Jump to content

Wikipedia talk:Protection policy

Page contents not supported in other languages.
Page semi-protected
From Wikipedia, the free encyclopedia
(Redirected from Wikipedia talk:SEMI)

Can we add Signpost articles as an exception to WP:PREEMPTIVE??

Pre-emptively protecting Signpost articles (to confirmed/autoconfirmed) makes a lot of sense to me because they don't need to be updated post-publication and never really need to be edited except by bots/scripts.

Pinging @Frostly: because they may or may not have an opinion on this topic. Polygnotus (talk) 18:59, 18 October 2024 (UTC)[reply]

Please provide several examples of signpost pages that were edited unjustifiably, in order to demonstrate that a problem exists. --Redrose64 🌹 (talk) 20:33, 18 October 2024 (UTC) amended Redrose64 🌹 (talk) 09:13, 19 October 2024 (UTC)[reply]
On a related note, the current Signpost editor-in-chief has already confirmed/autoconfirmed protected a lot of the past issues, after they were granted administrative privileges. isaacl (talk) 22:08, 18 October 2024 (UTC)[reply]
I'll second Redrose64's request for some examples. We really need to know how often it's a problem before we consider changes to the policy. And even if there's consensus that preemptive protection makes sense, I would want the protections to be automated or done by the Signpost team rather than adding load to WP:RFPP. Daniel Quinlan (talk) 02:03, 19 October 2024 (UTC)[reply]
An example or two wouldn't tell you how often it's a problem. Polygnotus (talk) 08:34, 19 October 2024 (UTC)[reply]
No, which is why I asked for examples (which is open-ended), and not "an example or two" (which is closed). I've inserted the word "several" to clarify this. The more examples that you can provide, the easier it will be for us to observe that a problem exists that needs to be fixed. If it's happening on a daily basis, we probably do need to take action; but if it's only once a year, it should be quite sufficient to ask the person who made the edit to explain it; if not, a simple revert will do.
A general principle, that should be followed by anybody seeking to create a new rule or amend an existing rule, is: can you justify that new rule or amendment in a non-hypothetical form? In other words:
  1. demonstrate that a problem exists
  2. show that existing processes are insufficient to overcome the problem
  3. propose a method that will either solve the problem, or prevent it from re-occurring
  4. invite comments and be prepared to defend your proposal
  5. if sufficient parties are in agreement, implement the proposal
You've come up with (3) and (4), but not (1) or (2). I'm just asking for (1) at this stage. --Redrose64 🌹 (talk) 09:13, 19 October 2024 (UTC)[reply]
But its not a new rule, it is exception to an existing rule. And the idea is not to create one but to mention it. Polygnotus (talk) 09:29, 19 October 2024 (UTC)[reply]
New, exception or amendment, the principle is the same: don't mess with the system just because you want to. It applies right to the top - you don't get governments saying "I know, let's pass a law that says that you can't step on the cracks in the pavement", it sounds like they want to control people's lives for the sheer hell of it. No, first they identify the cracks in the pavement as an existing problem (people have been tripping up leading to injury); second, they show that existing processes (sending somebody with cement and a trowel) are insufficient (there is no more cement in the stores); third, they propose a new method (tell everybody not to step on the cracks); fourth, they invite discussion (somebody suggests buying more cement); fifth, they agree that the amended proposal is the better idea even though it costs more. Result: people don't trip up, and feel better about the lawmakers even though their taxes are slightly higher. --Redrose64 🌹 (talk) 10:02, 19 October 2024 (UTC)[reply]
Since you seem to be very passionate about this rather boring topic, do you have any reasons? Are there any downsides I am not aware of? Do you think new users are likely to try to make improvements to Signpost articles that would be prevented by protection, which would force them to use a template on the associated talkpage, which they won't do for which reasons? I'm not sure I understand that scenario. Oh, and can you provide a list of examples please. I'll tell you when you have enough of them (I won't specify a number but its more than 2). Polygnotus (talk) 10:17, 19 October 2024 (UTC)[reply]
If you think it's a boring topic, why did you raise it in the first place? Do you have any reasons other than they don't need to be updated post-publication and never really need to be edited except by bots/scripts? If I seem passionate, it's because I am trying to explain why I am not going to approve a policy change on the request of one person, especially when that person refuses to show why the policy as it stands requires any change.
As for Oh, and can you provide a list of examples please. I'll tell you when you have enough of them (I won't specify a number but its more than 2). - are you joking? Do you really want me to provide examples of signpost pages not being edited because nobody felt like vandalising them that day?
I am not the one that is seeking change to an existing policy, it's you. You need to justify your proposal, otherwise it's change for the sake of change; so, you need to explain exactly why. If you put forth reasons why this must be done, I (and others) can consider each point carefully, and offer reasons why that point doesn't stand up. Therefore, it's not my duty to provide examples - it's yours. --Redrose64 🌹 (talk) 14:02, 19 October 2024 (UTC)[reply]
I was making a point... As a volunteer here, it is not my duty to do anything. This is why we must fork Wikipedia so we can start over with little to no policy and make the same mistakes all over again. It is a cycle all long term online communities go through, from wild west to Kafkaesque (which rhymes if you are a rapper). Do you need citations for that? Polygnotus (talk) 16:26, 19 October 2024 (UTC)[reply]
Polygnotus, it's standard to request evidence of a problem before changing policy, and Redrose64 has been respectful in doing that. If you're not willing to provide any evidence, then it's unlikely the policy will change.
Also, it's one of Wikipedia's core principles to interact with one another in a polite and respectful manner and you're not doing that here. Your last comment is so uncivil that I suggest you reexamine what you've said and consider whether you want to strike it or apologize. Daniel Quinlan (talk) 17:29, 19 October 2024 (UTC)[reply]
@Daniel Quinlan: Are you familiar with Graham's Hierarchy of Disagreement? Your last comment is at the second lowest tier. I suggest you apologize and strike your comment. In future, try refuting the points people make without attacking them. Polygnotus (talk) 22:01, 19 October 2024 (UTC)[reply]
Let's look at the OP's original justification for protection (they don't need to be updated post-publication and never really need to be edited except by bots/scripts), since no evidence for it has been provided. These are the articles in a recent issue of the Signpost:
The OP's original premise appears to be faulty. – Jonesey95 (talk) 13:03, 21 October 2024 (UTC)[reply]
See Straw man. Which of those who made good edits are not confirmed/autoconfirmed? None in the list above... You can't just isolate a part of a sentence in a way that distorts its meaning (because of the lack of context) and then pretend you refuted anything. Polygnotus (talk) 09:42, 24 October 2024 (UTC)[reply]
Isolate a part of a sentence? I literally copied and pasted every part of the OP's rationale that followed the word "because". See WP:REFUSINGTOPROVIDEEXAMPLESEVENWHENASKEDNICELY. The OP has not provided any examples of the need for protection of these pages. I linked to a lot of productive edits, any of which could reasonably have been made by new and helpful editors. – Jonesey95 (talk) 15:08, 25 October 2024 (UTC)[reply]

Icon redirect

The BLACKLOCK icon at Asian News International vs. Wikimedia Foundation redirects to Wikipedia:Protection_policy#full, but Wikipedia:Protection_policy#Office_actions would be more on point. Gråbergs Gråa Sång (talk) 11:50, 21 October 2024 (UTC)[reply]

@Gråbergs Gråa Sång: I believe you are right, but the policy isn't the place to fix this. This needs to be raised as an issue at Module talk:Protection banner. Regards. Daniel Quinlan (talk) 18:37, 21 October 2024 (UTC)[reply]
Thanks, I thought someone here would know where (or even better, how, and do it). I'll move over there. Gråbergs Gråa Sång (talk) 18:40, 21 October 2024 (UTC)[reply]
That module is a bit convoluted so I probably should have qualified my answer. Anyhow, if that doesn't end up being the right place, you could also ask on WP:VPT. Daniel Quinlan (talk) 18:45, 21 October 2024 (UTC)[reply]

Operational pages

I propose updating the policy to explicitly cover the protection of operational pages used by bots and user scripts. While many of these pages are already protected, it would be better to include something in the policy. User:MDanielsBot/AIVStop is the most recent target of a disruptive edit, but this has also happened in the last year with User:Lowercase sigmabot III/Shutoff, User:ClueBot NG/AngryOptin, User:DatBot/Filter reporter/Run, User:GreenC bot/button, User:InternetArchiveBot/Dead-links, and User:Yapperbot/kill/FRS. If there is consensus, I would like to add a Protection of operational pages section under the Uncommon protections section as follows:

Operational pages used by software, including bots and user scripts, may be protected based on the type of use, content, and other considerations. This includes, but is not limited to, configuration pages, data pages, log pages, and status pages. However, personal CSS, personal JavaScript, and personal JSON are automatically protected and should not be protected for this reason.

Note that the intrapage links will omit the page name if this is added to the policy. Daniel Quinlan (talk) 00:01, 10 November 2024 (UTC)[reply]

I updated the proposed text slightly to remove Similar to templates which is unnecessary and I also updated the proposed location since it doesn't really fit in the Protection by namespace section. Most of the protections for this reason are in User space, but some are in Wikipedia space, Module space, and Template space. Daniel Quinlan (talk) 22:50, 18 November 2024 (UTC)[reply]
I went ahead and added the section with some minor rewording, and with one significant change: I added principally as an additional restriction to ensure this doesn't extend to cases where software happens to use a page (e.g., protecting Wikipedia:Arbitration Committee because a user script extracts the list of ArbCom members would not be covered).
If anyone has comments or concerns about this change to encode common practice in policy, please let me know. Thanks. Daniel Quinlan (talk) 20:12, 22 November 2024 (UTC)[reply]

Should some templates be fully protected?

Moved from Wikipedia talk:High-risk templates

Are there any "very" high risk templates or modules which need full protection or is template protection adequate? The guideline Wikipedia:High-risk templates is not clear on this matter, and there is an ongoing discussion about the protection of Module:WikiProject banner — Martin (MSGJ · talk) 20:52, 6 December 2024 (UTC)[reply]

Of course there are, and they mostly already are. Ones used on 10's of millions of pages which could cause severe disruption are good candidates, as are many that are part of the system interface. The FPROT request queue is seldom backlogged and serves as an effective check against protection. The policy (Wikipedia:Protection_policy#High-risk pages and templates) already makes allowances for this. If guideline text is outdated, it's a wiki..... — xaosflux Talk 21:24, 6 December 2024 (UTC)[reply]
In other policy, Wikipedia:Template editor makes passing mention of full protection. It's probably also worth re-reading the RfC, which talks about 'temporary', 'extraordinary', 'thousands of transclusions', and other things. -- zzuuzz (talk) 21:35, 6 December 2024 (UTC)[reply]
I think the 2013 RfC highlights the argument in favor of allowing template editors to edit many or all of the FPROT templates:
While full protection is an ideal temporary solution for articles that have demonstrated a state of overwhelming controversy, it is less ideal as a permanent precautionary measure for templates. Many editors who have shown an aptitude for coding templates, and have earned the trust of the community in doing their work, may not necessarily be administrators, nor even be interested in becoming administrators.
Non-administrators do have the ability to request edits at fully-protected templates for administrators to enact on their behalf, but there is a significant shortage of administrators who have the time and necessary skills to do this reliably. Coders also tend to find this extra step more than a mere annoyance: Technical work is largely rewarding to technically-minded people in that they value the hands-on experience. Many end up choosing to avoid having to verbalize uncontroversial edit requests made to convince someone else to enact an edit on their behalf, by simply avoiding work on fully-protected templates altogether.
I think WP will be better off letting template editors work on FPROT templates. If there are template editors who have lost the trust of the community, the right answer is to remove their template editor rights, rather than FPROT the templates they work on. (Bias alert: I am an template editor). — hike395 (talk) 06:36, 8 December 2024 (UTC)[reply]
I am of two minds on this one. I am a very experienced template editor, and I mostly know my limits. When I have doubts about my ability to successfully implement an edit, I test it in the sandbox, ask on the talk page or VPT, or both. Every once in a while, the Dunning-Kruger effect kicks in, and I do not have doubts but get something wrong. If I see the problem, I either quickly revert or quickly fix the problem. About once a year, I am dunned by other template editors and admins for such behavior, but nothing ventured, nothing gained, I figure. Template editors fix a lot of stuff around here.
A few times per year, I encounter a fully protected template or other transcluded page that I can't edit, so I put the edit in the sandbox and put in an edit request on the template's page, with a full explanation of the esoteric change that I am requesting. Admins have always gotten to these requests within a couple of days, IIRC. About half the time, a sensible admin will lower the page's protection to template protection because it no longer meets the FP requirements.
About the same number of times per year, I encounter some nice-to-have fix that I want to make to an FP page, but I don't bother with an edit request because it seems trivial. That, to me, is the only downside to denying FP template editing to template editors. Overall, I'd say that I can live with this inconvenience for the tradeoff of protecting these pages against template editors who are less careful than I am; there have been a few of them over the years. – Jonesey95 (talk) 04:52, 9 December 2024 (UTC)[reply]
Pulling a random number out of the air, but anything with >1 mil transclusions probably should be FPROT, if only as a final check to make sure folks are really paying attention to what they are doing. Raising the protection on a template because one template editor made mistakes one time seems a bit overkill, and a word to said template editor would probably be more effective, especially as it would leave a (digital) paper trail for if those sorts of things were a regular occurrence. The above being said, I have no issue with having TPER on pages with >1mil transclusions if they're fairly static or there's a TPE that has demonstrated they can update the template properly (i.e. sandboxing first, etc). Primefac (talk) 15:14, 9 December 2024 (UTC)[reply]
According to Wikipedia:Database reports/Templates transcluded on the most pages, there are just 198 template and module pages with 999K or more transclusions. Looking at the list, I see a lot of low-level, sometimes intricately used, tools that I (a template editor) wouldn't dare mess with without a discussion, along with some pretty simple templates that should not need frequent maintenance. Heck, I even created one of them, {{Short description/lowercasecheck}}, but now that it is used in six million pages, I wouldn't like to see template editors monkeying with it casually. I don't see a problem with one million as a threshold for automatic FP. – Jonesey95 (talk) 06:23, 12 December 2024 (UTC)[reply]
Under WP:FULL, the protection policy includes Pages that are transcluded very frequently on the list of pages that are usually fully protected for an indefinite period of time. That seems pretty clear. It might be worth having a report somewhere for pages with more than one million transclusions that aren't fully protected, though. Daniel Quinlan (talk) 08:14, 12 December 2024 (UTC)[reply]
Retrieved from "https://en.wikipedia.org/w/index.php?title=Wikipedia_talk:Protection_policy&oldid=1262614277"