direct quote of a non-English source - it is not particularly burdensome to include an English translation inline or in a footnote, and in any case that has nothing at all to do with copying chunks of a source, it is just a translation of a quote that is already included. Moreover, including the English translation helps any reader who does not speak the non-English language well enough to read the quote, so it helps not only with verifiability, but with general readability. — Carl (CBM · talk) 12:37, 24 March 2013 (UTC)

Agreed. Would this effect the wording about a dispute arising? Is that needed now if we are not just suggesting an English translation but pretty much requiring it?--Amadscientist (talk) 12:41, 24 March 2013 (UTC)

It should be a suggestion not a requirement.—S Marshall T/C 21:53, 24 March 2013 (UTC)

different approach?

I have been bold and tried a different approach...

Non-English sources

Citing non-English sources:
Citations to non-English sources are allowed. However, because this is the English language Wikipedia, English-language sources are preferred over non-English ones, assuming English sources of equal quality and relevance are available. If a dispute arises involving a citation to a non-English source, quote relevant portions of the non-English source (in its original language) as a footnote.

Quoting non-English sources:
When quoting a non-English source (whether in the main text or in a footnote), a translation into English should always accompany the qoute. Translations published by reliable sources are preferred over translations by Wikipedians, but translations by Wikipedians are preferred over machine translations. When using a machine translation of source material, editors should be reasonably certain that the translation is accurate and the source is appropriate. Do not use machine translations of non-English sources in contentious articles or biographies of living people—ask an editor who speaks the language to translate it for you.

When quoting any material, whether in English or in some other language, be careful not to violate copyright; see the fair-use guideline.

Blueboar (talk) 13:15, 24 March 2013 (UTC)

I like your formatting and most of the prose but have altered a small portion. We cannot be stricter with wording than the BLP policies and an editor that can speak a language may not be able to write it.--Amadscientist (talk) 21:51, 24 March 2013 (UTC)

Machine translations can't be considered reliable sources.—S Marshall T/C 21:53, 24 March 2013 (UTC)

True but neither are Wikipedian translations.--Amadscientist (talk) 21:54, 24 March 2013 (UTC)

On the contrary, the fact that a Wikipedian has translated something does not make it unreliable. We routinely trust Wikipedians to read and summarise sources. That's the nature of being a Wikipedia editor. If you add something to an article based on a reliable source, and you cite that source fully, then your contribution is considered reliably-sourced. The only exception is for certain editors who've been shown to misuse or falsify sources, but that's a rather serious conduct issue that's outside the scope of WP:V.

Machine translations are inherently unreliable (although the degree of reliability depends on the language). I've just run a machine translation of a randomly-selected article (Louis XIV on ja.wiki) and the opening paragraph translated into English via google translate reads: "Louis XIV ( French : Louis XIV , 1638 September 5 - 1715 , September 01 ), the Bourbon of the 3rd French king reign ( 1643 May 14 - 1715 September 01 ). Louis XIII firstborn. Princess is Spanish king Philip IV 's daughter Marie-Therese Dauphin Trish (Maria Theresa). In the heyday of the Bourbon king sun king was called (Roi-Soleil)." See? It is not right to base content in a contentious article or BLP on a translation of this quality. But I'd trust a japanese-speaking Wikipedian to translate it.—S Marshall T/C 00:02, 25 March 2013 (UTC)

I'm sorry but you are shifting the discussion now. No offense but please stick to the actual wording and meaning that you began with. Wikipedians are NOT reliable sources. Period. Do we allow them to make translations on their own? Undeniably, but they are never considered to be "Reliable sources". In order for anything to be considered a reliable source it must have been previously published and there are indeed many reliable published translations of many non English sources. There is a reason why machine translation is not preferred...because they translate words, not meaning. But they are accepted, while not being preferred. Also, what you may trust is not the same level of trust I would have. Being able to speak Japanese does not mean that one is capable of an English translation in written form. One would have to be able to read Japanese, and that really is a very big difference. So, yeah, I would trust a machine translation of Japanese text above anyone that just knows how to speak Japanese.--Amadscientist (talk) 00:53, 25 March 2013 (UTC)

By the way, technically, a translation is an analysis and interpretation of text. It isn't a summary of the text. --Amadscientist (talk) 00:57, 25 March 2013 (UTC)

What are you talking about, Amadscientist?—S Marshall T/C 01:13, 25 March 2013 (UTC)

Not a bunny with a pancake on its head.--Amadscientist (talk) 01:15, 25 March 2013 (UTC)

Amadscientist... you are correct... Wikipedians are not reliable sources. However, a Wikipedian is not the source for what we are talking about... the original non-English text is the source. The Wikipedian is simply providing a courtesy translation of that original source, for the benefit of those who don't know the original language. If someone questions that translation, we can get other wikipedians who know the original language to check it and provide a better translation. But the source remains the original non-English text. Blueboar (talk) 01:33, 25 March 2013 (UTC)

Oh yeah. I'm not questioning that at all. Just replying to the statement that Marshal was stating. But...a translation by a Wikipedian is the one area were we allow an analysis and interpretation of the source, that would normally require a secondary source to make. The original intent of the OP was to claim that the policy page was stating something that was a copyright issue. My point is that technically...Wikipedians are not allowed to make such analysis and interpretations of sources. Now, I am not really questioning that this should be allowed, but it is clearly one policy that ignores another and I am not exactly sure if it is an improvement to the project to allow it. But I am not really aiming at changing it.--Amadscientist (talk) 01:43, 25 March 2013 (UTC)

Evaluating, summarising and interpreting sources is the main task of the encyclopaedist and it's exactly what we should be doing.—S Marshall T/C 01:49, 25 March 2013 (UTC)

We don't interpret the sources. We summarize them. Any analysis and interpretation must be made by a reliable source.--Amadscientist (talk) 01:51, 25 March 2013 (UTC)

"All interpretive claims, analyses, or synthetic claims about primary sources must be referenced to a secondary source, rather than to the original analysis of the primary-source material by Wikipedia editors." per Wikipedia:No original research in the section WP:PSTS.--Amadscientist (talk) 01:53, 25 March 2013 (UTC)

WP:OR states in the lede: "Wikipedia articles must not contain original research. The term "original research" (OR) is used on Wikipedia to refer to material—such as facts, allegations, and ideas—for which no reliable, published sources exist."--Amadscientist (talk) 01:59, 25 March 2013 (UTC)

And: "To demonstrate that you are not adding OR, you must be able to cite reliable, published sources that are directly related to the topic of the article, and directly support the material being presented"--Amadscientist (talk) 02:11, 25 March 2013 (UTC)

The fact that sources are necessary doesn't mean the encyclopaedist isn't interpreting them. The process is: Read, comprehend, think, then write.

You can't possibly create decent articles by taking snippets that other people have written, rephrasing them in your own words and stringing them together into paragraphs. Such an article would be incoherent, and in breach of our guidelines on close paraphrasing as well. The only way to write a decent article is to research your topic, learn enough about it to gain a balanced understanding of it, write your text, and then go back through it checking against your sources and adding citations where appropriate. And this means that an article is, and can only ever be, Wikipedians' interpretation of the topic. Yes, we still have to be neutral and no, we can't carry out original research, but that doesn't eliminate interpretation. The only viable process for writing articles is: read, comprehend, think, write.—S Marshall T/C 09:41, 25 March 2013 (UTC)

I am beginning to suspect we are arguing about abstract concepts, and not the current language of the page. So let's re-focus. Does anyone have a concern with the current language of Wikipedia:Verifiability#Non-English sources. If so, what is the concern and how would you suggest we resolve it? Blueboar (talk) 16:59, 25 March 2013 (UTC)

Not sure how abstract it really is. Simply put, we don't attempt to interpret the information we just use original prose to summarize the facts. This really is an issue that may be a slight loophole where our policies are promoting a behavior that really isn't appropriate. I don't object to the wording per say...just that, when you come right down to it, we probably shouldn't be allowing editors to create original translations. We really should only allow previously published translations that can be verified.--Amadscientist (talk) 05:44, 26 March 2013 (UTC)

@Blueboar: I think the current wording is quite good.

@Amadscientist: I don't agree. The source being in a different language is an accessibility problem similar to the source being available only in a few libraries. But translations can be checked by others if the original is provided, so these sources can be even easier to verify than many English ones. — HHHIPPO 08:59, 26 March 2013 (UTC)

Absolutely. I wasn't raised in a multilingual household---I worked hard in school to learn my languages. If other Wikipedians can't read foreign languages, then that's their problem and not mine. In building articles, I should be entitled to use all the reliable sources that I can read.—S Marshall T/C 12:20, 26 March 2013 (UTC)

@Amadscientist - Re: "We really should only allow previously published translations that can be verified"... I think you are half right... I think we all agree that we should quote previously published translations if they exist... but what if no previously published translations do exist? If we quote a non-English source, we really should provide a translation as a courtesy to our readers (who may not understand the original language). And if no published translations exists, we have to rely on translations by Wikipedians. Now... If someone thinks there is a flaw with that translation, the solution is quite simple... swap in a better translation. And if a published translation becomes available, swap in that published translation.

In some ways (and I admit the analogy is not perfect) this reminds me of the debates we had over user-created images. Suppose you want to include a image of a notable person in an article... and if we can find a photograph (and are allowed to use it, per copyright)... great. But, what if we can not find a photograph that we are allowed to use. It is OK for a Wikipedian to create a drawing of the person and use that drawing. Now... suppose someone objects because that drawing is not a good likeness - no problem... they can swap in another drawing that is a better likeness. And if a photograph becomes available, we can swap in that photograph. Blueboar (talk) 13:04, 26 March 2013 (UTC)

In politically contentious fields, mistranslation is one of the most common weapons of propagandists. We need to be careful not to mandate the use of a mistranslation just because it appears in a "reliable source" when we are looking at the original and know for a fact that it says something else. Of course this becomes original research when any depth of textual analysis is required, but I've seen plenty of cases where the foreign text is completely plain (i.e. every fluent reader of that language would understand it the same way) and yet the only published translation claims something else. I'm not sure what I'm proposing here, maybe nothing, but the point I want to make is that the original source itself has the highest status and it is the meaning of that source that we should strive to get into the article. I don't think it is much different from when English source B misquotes English source A; if A is in itself a reliable source we should use it. Zero^talk 00:23, 27 March 2013 (UTC)

Zero, I agree but that issue exists even without translations. We call it misrepresentation. @Aman: I also have to disagree with you. A source exists even if it is not in English. If we were to create an artificial burden that sources must be translated to English by a secondary source, we are going to bias ourselves toward English-language point of view and also limit our pool of sources. WP:V requires that verification be possible, not that it be easy. Using non-English sources is equivalent, if not a lesser evil than, using pay-wall sources. At least Wikipedia already has a large pool of bi-lingual editors of just about every language and English. At some point, as with paywalls, WP:AGF needs to come into play for the translator.--v/r - TP 14:01, 27 March 2013 (UTC)

I object to the existence of this direct order: "If a dispute arises involving a citation to a non-English source, quote relevant portions of the non-English source (in its original language) as a footnote."

Why is it only non-English sources that require this? What should you do this for a non-English online source but not for an offline English source with extremely limited circulation? Why must it be in a footnote, rather than on the talk page?

We agreed last round that non-English sources should be treated exactly like any English-language source, meaning that if there is a dispute, you should provide a quotation as a courtesy (assuming it wouldn't require quoting whole pages). This is exactly what you would do with a dispute involving an English-language source. WhatamIdoing (talk) 20:24, 29 March 2013 (UTC)

Our policy as written seems to allow something here we don't allow in other areas. I disagree with TParis here but I do agree with Zero0000. Its a slippery slope regardless which direction we take but I am still concerned.--Amadscientist (talk) 01:34, 30 March 2013 (UTC)

WAID - I am not sure if providing a quote is "just a courtesy"... Your point that the language does not matter is valid, but providing a quote when someone requests one is standard procedure per WP:BURDEN (I am not sure if this is explicitly stated anywhere, but it is accepted practice... and we have Template:Request quotation in order to make the request). What is a courtesy (but one that we should strongly push) is providing a translation of the quote, when the text we are quoting is non-English. Blueboar (talk) 02:32, 30 March 2013 (UTC)

Footnote 9 says that all quotations are provided as a courtesy, and that's not specific to NONENG.

Quotations are not required under BURDEN. Your sole requirement under BURDEN is to provide an inline citation to a source that you believe supports it. After that, if I want to say that your cited source doesn't support it, then you have the option of courteously providing me with quotations (assuming it's possible to support it with a non-copyvio excerpt, which is not always the case) or the option of telling me to look it up myself. There is no requirement that you make it easy on me to verify your sources. WhatamIdoing (talk) 05:04, 31 March 2013 (UTC)

(ec) Then why do we have Template:Request quotation? My understanding has always been that this is a legitimate challenge under WP:BURDEN, (if that tag is placed on a citation, someone is challenging whether the source actually does support the statement... and the burden to "prove" that it does falls on those who cited it). My understanding was that this was not a courtesy... but a necessary response to a legitimate challenge. Blueboar (talk) 14:09, 31 March 2013 (UTC)

If we're going to allow machine translations in contentious articles and BLPs (as is apparently the case from Amadscientist's revert), then I think we probably do need to introduce special rules for foreign-language sources. We do need the original text so that someone who speaks the language can more easily check for translation issues.

Let me give you a couple of examples of potential problems. As of today's date, Google translate says "Ich bin warm" is German for "I am warm". (Actual meaning: "I am gay".) It gives exactly the same translation for "ertrinken" and "ertränken" ("drown"). They both do translate as "drown", but getting confused between those two can turn a description of a tragedy into an accusation of murder, or vice versa.—S Marshall T/C 13:58, 31 March 2013 (UTC)

So let's require that foreign language sources require a citation to that foreign language at the very least with all other ideas (quotes, machine translations, expert translations, ect) being optional. But the original source should be the primary citation so the reader can go directly to the original (and hire their own translator if they are in doubt).--v/r - TP 13:12, 1 April 2013 (UTC)

This is actually very simple.... first, we have to distinguish between citations and quotations.

Citations: We must cite reliable published sources, and English language published sources are preferred... so... if you can find a reliable published translation, cite that. If you can not find a reliable published translation, cite the original non-English published source. Do not cite a user-created translation or a machine translation (neither is published).

Quotations: This is a separate issue from citation. Whenever you type non-English text into an article or talk page discussion (which is what you do when you quote a non-English source), you should provide a translation into English as a courtesy to readers and/or fellow editors who might not understand the original language. Published translations are best for this, user translations are second best... but machine translations (while discouraged) are allowed.

The only remaining question is... must we provide a quote if another editor demands one (example: am editor leaves a tag)? Blueboar (talk) 13:43, 1 April 2013 (UTC)

100% agree with you, Blueboar. As far as must provide a quote? I think we are no more obligated to provide one than if we were using a book as a citation and requested to provide one. It is a courtesy, but if the citation is there than the obligation of the content provider per WP:V is met and it becomes the obligation of the person challenging the source to look it up.--v/r - TP 15:13, 1 April 2013 (UTC)

I fully agree. The question of providing a quote from the original source for verification is independent of the language issue. And a quote has limited value as verification anyway: one can check if the questioned statement in an article is supported by the quoted text, but not if the quoted text is representative of the entire source or even if it is quoted correctly. So it should be encouraged, but not required. — HHHIPPO 15:45, 1 April 2013 (UTC)

I think that someday we are going to evolve to where we need to recognize that there are "degrees of strength" of verifiability, and that greater 'strength" is required when the material itself is challenged. In the discussed area, things that make it harder to "check" (source is behind a paywall, source is off line, source is obscure and off line, source is in a non-English language) all make verifiability weaker, but should not categorically rule out use of those sources. And the measures / recommendations described in this thread / above tend to strengthen the ability-to-verify (reduce the weakening of verifiability caused by use of a non-English source.) North8000 (talk) 16:01, 1 April 2013 (UTC)

No disagreement... there is no question that a "citation request" is a different type (or level) of challenge than a "quote request". But is there no burden to respond to a quotations request? Surely if I quote something, and someone else challenges the quotation (saying I either misquoted it or took the quote out of context), there is some sort of burden on me to "prove" my quotation is accurate. Why else do we have a tag to request the original text? Blueboar (talk) 16:27, 1 April 2013 (UTC)

The issue here is that requesting a quotation to support a claim is independent of language, and so we should not have 'special' rules for non-English sources.

As for whether a request for a quotation is actually a challenge, i.e., if you don't provide a quotation then the cited text can be removed, I don't believe that's a good rule to set no matter what the source is. What if providing the quotation requires such a long (and/or important) stretch of text that it would be a copyvio? Is my unwillingness to violate a copyright grounds for removing the cited text? What if the primary purpose of the request is to accommodate laziness on the part of a POV pusher? Should you have to type up a long stretch of text solely to save someone else a trip to the library? I don't think so. WhatamIdoing (talk) 23:40, 4 April 2013 (UTC)

I have started a separate section on the "citation request" tag... it's an issue that goes beyond non-English sources. Blueboar (talk) 16:42, 1 April 2013 (UTC)

Re: Template:Request quotation

This came up above in the context of discussing non-English sources... but I think we need to discuss it in a broader context:

When should we use this template? How should we respond to it if someone adds it to an article? Is there any burden to respond to it? Under what circumstances can we remove it, and under what circumstances can we not remove it? Please discuss. Blueboar (talk) 16:40, 1 April 2013 (UTC)

• TO my mind it seems redundant with Template:Verify source (an altogether superior template), and subject to the outcome of this discussion, I'm a bit tempted to TFD it.—S Marshall T/C 17:43, 1 April 2013 (UTC)
  • How is verification...ahem...verified with the verification template? How do you know that User B verified User A's citation? Do you verify that user B cited it? To what extent can it be added in the first place and how is it logged/tracked which sources were verified?--v/r - TP 17:57, 1 April 2013 (UTC)

I would imagine the intended use of the template {{Request quotation}} to be, for instance when the tagger is a domain or language "expert" and intuitively suspects that the editor who cited the source in some way misinterpreted the source, especially if it seems reasonably likely that another editor will also misinterpret the source in the same way . I can imagine this being more likely with foreign-language sources, because many editors will not have the required mastery of the source language. I'm not sure I can think of a very good example off the top of my head, but try the following (in all cases assuming that the cited source is not easily available):

• An article describes a German politician as being regarded as "sensible and sympathetic . . ." , citing a German source.
• An article describes something as being "a legend" , citing a German source.
• An article states that a cold-war treaty granted everybody free access to Berlin, citing a West German source and implying that East Germany violated this treaty.
• An article refers to an English MP having stated something elsewhere and indirectly cites a speech in Parliament.
• An article talks of an accusation about a French politician involving kissing, citing a French investigative journalist.

Most real cases would probably be more subtle than this. I am trying to find examples that are obvious in retrospect. --Boson (talk) 23:30, 1 April 2013 (UTC)

I have seen the tag used in situations where the cited source was in English... and the tagger questions whether the cited source actually does support the statement. Example: when you suspect that the cited source has been taken out of context. Blueboar (talk) 15:09, 2 April 2013 (UTC)

Note: I am really more concerned with exploring the questions about how to respond to the tag. How should we respond to it if someone adds it to an article? Is there any burden to respond to it (and if so what is that burden)? Under what circumstances can we remove it, and under what circumstances can we not remove it? Blueboar (talk) 15:17, 2 April 2013 (UTC)

I think that really depends on the circumstances, and comes down to what North8000 said above about degrees of strength of verifiability. A quote increases the chance that a citation gets verified, and thus that it survives a challenge. However, there has to be a specific challenge to the material for it to require verification, so if there's no dispute about the content, apart from an unanswered {{Request quotation}}, I don't think that's enough grounds for removing the material.

Now I didn't really answer your question, when should the tag be removed? If the tag is set in the course of a content dispute, I'd say it can be removed once the dispute is settled. If there's no dispute, but only this tag, it should probably stay for a while to give people a chance to respond, but it feels wrong to leave it there forever. The template documentation suggests to use the tag for references that you think may be inaccurate. Maybe a way forward could be to challenge the tag on the talk page, requesting an explanation on why the tagger suspects the reference to be inaccurate. From there, consensus should decide if the tag should be removed or escalated to {{Verify source}}. — HHHIPPO 21:55, 4 April 2013 (UTC)

Makes sense. Blueboar (talk) 13:58, 5 April 2013 (UTC)

The first sentence

banned
The following discussion has been closed. Please do not modify it.
I think that the first sentence should be expanded from: In Wikipedia, verifiability means that people reading and editing the encyclopedia can check that the information comes from a reliable source. to: In Wikipedia, verifiability means that people reading and editing the encyclopedia can check that the information comes from a reliable source, which is based on the reputation of typically the publisher or domain name. Feedback?--Knockatock (talk) 01:20, 8 April 2013 (UTC) When I undid I included a criticism of the change in the edit summary: "And it is also misleading: There are more aspects than just those two which are so typical." I can expand on this here. First, I which to clarify the grammar. "Which" is a relative pronoun which preferably refers to something mentioned explicitly in the preceding clause. But there's no such thing in the preceding clause which can sensibly fill this role. In this case the "which" is expected to refer to "the reliability of the source", but this is not mentioned explicitly. So the simple "which" is better replaced with an entire relative clause "the reliability of which". Indeed, that's how I read this sentence. The reliability of a source is in some cases based on the reputation of the publisher. But the reliability is also in some cases based on the reputation of the author, or that of the editor, or that of reviewers. There's no evidence that the reliability is any more typically based on that of a publisher as it is on any of these others just mentioned. In a common usage, to say that something is typically done in some way is to suggest that that way is more typical than any other way. So, to say that the reliability is typically based on some set of things excluding another set of things that are just as well-regarded as ones on which the reliability is typically based is misleading. --Atethnekos (Discussion, Contributions) 02:21, 8 April 2013 (UTC) Evaluating sources for reliability is nuanced and complicated. It's also the central task of the encyclopaedist—the single most important thing we do. We have a whole page that explains how to go about it at WP:RS. I don't think we should try to summarise that page in one sentence here. What we should simply link to WP:RS as we already do. (I don't think we should try to summarise WP:N either and would be in favour of removing that part from WP:V, but when I have tried to do so in the past I have been reverted.)—S Marshall T/C 11:24, 8 April 2013 (UTC) In case anyone doesn't realize it, the opening poster has been indefinitely blocked. --Tryptofish (talk) 21:05, 8 April 2013 (UTC) Just for the sake of anyone looking through the archives in the future: 'Source' means three things on Wikipedia, and 'publisher' (which includes the publisher's editors) is only one of them. So naturally the publisher's identity is not the sole factor, or even always a relevant one. WhatamIdoing (talk) 05:38, 9 April 2013 (UTC)

Possiblities

The second paragraph in the lead says this:

All material in Wikipedia mainspace, including everything in articles, lists and captions, must be verifiable.

I'd like to clarify the meaning of verifiable (specifically, the difference between what is verifiABLE and what is CITED). My idea for how to do this is to change this sentence to read:

It must be possible for other people to verify all material in Wikipedia mainspace, including everything in articles, lists and captions.

What do you think? WhatamIdoing (talk) 20:19, 29 March 2013 (UTC)

It seems to me that the "possible for other people" language is repetitive of other language there, and I'm not really persuaded that there is a problem here that needs to be fixed. --Tryptofish (talk) 01:16, 30 March 2013 (UTC)

I'm thinking that the meaning of "verifiable" there means "per the specifics of this policy", whereas "It must be possible for other people to verify all material" seems to creating or trying to create its own stand-alone definition, which is impossible to do well in one sentence. A second issue is that IMHO it essentially says that all material must be cited. Sincerely, North8000 (talk) 01:23, 30 March 2013 (UTC)

That's not really how I read it, North----I think that's exactly what it's trying to avoid saying----but for the moment I'm with Tryptofish (on both the things he says).—S Marshall T/C 01:59, 30 March 2013 (UTC)

I would take it one step further:

It must be possible for other people to verify that all material in Wikipedia mainspace (including everything in articles, lists and captions) is supported by a reliable source.

That's really what we mean by "verifiable". Blueboar (talk) 02:16, 30 March 2013 (UTC)

It must be possible, but not always easy, for other people to verify material in Wikipedia mainspace (including everything in articles, lists and captions) is supported by a reliable source.

Minor change.--v/r - TP 15:10, 30 March 2013 (UTC)

I would suggest "necessarily" instead of "always". Question... are we trying to squeeze too many concepts and ideas into one single sentence? Blueboar (talk) 17:06, 30 March 2013 (UTC)

If the meaning of verifiable is unclear, then it should be fixed in the first paragraph of the lead where the meaning is given, specifically the first three sentences of the lead.

"In Wikipedia, verifiability means that people reading and editing the encyclopedia can check that the information comes from a reliable source. Wikipedia does not publish original research. Its content is determined by previously published information rather than the beliefs or experiences of its editors."

Good luck. --Bob K31416 (talk) 18:03, 30 March 2013 (UTC)

I might say "lotsa luck", as in don't bother with any of this. The lead has been thankfully stable since that last drama-fest, I mean, RfC. Until someone can actually show that the existing language is causing misunderstandings or other problems, all of this twiddling with the language is going to go nowhere. --Tryptofish (talk) 22:58, 30 March 2013 (UTC)

I have to agree. I am not clear what the purpose of the change is for. Can it be demonstrated in some manner?--Amadscientist (talk) 01:50, 31 March 2013 (UTC)

As I said above, the purpose is to clarify the difference between what is verifiABLE and what is CITED. This policy requires material to be verifiABLE. It does not require material to be CITED. We fairly frequently encounter people who believe that 100% of the articles in Category:Articles lacking sources are in active violation of this policy solely because there are no little blue numbers to click on. It's not true, everybody who hangs out at this page knows it's not true, and we could (I believe) improve comprehension on this point by not assuming that all of our editors grasp the meaning of the -able suffix in that sentence, and instead spell it out: rather than "verifiable", we say "possible to verify".

I do admit that I'm leery of tinkering with anything in the lead after the last round of pain. WhatamIdoing (talk) 23:47, 4 April 2013 (UTC)

It's actually very helpful that you described the problem as you see it. Instead of trying to change the lead, which I suspect would do little or nothing to educate the users who think that way, we could consider adding a statement lower down on the policy page, specifically addressing that misunderstanding. I could very likely support that. --Tryptofish (talk) 01:00, 5 April 2013 (UTC)

Yeah... a section on this lower down on the page would be much better. I have always thought that we should say something along the lines of: "All material must be verifiable, but not all verifiable material must be cited."... followed by a short paragraph of explanation so editors better understand what kinds of material a) must be cited, b) should be cited, and c) do not need to be cited... (and why). Blueboar (talk) 01:30, 5 April 2013 (UTC)

Well, a logician would say that that is just repeating what's already in the policy, but it's often not a good idea to listen to logicians. I think that if we can come up with a simple statement which says something which is only abstractly in the policy it's a good idea. I think that your "All material must be verifiable, but not all verifiable material must be cited." is that. Maybe also summary whihc you followed with, but I'm not as sure about that. North8000 (talk) 02:03, 5 April 2013 (UTC)

Just thinking out loud here, but an alternative approach would be to frame it in practical instead of conceptual terms, because our target readership might better understand it that way. In other words, something like: "When you encounter uncited content, do this and don't do that.". --Tryptofish (talk) 22:52, 5 April 2013 (UTC)

There are so many potential scenarios that I think that trying to write prescriptively might tend to be too complex.difficult. North8000 (talk) 21:56, 8 April 2013 (UTC)

I'd love to have a section on this, but I figured that would be asking for too much. IMO the endless list of scenarios would be better placed on a {{supplement}} page, but a short (one paragraph? two?) section could cover the principles. It might be possible to re-use a few sentences from Wikipedia:Inline citation#When you must use inline citations for this. WhatamIdoing (talk) 05:36, 9 April 2013 (UTC)

I agree with both of you, and I'm simply thinking about it in terms of a practical rather than conceptual approach, but not going into a long list of specific circumstances. I'll try to post here in talk what I mean by that, but I probably won't get around to it for a couple of days. --Tryptofish (talk) 22:30, 9 April 2013 (UTC)

YouTube videos as a source citation

I have a question about if YouTube videos can be used as a factual source. Please help guide me on this matter. I am working on an article Erica Andrews. Andrews had appeared on the Tyra Banks, Maury Povich, and in the music video of some singers. None of the videos were uploaded by the original author of the shows, i.e. not by Tyra Banks or Maury Povich but by other people into YouTube. When you view the YouTube videos, it is very evident that she appeared in the the video clips from those shows that were aired. I feel this video content has value to the Wikipedia article on her. However, the content that I had written into Wikipedia leveraging these YouTube videos as a source to prove that Andrews did indeed appear in those shows were removed by other Wikipedia authors who insist that YouTube videos are not allowed on Wikipedia. I had even noted in the Wikipedia article at what time marker you could see Andrews in the music videos that she appeared in. She did not play a large part in some of the music videos - she had a cameo role and in one case, she did have a larger supporting role. The music videos itself (not in any other printed or online published article) is all that is evident that she did partake in those projects. Please guide me on how videos can be sourced.

I also have a question as at what point does a content that is very well known by people, and is published on a blog can be used as a source. One of the problems encountered when authoring this Andrews article is that while her performances and work was very known to people in the LGBT community, they were not often mentioned or published in "well-known" print/online publications and if they were, they were by smaller publications or social media sites (MySpace, Facebook, blogs). I used the information from http://en.wikipedia.org/wiki/Wikipedia:Verifiability#Self-published_or_questionable_sources_as_sources_on_themselves as a guide on this matter. The content found on her on these social media sites does not violate the points mentioned under "Self-published or questionable sources as sources on themselves". However, these other authors disagree vehemently. I argued that the data points written into the article is informative to a reader about who Andrews was and her achievements in her career. Again, please guide me. If what I am stating here sounds confusing, please let me know and I'll be happy to re-explain again. Thank you. Lightspeedx (talk) 14:18, 15 April 2013 (UTC)

Thank you. Lightspeedx (talk) 14:18, 15 April 2013 (UTC)

If the videos were not uploaded by the original shows, they are a copyright violation. Regardless of any question about the merits as a source, we do not link to copyright violations for legal reasons (there is a possibility, though it is unclear, that knowingly linking to copyright violations is itself a copyright violation). As for the social media sites, if it is known that the accounts were directly under Andrews control (i.e., not fan created pages, not pages created by her agency) and you are verifying only information about her, then they could be used as a source, maybe. That question is better raised at WP:RSN, however, where you'll need to state exactly what you want to verify with SNS and exact details about the source. In any event, this page is really only for use in discussing the verifiability policy, not for questions of it's application. Qwyrxian (talk) 22:41, 15 April 2013 (UTC)

Re "If the videos were not uploaded by the original shows, they are a copyright violation." — I'm not sure that's always true. If someone uploads copyrighted material to Youtube, it seems that the copyright holder can allow it and make a profit from it, for example by receiving the profit from a video advertisement that is added before the video. See http://www.youtube.com/t/contentid .--Bob K31416 (talk) 18:30, 16 April 2013 (UTC)

Bob K31416 you make a good point. The guidance I have received here has been helpful but I am still wondering what does one do if a performer is involved in a music video and they have only played a cameo role and their role is not mentioned in any print/online publication but they are really in the video. Andrews appeared in Jennifer Lopez's and Maroon5's music videos in a cameo role. In music videos, there are often many actors/performers who play a smaller part in the videos and their names are often not mentioned in a press release or by print/online publication. (Motion pictures/films tend to be more comprehensive in their cast listing and even if you played a small part, you are listed in the movie credits.) Do you not at all even mention their work in such music videos? Some authors have said since it's a small role, don't list it. I mean is that right at all? Who are we to judge about the size of that role or its significance? This is not an audition/casting call that we are running here. It would seem to me to be wrong that since all these videos are part of the body of work they have done as a performer and in an encyclopedic article about a performer, wouldn't it be right that the article be as detailed as possible about the person so that ideally the Wikipedia article about them could be used as a reference point if someone were to do research about that person? I mean what good would the Wikipedia article on them be if I am trying to research and learn more about them and I've come to find out that many of their performances aren't even listed in the article. I've also looked at other actors and performers' Wikipedia article and often times, Wikipedians list out as detailed as they can if they know of their work, even if all of their work isn't sourced to or mentioned in any "notable" online/print publication. Lightspeedx (talk) 19:10, 16 April 2013 (UTC)

I recall a case where someone put in The Great Escape that Harrison Ford was an extra in the film.[17] There was an objection on the Talk page and it was deleted. The extra looked like him, but there wasn't any evidence for that besides the resemblance. (The extra was the young man with the Nazi arm band at the beginning few seconds of this scene.) So there's that potential problem of saying Andrews was in a video without being mentioned in the credits or a reliable source. An editor could object according to WP:NOR. --Bob K31416 (talk) 19:25, 16 April 2013 (UTC)

• I think a helpful clarification to be made is that YouTube is not, itself, a source. It is a medium by which some sources can be found. Some of the videos on YouTube are high quality sources, and some of the videos are complete crap. It isn't YouTube that matters, it's the reliability of the video itself, it's official status (so we aren't linking copyright violations) and the person or organization responsible for creating the video. Trying to debate the appropriateness of YouTube as a source is like trying to debate the appropriateness of "The Library" as a source: What specific book in the library are we discussing? Likewise, when dealing with videos hosted at YouTube, the fact that YouTube itself hosts the video is entirely irrelevent. It's the specific video itself we need to judge. --Jayron32 03:50, 17 April 2013 (UTC)

Perhaps a side comment, or maybe sorta on point: I just came across this in an article at AfD whose notability is probably going to turn on three videos of TV broadcasts currently linked to Vimeo and which are almost certainly copyvios. A distinction has to be made between those videos as RS's and the links to those videos at Vimeo. The links are almost certainly improper as linking to a copyvio, but the existence of the videos proves that the article is probably verifiable, and should be cited to the original source of the videos rather than to the copies at Vimeo. Or maybe someone has said this already and I'm just being redundant. Regards, TransporterMan (TALK) 15:39, 17 April 2013 (UTC)

Wikipedia and sources that mirror or use it

The following discussion is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.

---

A reputable journal will usually point out where a company or person mentioned in an article is associated in some way with the journal. Thus an article in The Economist may include a warning like (the Financial Times is a part owner of The Economist) or (Johnstone is a former science reporter with The Economist). This is to warn readers that there may be some bias or conflict of interest. Similarly, Wikipedia may be used as a primary source for articles that discuss Wikipedia itself. This is to ask for comment on whether we should recommend inline identification of cases where this is done, changing WP:CIRCULAR to read:

'Do not use articles from Wikipedia or from websites that mirror its content as sources, because this would amount to self-reference. Similarly, do not use sources that present material originating from Wikipedia to support that same material in Wikipedia, as this would create circular sourcing. Citing Wikipedia to support a descriptive statement about Wikipedia itself is not a form of circular referencing. Instead, it is the use of Wikipedia as a primary source and should follow the relevant policy for the use of primary sources.

Where Wikipedia or a sister Wikimedia project is used as a primary source for a statement about some aspect of Wikipedia, or some event associated with Wikipedia, the article text should make it clear that the material is sourced from Wikipedia or the sister project so the reader is made aware of the potential bias.

Comments? Aymatth2 (talk) 11:23, 24 April 2013 (UTC)

Some of this looks like good advice and good practice. (except why would Wikipedia as a primary source need that notice more than any other primary source) But I'm not so sure that it should be written as policy, and IMO it is too specific and detailed to be in the the top level wp:ver policy. Sincerely, North8000 (talk) 12:26, 24 April 2013 (UTC)

• I should be clear that the first paragraph is already present. I am proposing to add the last sentence. Wikipedia is a bit of a special case as a source for a Wikipedia article, which is why this section exists. Aymatth2 (talk) 13:50, 24 April 2013 (UTC)

The case your second paragraph considers is already described in the first paragraph, "Citing Wikipedia to support a descriptive statement about Wikipedia itself...". So you don't need to describe it again with, "Where Wikipedia or a sister Wikimedia project is used as a primary source for a statement about some aspect of Wikipedia, or some event associated with Wikipedia...".

Regarding bias, I don't see how it can occur under the conditions described in the first paragraph. Perhaps you could give a hypothetical or real example? --Bob K31416 (talk) 01:29, 25 April 2013 (UTC)

• It is a bit repetitive. A simpler version follows:

ALT1. Do not use articles from Wikipedia or from websites that mirror its content as sources, because this would amount to self-reference. Similarly, do not use sources that present material originating from Wikipedia to support that same material in Wikipedia, as this would create circular sourcing. Citing Wikipedia to support a descriptive statement about Wikipedia itself is not a form of circular referencing. Instead, it is the use of Wikipedia as a primary source and should follow the relevant policy for the use of primary sources. In this case the article text should make it clear that the material is sourced from Wikipedia so the reader is made aware of the potential bias.

Suppose a new internet regulation is being discussed that changes the balance between protecting freedom of speech, respecting privacy and preventing crime, and Wikipedia could be significantly impacted. Alternatives:

1. Both houses are in favor.^[1] Others strongly oppose the measure.^[2]
2. Both houses are in favor.^[1] However, a survey by Wikipedia shows that their editors strongly oppose the measure.^[2]

The difference is maybe subtle, but I think important. By clearly identifying that Wikipedia is citing itself, we convey a greater sense of openness and neutrality. If we are not up-front, and the reader scrolls to the bottom and sees we are citing ourselves, we lose credibility. Aymatth2 (talk) 02:22, 25 April 2013 (UTC)

The policy says "Do not use articles from Wikipedia or from websites that mirror its content as sources...". In your example, it doesn't appear that an item from a Wikipedia article is being used. The report of the survey originates from outside a Wikipedia article, since the survey is original research and I presume the article complies with WP:NOR. So this example doesn't fit into the case that is being considered since the source of the info is outside a Wikipedia article. --Bob K31416 (talk) 03:47, 25 April 2013 (UTC)

That is a good point. The policy awkwardly worded. The first part refers only to articles, but the second part refers to any material from Wikipedia. A Wikipedia survey, or for that matter a policy statement or a discussion like this one, is not original research from the viewpoint of an article about Wikipedia. It may be used as primary material. An attempt to analyze or interpret the primary source is wrong, but the source can be cited. A Wikipedia survey may be deeply flawed in the way the questions were phrased, respondents were selected and results counted, but the survey and its results exist as a fact that can be reported. Ditto with a discussion: the participants may have been biased, ill-informed and stubborn, reaching a dubious conclusion, but the discussion and its outcome exist and can be reported. I am trying to slightly tighten up the rules by saying it must be made very clear that the primary source is Wikipedia. Another shot at wording below:

ALT2. Do not use articles from Wikipedia or from websites that mirror its content as sources, because this would amount to self-reference. Similarly, do not use sources that present material originating from Wikipedia to support that same material in Wikipedia, as this would create circular sourcing.

Citing a Wikipedia article, guideline, discussion or other content from Wikipedia or a sister project to support a descriptive statement about Wikipedia itself is not a form of circular referencing. Instead, it is the use of Wikipedia as a primary source and should follow the relevant policy for the use of primary sources. In this case the article text should make it clear that the material is sourced from Wikipedia so the reader is made aware of the potential bias.

Comments? Aymatth2 (talk) 12:44, 25 April 2013 (UTC)

I don't think the passage was intended to address things like the survey in your example that are outside Wikipedia articles. This can be clarified with the change that is underlined in the following,

"Citing a Wikipedia article to support a descriptive statement about Wikipedia itself is not a form of circular referencing."

--Bob K31416 (talk) 13:54, 25 April 2013 (UTC)

• If the intent were to limit the scope to articles, the policy would have been worded as you suggest. An article that discusses Wikipedia may well cite Wikipedia/media sources other than than articles. This would seem to be the norm rather than the exception. For example, the article on Protests against SOPA and PIPA includes "Wikimedia Executive Director Sue Gardner posted an announcement of the Foundation's support for the blackout proposal on Wikimedia's blog." This is supported by citing the blog, a primary source in this context. Rather than change the policy to deliberately leave open the question of how non-article Wikipedia content is to be used, it seems best to spell it out. (I have tweaked the ALT2 wording to add "or a sister project".) Aymatth2 (talk) 14:43, 25 April 2013 (UTC)

"If the intent were to limit the scope to articles, the policy would have been worded as you suggest" LOL. Grasshopper, you are clearly an apprentice. If you hang around Wikipedia policy talk pages for long enough you will appreciate that wording is driven by "Events, dear boy, events" (conversations such as this) and that details of polices are in a constant flux. --PBS (talk) 09:08, 26 April 2013 (UTC)

• It seems to me, the more appropriate thing (or at least content that addresses the more common failings that I have seen in the use of Wikipedia) would be to strongly advise against such use of primary sources, per WP:OR. -- TRPoD aka The Red Pen of Doom 15:01, 25 April 2013 (UTC)

While the use of a Wikipedia article, guideline, discussion or other content from Wikipedia or a sister project to support a descriptive statement about Wikipedia itself is not a form of circular referencing, great care needs to be taken with the use of such primary source material to avoid making claims based on original research, giving undue emphasis to the role or status of Wikipedia's impact on the topic, or making inappropriate self reference to Wikipedia's internal workings.

Aymatth2m, Using the survey example that you mentioned, the source would be treated just like any other source that is outside a Wikipedia article and might be biased. For example, the material could state, "According to a survey conducted by the staff of Wikipedia, editors of Wikipedia strongly oppose the measure.[2]" The simple change I suggested in my previous message thus wouldn't be changing policy. --Bob K31416 (talk) 15:30, 25 April 2013 (UTC)

• The change would be significant, restricting the policy to references to articles where at present it covers references to anything in Wikipedia. It is best to be clear that all references to Wikipedia - articles, discussions, policies, blogs, whatever - must conform to the rules for primary sources and must be clearly identified. I find it hard to disagree with Pen of Doom's comment. Even when writing about Wikipedia, using it as a primary source should generally be avoided. Below is an attempt to incorporate the warnings, toned down a bit:

ALT3. Do not use articles from Wikipedia or from websites that mirror its content as sources, because this would amount to self-reference. Similarly, do not use sources that present material originating from Wikipedia to support that same material in Wikipedia, as this would create circular sourcing.

Citing a Wikipedia article, guideline, discussion or other content from Wikipedia or a sister project to support a descriptive statement about Wikipedia itself is not a form of circular referencing. Instead, it is the use of Wikipedia as a primary source and should follow the relevant policy for the use of primary sources. Any such use should avoid original research, avoid undue emphasis on Wikipedia's role or views, and avoid inappropriate self reference. The article text should make it clear that the material is sourced from Wikipedia so the reader is made aware of the potential bias.

Aymatth2 (talk) 15:53, 25 April 2013 (UTC)

Please note that in my last message I showed how the simple change I proposed wouldn't change policy. If you think my reasoning was incorrect, please explain. If you need clarification of what I wrote, please identify what is unclear to you and I'll do what I can to communicate better with you. --Bob K31416 (talk) 18:32, 25 April 2013 (UTC)

It seems to me that Bob K31416 proposed change will be enough. -- PBS (talk) 09:08, 26 April 2013 (UTC)

• An article that discusses an aspect of Wikipedia is less likely to cite a specific article than to cite a guideline, procedure, announcement, statistic or other type of content. The current "Citing Wikipedia to support a descriptive statement about Wikipedia itself is not a form of circular referencing" applies to citing any part of Wikipedia. There is value in spelling that out to make it even more explicit. But if the policy is narrowed down to "Citing a Wikipedia article to support a descriptive statement about Wikipedia itself is not a form of circular referencing", it opens the question of how citing a non-article is viewed. Is that a circular reference? Should a non-article should be treated as a primary source? The change would needlessly open a can of worms. Aymatth2 (talk) 20:12, 25 April 2013 (UTC)

Aymatth2 When replying to another's point please stop using bullet points as it messes up the indentation and hence threads are more difficult to follow --PBS (talk) 09:08, 26 April 2013 (UTC)

@PBS: What is this "Grasshopper"? I do not understand the allusion. It is some sort of insult maybe? Please explain this. Aymatth2 (talk) 15:22, 26 April 2013 (UTC)

It's not an insult.  It's from an old American television programme called Kung Fu.  In that programme, a child is instructed by blind Master Po, a senior monk.  "Grasshopper" or "Young grasshopper" is Po's name for the child.  It's meant as whimsical and humorous rather than offensive.—S Marshall T/C 16:06, 26 April 2013 (UTC)

I'm not sure that citing a Wikipedia article (i.e., adding Cupcake to List of cakes and putting http://en.wikipedia.org/wiki/Cupcake in between ref tags on that list) is truly a "self-reference". It's bad, but I'm not sure that "self-reference" really describes it. WhatamIdoing (talk) 16:01, 26 April 2013 (UTC)

"Self reference" does seem inaccurate, based on WP:SELF. I think it applies more to the second part, the exception. "Circular sourcing" may also sometimes be inaccurate (e.g. WP article A cites dodgy source B which cites WP article C which cites reliable source D: not circular). I think the real reason not to cite Wikipedia (except as a primary source when writing about Wikipedia) is that Wikipedia is not a reliable source. Nor are sources that use Wikipedia as a source. Following is an attempt to fix this:

ALT4. Do not use articles from Wikipedia as sources. Also, do not use websites that mirror Wikipedia content or publications that rely on material from Wikipedia as sources. Content from a Wikipedia article is not considered reliable unless it is backed up by citing reliable sources. Confirm that these sources support the content, then use them directly. (There is also a risk of circular reference when using a Wikipedia article or derivative work as a source.)

An exception is allowed when Wikipedia is being discussed in the article, which may cite an article, guideline, discussion, statistic or other content from Wikipedia or a sister project to support a statement about Wikipedia. Wikipedia or the sister project is a primary source in this case, and may be used following the policy for primary sources. Any such use should avoid original research, avoid undue emphasis on Wikipedia's role or views, and avoid inappropriate self reference. The article text should make it clear that the material is sourced from Wikipedia so the reader is made aware of the potential bias.

I am not sure if this is the most tactful way of putting it. Aymatth2 (talk) 13:06, 27 April 2013 (UTC)

This proposal started as a simple suggestion that sourcing from Wikipedia (when allowable) should be clearly identified. It turned into a productive review of the entire section that showed various issues, and several editors have given valuable feedback. Despite the title, it was not obvious that the scope is Wikipedia as a whole, not just articles. The section was missing explicit warnings about undue emphasis and original research, and the wording on self-reference and circular sourcing was incorrect. ALT4 tries to address all these problems. Assuming no major objections are raised, I propose to implement ALT4 on Sunday 5 May 2013. Aymatth2 (talk) 00:25, 1 May 2013 (UTC)

ALT4 comments

RE two sentences in the first paragraph, "Content from a Wikipedia article is not considered reliable unless it is backed up by citing reliable sources. Confirm that these sources support the content, then use them directly." — This is almost entirely a repeat of the ideas expressed elsewhere in the policy, and the relation of "use them directly" to the previous statements isn't very clear.

Also, I don't see why you rewrote the part of the present policy, "Do not use articles from Wikipedia or from websites that mirror its content as sources, because this would amount to self-reference. Similarly, do not use sources that present material originating from Wikipedia to support that same material in Wikipedia, as this would create circular sourcing." Seems like you could have just as well have kept this and added your idea of suggesting the use of the sources from Wikipedia articles or its mirrors. For example, adding the underlined part in the following,

Do not use articles from Wikipedia or from websites that mirror its content as sources, because this would amount to self-reference. Similarly, do not use sources that present material originating from Wikipedia to support that same material in Wikipedia, as this would create circular sourcing. Instead, you might be able to use the same reliable sources that Wikipedia uses, without citing a Wikipedia article or one of its mirrors.

--Bob K31416 (talk) 23:48, 1 May 2013 (UTC)

I do not see where the concept of "Confirm that these sources support the content, then use them directly" is given elsewhere in the policy. It seems fairly specific to use of WP as a source: A WP article should cite a reliable source directly rather than cite another WP article that in turn cites (or does not cite) a reliable source. But if there is some overlap with other statements in the policy, that is not a problem as long as this section on "Wikipedia and sources that mirror or use it" is clear and consistent with the overall policy.

The point made by WhatamIdoing, just above ALT4, is that using content from WP or mirror sites is usually not self-referencing, a WP:MOS concern about referring to Wikipedia in the text because it may restrict re-usability of the text by other sites. Using content from WP or mirrors would only be self-referencing if the text refers to WP. That type of self-reference is recommended in the second part, the case where an article discusses WP and should clearly identify where WP is used as a primary source. As also noted, using "sources that present material originating from Wikipedia to support that same material in Wikipedia" is not necessarily circular sourcing. The scenario "WP article A cites dodgy source B which cites WP article C which cites reliable source D" is not circular.

It seems sensible to fix all the defects that have been identified in the current section during this discussion, hence ALT4. Aymatth2 (talk) 01:17, 2 May 2013 (UTC)

Let's see what others think of my comment. --Bob K31416 (talk) 01:20, 2 May 2013 (UTC)

Side discussion on OR

This sentence from ALT4 has a problem: "Any such use should avoid original research, avoid undue emphasis on Wikipedia's role or views, and avoid inappropriate self reference." But when we legitimately use Wikipedia as a primary source, chances are the material will be original research. That's what primary sources usually are, original research. Jc3s5h (talk) 01:53, 1 May 2013 (UTC)

A source can contain original research... and WP:NOR explicitly allows us to describe what a primary source says. So... it is not original research on our part to simply describe the original research found in sources. NOR is about what we say in an article... not about what our sources say. Blueboar (talk) 11:28, 1 May 2013 (UTC)

I agree with Blueboar. WP:PRIMARY says: "Do not analyze, synthesize, interpret, or evaluate material found in a primary source" - in other words, "no original research". For example, an article about Wikipedia could mention and cite Wikipedia's Verifiability policy, which is self-published original research as Jc3s5h says, as long as the article uses secondary sources for any interpretation of the policy. Aymatth2 (talk) 11:56, 1 May 2013 (UTC)

Although... this can be tricky... for example, suppose that on Monday the Wikimedia foundation says something that causes half of our Admins to resign in protest on Tuesday... we might want to mention this mass resignation in our article about the Wikimedia Foundation, saying: "In protest over the WMF's actions, half of Wikipedia's Administrators resigned." . We would need a citation to support the statement. So lets look at two possible citations:

1. a user page containing a user generated list of Admins who resigned on Tuesday...
2. a group letter posted on the same user page - jointly signed by all the resigning Admins, explaining how the WMF's actions will harm WP and explicitly stating that they are resigning because of this.

Both citations contain OR... in citation 1 the OR is the compilation of Admins who resigned, in citation 2 the OR is the analysis of the harm the WMF's actions will cause.

Now to examine both in light of the WP:V policy...

Citation 1 would be an improper citation under WP:V. The creator of the list is assuming that all of the resigning Admins he lists resigned due to the WMF's actions. It may be a logical assuption, but it an assumption never the less. It does not verify the statement because at least some of the admins listed might have resigned for other reasons.

Citation 2 (the letter) would not be an improper citation under WP:V ... the letter directly supports the statement that the Admins who signed it resigned due to the WMF's actions. Yes, the analysis of harm is Original Research... however that analysis is not what is being cited... what is being cited is the number of Admins who signed the letter and stated that they are resigning due to the WMF's actions. Their analysis is simply background information... their opinion on why their resignation is necessary. We can even describe that analysis... as long as we make it clear that it is opinion and not necessarily fact.

All that said... I would suggest NOT mentioning any of this the section. Leave tricky OR issues for the NOR policy page. Blueboar (talk) 14:36, 1 May 2013 (UTC)

It seems useful to leave the basic "avoid original research" statement here as a pointer to the WP:NOR policy, which covers the topic in depth. The warning applies to all articles, but perhaps the temptation is greater when using a primary source. In the example above, I agree that it would be hard to cite #1 in context of an article about mass resignations without either implying they were all due to the WMF action, or else stating that some may not have been, both of which are OR. Aymatth2 (talk) 15:37, 1 May 2013 (UTC)

The discussion above is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.

Magazines as reliable sources

I am not sure whether this should be in WP:Verifiablity or in WP:Reliable Sources, but should there be a guideline that states that supermarket tabloid magazines are not considered reliable sources for information about celebrities? Robert McClenon (talk) 17:22, 4 May 2013 (UTC)

Unreliable magazines

I am not sure whether this should be in WP:Verifiablity or in WP:Reliable Sources, but should there be a guideline that states that supermarket tabloid magazines are not considered reliable sources for information about celebrities? That could go in the section on questionable sources. Robert McClenon (talk) 17:23, 4 May 2013 (UTC)

It already explicitly says it " Such sources include ... [those] that rely heavily on rumor". That's good enough to keep such tabloids from being used as sources. --Jayron32 17:26, 4 May 2013 (UTC)

Added as an example.

It's been reverted. Are you actually having a pervasive problem with this? I don't think that's common, but I tend not to frequent celebrity articles, so perhaps I'm just unaware of the scale of the problem. WhatamIdoing (talk) 20:08, 4 May 2013 (UTC)

Proposed change of wording to {{unreferenced}}

Please see the discussion at Template talk:Unreferenced#Inline citations to reliable sources some additional input by other editors should help us reach a consensus on whether to change the wording of {{unreferenced}}. -- PBS (talk) 10:00, 30 April 2013 (UTC)

The question is whether a substub containing purely obvious material and looking like this:

A bone fracture is a broken place in a bone.

References
• Expert, Alice (2012) "What is a Broken Bone?" J Imp Medicine

is entirely "unreferenced" or merely lacking in inline citations. If there are no little blue numbers, is it possible for references to still be listed on the page? WhatamIdoing (talk) 15:58, 1 May 2013 (UTC)

The actual question is should the wording in the second sentence of the template be changed from "adding citations to reliable sources" to "adding inline citations to reliable sources"? -- PBS (talk) 09:01, 4 May 2013 (UTC)

I would say the answer to PBS's last question is: no... newer editors may not yet know how to format inline citations... but we still want them to provide reliable sources to support the article. If they don't do it "properly", someone more experienced can follow up and fix it so its in line. So... the template should simply ask for reliable sources (without specifying "in line"). Blueboar (talk) 16:33, 4 May 2013 (UTC)

On a related note, perhaps it would be helpful to add a sentence to WP:V that says something like "General references are permitted, although they are not sufficient for the four types of material that are specifically required to be supported by inline citations". WhatamIdoing (talk) 20:06, 4 May 2013 (UTC)

I do not think it would be helpful. -- PBS (talk) 13:40, 10 May 2013 (UTC)

I'm not surprised, since you keep telling people that general references are not acceptable. But does anyone else have any opinions about this? For example, Early life of Joseph Smith is a Featured Article that contains some general references (e.g., Grant Palmer's book). These are sources that were used in the work of building the article but which are not cited in support of any particular sentence. Should listing these as references be explicitly permitted, or should we remain silent on the point, with the usual risk that some editors will interpret that silence as proof that general references are not acceptable and "depreciated (sic) for a number of years"? WhatamIdoing (talk) 00:54, 14 May 2013 (UTC)

Non-English quotations and translations

Wikipedia:Verifiability/Archive 60#Non-English sources indicates that original, non-professional translations of non-English sources are acceptable when no reliable source has one. For translated quotations, should the translator be credited? And must the original-language quotation be included as well? —Frungi (talk) 19:27, 16 May 2013 (UTC)

The usual approach is this: people quote the original and provide the translation, and do not name the editor (usually themselves) who created the translation. You may omit the original-language quotation, and you may add the Wikipedian's name to the citation, but it is not normally done that way. WhatamIdoing (talk) 21:03, 16 May 2013 (UTC)

If there is a published translation, I assume that it, rather than the original, should be listed as the (English) source, and both the original author and the translator are credited. (Or is that too obvious?) Robert McClenon (talk) 21:40, 16 May 2013 (UTC)

Right, but I was asking about when there isn’t a published translation, and an editor takes it upon itself to translate a quote for Wikipedia. —Frungi (talk) 02:08, 17 May 2013 (UTC)

All right, thanks. The page isn’t clear on those points. —Frungi (talk) 02:08, 17 May 2013 (UTC)

And now it is. —Frungi (talk) 20:56, 17 May 2013 (UTC)

It's a content guideline

S. Marshall has reverted the restoration of an older version of one sentence, namely the change from

Articles should be based on reliable, third-party, published sources with a reputation for fact-checking and accuracy.

to the more recent

Base articles on reliable, third-party, published sources with a reputation for fact-checking and accuracy.

I thought we had achieved consensus to restore the earlier version some time ago. Apparently not, though, as the reversion never actually occurred. This guideline is a guideline about content, not behaviour. It needs to address what the content of an article should be. It shouldn't be issuing directives to the writer, it should be describing what an article needs to contain.—Kww(talk) 19:35, 21 May 2013 (UTC)

• The dispute is about mood: the extent (if any) to which the policy should be phrased in the active imperative ("always do this, never do that, and consider doing the other") and the active indicative ("this is this, that's that and the other is the other") -vs- the extent to which it should be phrased in the passive hortative ("this should always be done, that should never be done, and editors should consider whether to do the other").

  My position is that although policies apply to content, they're meant to be read by people. They should be phrased in simple, accessible language that's directly applicable to cases.—S Marshall T/C 19:42, 21 May 2013 (UTC)

• If someone's language skills are so limited that they are baffled or confused by "Articles should be based ...", he is below the minimum competence level to edit Wikipedia.—Kww(talk) 19:59, 21 May 2013 (UTC)

I support S Marshall, at least in his conclusions, on this one. This is one of the most basic and important tenets at WP. "Should" makes it discretionary and it needs to be mandatory. That's one reason this is a policy and not a guideline. Regards, TransporterMan (TALK) 20:43, 21 May 2013 (UTC)

Agree. If we were using the longer imperative we would say, "Articles shall be based . . .", but Smarshall has the better, simpler mood. Alanscottwalker (talk) 21:02, 21 May 2013 (UTC)

I'd be thrilled with "Articles must be based ...". I don't think the question of whether something is mandatory should be muddled up with whether the guideline should address behaviour or content.—Kww(talk) 21:08, 21 May 2013 (UTC)

I lean towards Kww's take on this... the tone should focus on the content not behavior. However, I don't feel strongly about it where this particular sentence is concerned. Blueboar (talk) 21:13, 21 May 2013 (UTC)

I'd have additional objections to "articles must be based..." because of the risk that it'd be taken as a licence for idiots to go on an AfD-spree of older content that's verifiable but not verified. "Base articles..." is an instruction applying to new content being written today.—S Marshall T/C 21:24, 21 May 2013 (UTC)

That's why I think the two issues shouldn't be coupled. I disagree with you on both counts, but I think two reasonable people could agree on one and disagree on the other.—Kww(talk) 21:36, 21 May 2013 (UTC)

I share S Marshall's concern about the potential for misunderstanding here. We don't require that articles demonstrate what they were based on, and we don't want people to believe that unref'd pages are automatically in violation of this policy.

Actually, we don't technically require that editors actually write most of the content from these good sources, because that's unenforceable. What we really require is that the articles match what would have been written if the editors had used good sources. An example might make that tortured sentence clear: If I actually read a personal blog and write an article that says "Alice Expert is a notable professor who invented warm ice and cold sunshine. Alice and Bob have a shared secret.", then I've not actually used good sources, but that's okay—if (and only if) using good sources would have produced the same article. If the product is indistinguishable, then we don't get hung up on the process per WP:PPP. WhatamIdoing (talk) 22:46, 21 May 2013 (UTC)

No, what is required is that the article use either general references or inline citations. You've taken the notion that we don't immediately delete articles for not having them and expanded it into some kind of notion that they are inherently acceptable. Certainly some bad articles can be repaired, that doesn't mean that they aren't bad articles. I really don't understand why you campaign so relentlessly for the retention of unsourced material.—Kww(talk) 23:03, 21 May 2013 (UTC)

Kww, that's wrong. Citations are only required if the article contains quotations, close paraphrases, or material that is likely to be challenged. Jc3s5h (talk) 23:18, 21 May 2013 (UTC)

Such articles still require at least a general reference. I agree that inline citations are only best practice, but completely unsourced articles aren't acceptable under any reasonable reading of policy.—Kww(talk) 23:21, 21 May 2013 (UTC)

What's unacceptable is articles that are unsourced and unsourceable. Nobody reasonable wants to keep those. As far as I can see the dividing line is over content that's unsourced but sourceable: can it remain in the mainspace until the sourcing chores are done? WAID's position (and mine) is that it can.—S Marshall T/C 23:29, 21 May 2013 (UTC)

I've never interpreted WAID's comments that way, and if that's what she means, that's what she should say. Instead, she campaigns for the patently absurd notion that articles need not give any clue as to where the material came from so long as a diligent researcher could eventually uncover the same material.—Kww(talk) 23:37, 21 May 2013 (UTC)

I see no practical difference between the two. An article whose contents are actually verifiable, but which is currently unsourced, is an article that does not give any clue where the material came from. It is my position that an article that somehow manages to be entirely composed of material that doesn't cross the lines described at WP:MINREF does not require any sources to be cited at all. You may describe this as "leaving an unsourced but sourceable article in the mainspace until the sourcing chores are done" or you may describe this as "articles need not give any clue as to where the material came from so long as a diligent researcher could eventually uncover the same material", but it's the same thing in practice. WhatamIdoing (talk) 02:28, 22 May 2013 (UTC)

The primary difference is whether you view an editor that frequently includes material without sources to be disruptive or constructive. I consider such editing to be readily excusable for a new editor but consider any experienced editor that produces such material to be either incompetent or intentionally disruptive. If you view the content as acceptable, it generally follows that you don't find producing such content to be a problem.—Kww(talk) 02:58, 22 May 2013 (UTC)

I thought I was fairly careful to draw a dividing line between new content produced now, under the current rules, and content written historically when the rules were different. Personally, I would view someone in 2013 who adds articles that are totally unsourced, as someone in need of guidance. But a decade ago the rules were different, and some of this legacy content is valuable. I wouldn't want policy to grant an implied licence to delete it all.—S Marshall T/C 10:14, 22 May 2013 (UTC)

That's why my disagreement with you is relatively small. I don't disagree with tolerating older material, my primary objection is that retaining it based on " ooh ooh ooh I did a google search and the title came up zillions of times" isn't enough to keep it. It think that participants at AFD (including the closing admin) have to do enough due diligence to be able to make a good faith assessment that the content of the article is, indeed, supported by those zillions of references and that a natural consequence of doing that due diligence is adding at least one of those zillions and zillions of references to the article. WhatamIdoing has specifically stated that no such examination is required, and I'm unaware of any statement that she views it as a way to tolerate older material: she simply states that sourcing isn't required.—Kww(talk) 22:22, 22 May 2013 (UTC)

I think we should stick with "should" rather than "must" because "third-party" is a concept from contract law which cannot be rigorously defined when it comes to deciding upon sources for an encyclopedia article. Jc3s5h (talk) 21:38, 21 May 2013 (UTC)

While "third party" is a term used in contract law... it is not a term exclusive to contract law. We are not using it in its legal sense. Blueboar (talk) 21:47, 21 May 2013 (UTC)

One of these days, I'll actually merge WP:Third-party sources into WP:INDY. There is a small distinction between the two, but it's not important for Wikipedia. In the meantime, we often say "third-party, independent source" because some people understand one concept or the other, and using both increases the odds that they'll understand. If we were going to simplify, then I'd pick "independent". WhatamIdoing (talk) 22:46, 21 May 2013 (UTC)

Not sure if this will fly, but it sometimes helps to think "outside the box" for a moment ... is there a way to completely reword the phrase, to avoid the entire debate? Perhaps: "Wikipeida articles are supposed to be based on..." (or something along those lines). Just a thought. Blueboar (talk) 21:59, 21 May 2013 (UTC)

• I agree that this is primarily a content guideline, not a behavioural one. Reyk _YO! 23:35, 21 May 2013 (UTC)