Wikipedia talk:Disambiguation/Archive 32

This is an archive of past discussions on Wikipedia:Disambiguation. Do not edit the contents of this page. If you wish to start a new discussion or revive an old one, please do so on the current talk page.

Archive 25

←

Archive 30

Redirects to disambiguation pages being tagged for deletion

A particular user, apparently unhappy about the bot that tags intentional links to disambiguation pages in hatnotes, has started tagging "foo (disambiguation)"-type redirects for deletion, such as Delirious and The Choir. The same user has been changing other intentional dab links to bypass the redirects. I have tried to reason with this person on his talk page, but so far unsuccessfully. Other interested editors may wish to monitor these pages to make sure that not too much harm is done, or to try to be more persuasive than I was able to be. --R'n'B (call me Russ) 17:49, 19 October 2010 (UTC)

Delirious (disambiguation) is now up for discussion at RfD. Hopefully concluding a discussion there will settle this nonsense. bd2412 T 19:31, 19 October 2010 (UTC)

The nominations for this and The Choir (disambiguation) were incomplete, and I have now completed them so that we can proceed to discussion at Wikipedia:Redirects for discussion/Log/2010 October 19. Cheers! bd2412 T 19:44, 19 October 2010 (UTC)

From Mr. Görlitz: "NO I'm sane. You are the ones who are not sane." RARR!! (Emphasis mine.) Actually I'm surprised we haven't seen more of this. Back when I was working for the Orphanage, phew, they came a-runnin' with torches and pitchforks... --JaGa^talk 19:46, 19 October 2010 (UTC)

The policy is stupid, at least in these two cases. It should be changed. I will withdraw the request and suggest that you change the policy immediately and make it rational. --Walter Görlitz (talk) 19:51, 19 October 2010 (UTC)

Please fix your bot as well. The hatnote does not need to point to the disambiguation redirector if another disambiguation page exists. I will not be coming back to discuss this. It's my opinion. I will also ignore and remove any comments related to this insane policy on my talk page. Cheers. --Walter Görlitz (talk) 19:55, 19 October 2010 (UTC)

I agree that it is your opinion. --R'n'B (call me Russ) 20:54, 19 October 2010 (UTC)

Since this does crop up from time to time, perhaps we need to expand and clarify the explanation at WP:INTDABLINK, to make it clear why this is necessary and how it helps. bd2412 T 06:02, 20 October 2010 (UTC)

Good idea. Every once in a while I get some arguments from "policy lawyers" who try to convince me that the policy doesn't cover hatnotes, or some such thing. We need to have it spelled out so clearly that no one could possibly believe it doesn't apply, as well as providing the reasons behind it for those who don't see how it is helpful. --R'n'B (call me Russ) 10:25, 20 October 2010 (UTC)

Primary topic and traffic stats

The traffic statistics are typically the major fact used in primary topic discussions at present, almost to the point of giving it overwhelming weight and ignoring other factors entirely at times. However, the guideline does not indicate this and says that the stats are one of a few measures that should be borne in mind to aid discussion.

This strikes me as a break between the guidance and the current implementation of it: Either the current interpretation of the guideline is drifting away from the community-derived consensus and people are slightly distorting the exact meaning of WP:PRIMARYTOPIC or the consensus has changed and its the guideline that is outdated. If the latter is the case, the guideline should be updated to reflect this. In particular, more detailed information on the sort of figures a primary topic should be obtaining should also be written into the guidance IMO.--Nilf anion (talk) 08:40, 23 October 2010 (UTC)

I'm typically use traffic stats as a major factor in determining a primary topic, but I wouldn't want the guideline changed to put them as the sole determinant. Traffic statistics have the benefits of being easy to determine (thanks to the tools at stats.grok.se) and being good (but not perfect) predictors of the other factors. Most primary topic determinations do continue to follow the guidelines, so I don't think they need to be updated. Do you have in mind a case where the traffic stats were used and the other measurements disagreed with its results but were ignored? -- JHunterJ (talk) 14:13, 23 October 2010 (UTC)

I certainly agree that traffic stats shouldn't be the sole determinant, but at same time I get impression that they are considered more valuable than the other two listed terms (or the "importance" concept the guideline fails to pin down and describe effectively). The ease of generation is partially the cause of this, but probably also because they are genuinely the most powerful: the test redirects with Lincoln and EA are examples of this. The guideline doesn't really make clear the fact that the stats are seen as more valuable by a significant portion^{[citation needed]} of the community that participate in these discussions.

In any case, providing figures on the lines "If an article has over 70% of the traffic it is probably the primary topic" and "If the most popular article has under 30% of the traffic there is unlikely to be a primary topic" (or whatever past experience indicates are clear-cut numbers) would benefit those not that familiar with how to apply the primary topic guidelines. Providing number for this - and indeed for the other two measures wouldn't hurt.

I can't immediately provide examples where the traffic and the others disagreed, but can certainly provide examples where the others have not been investigated despite the discussion being contentious (eg Talk:Cambridge#Requested move). (As an aside: I've found AWB to be a simple way to generate inward link counts, easier and more powerful than going from the special page).--Nilf anion (talk) 19:21, 23 October 2010 (UTC)

If no one (on any "side") is bothering with the other measurements, I'm not sure what we should do. The other measurements are still available and useful, but if everyone in a particular discussion is happy using just the traffic stats, that will work too.

As far as clear-cut numbers, I don't know what they'd be. If there are 50 ambiguous topics for a title, one gets 51% of the traffic and each of the others gets 1%, I'd say it was the primary topic, but if there are 2 ambiguous topics, one gets 51% and the other gets 49%, I would say there's no primary topic. That's why the current language (much more than any other, and more than all the others combined) is there. -- JHunterJ (talk) 21:05, 23 October 2010 (UTC)

I think the problem is the pro-dab "side" has the traffic stats in its favour - so they won't look further; why bother? Whereas the anti-dab side aren't familiar with the guidance and will think about what comes more naturally (that is their perception of importance). The "importance" of a topic is relevant to primary topic resolution but is much harder to gauge due to obvious subjectivity.

And yeah, good point on clear-cut numbers it depends on how many articles there are, how many are significant etc etc. Incidentally this discussion is one where traffic stats (reasonably ambiguous) and incoming links (clear "winner") do indicate completely different things.--Nilf anion (talk) 00:15, 24 October 2010 (UTC)

Can I just ask does the guideline, with regards to traffic stats, state that a primary topic requires clear majority (over half the total traffic), or a clear plurality (a lot more than the next greatest)?

The guideline does say "much more likely than any other, and more likely than all the others combined". Appearing that either the "much more likely than any other" line is superfluous or that the "more likely than all the others combined" line is erroneous when counted with the previous. Zangar (talk) 17:48, 25 October 2010 (UTC)

An example. Say we were able to get the exact usage statistics coming out of a dab page. In one case:

Header text	Article
A1	51%
A2	49%

and

Article	Percentage
B1	51%
B2	7%
B3	7%
B4	7%
B5	7%
B6	7%
B7	7%
B8	7%

As HJH says, we generally have a dab page for A and not for B. A majority and a significant advantage over any individual article is generally required. NOTE: Some insist the primary should be significantly greater than 50% of the total outlink hits. --John (User:Jwy/talk) 18:01, 25 October 2010 (UTC)

(edit conflict)As illustrated by JHunterJ above, if one topic has 51% and all the other 50 topics have 1% each, then not only is "more likely than all the others combined" true, but so is "much more likely than any other", and, so, the 51% is a primary topic. But if there are only two topics (51% vs 49%) then "much more likely than any other" is not true, so it's not the primary topic.

Further, if one topic has 40% but 30 others have 2% each, then "more likely than all the others combined" is not true and so the 40% one is not the primary topic even though "much more likely than any other" is true. Both phrases are needed. Neither is erroneous nor superfluous. --Born2cycle (talk) 18:02, 25 October 2010 (UTC)

Cheers for clarification on that Zangar (talk) 23:40, 25 October 2010 (UTC)

Getting back to the original question. We don't have an easy way to gather the statistics used in the examples here. In special cases, we have set up temporary redirects coming out of a dab page to understand the "real" usage of the dab links, but that is rare. So, in part, the other "non-statistical" tools are used to help approximate the usage. Personally, I believe that is pretty much ALL we should use them for, but I know there are those that disagree. If we could get these percentage statistics, I believe we would have a very effective, useful, understandable and efficient implementation of primary topic matters using a formula similar to discussion above - adjusting the required interval between primary and second choice, and setting an agreed upon level of majority (51%? 60%?). I'm not suggesting we do this - we can't without significant Media Wiki changes and performance issues. But (see the next section) I think discussion of how we would proceed if we COULD do that would be useful in fleshing out what other factors others believe we should use in primary topic determination. --John (User:Jwy/talk) 18:30, 25 October 2010 (UTC)

To further emphasize what others have said above, traffic view statistics are only one tool used to help judge which topic is most likely (or more likely) to be viewed by Wikipedia readers, and in no case should be considered the sole determining factor. --R'n'B (call me Russ) 18:37, 25 October 2010 (UTC)

I'm not suggesting that we solely use current statistics. I'm trying to understand your position further. If we had precise statistics that told us how people navigated out of dab pages, what more should we consider? Such a statistic should come quite close to giving us information about which topic is most likely to be viewed by readers. What would missing? That's what I'm asking below --John (User:Jwy/talk) 18:55, 25 October 2010 (UTC)

There are many cases where they can be and should be the sole determining factor. If their the only factor brought forward, there's no reason for them not to use them. There are too many discussions where the traffic stats are attacked but no other appropriate rationale (just anecdotes, origins, and age, usually) is used to contradict them. In those cases, the stats should be the determinant. -- JHunterJ (talk) 23:06, 25 October 2010 (UTC)

Total confusion

Please see this diff, where the hatnote links to the DAB page, but displays the undisambiguated title. Is this correct? Why or why not? I have no clue. - BilCat (talk) 16:46, 27 October 2010 (UTC)

Its as intended. Its for maintenance of dab pages: By going through the (disambiguation) redirect, someone looking at http://en.wikipedia.org/wiki/Special:WhatLinksHere/Brave can tell that the link from Atlanta Braves is intentional - needing no correction. Most links to a DAB page are un-intentional and should be piped to point directly to the appropriate article. By going through the redirect, we know that link is okay and need not navigate to the page to figure that out by context. --John (User:Jwy/talk) 17:03, 27 October 2010 (UTC)

That's far too confusing for me to try to understand while sober. And I don't drink. - BilCat (talk) 17:14, 27 October 2010 (UTC)

Yeah, I was slow to figure out the use, as well, many moons ago. Let's try another explanation: Disambiguation pages attract incoming links (many, many links; see the tables at WP:TDD). Most of them are accidental, because the links are supposed to go to a particular article, not to the dab page. These links need to be fixed, and there are many editors who work tirelessly at this task. However, sometimes we want to link to the dab page (as in hatnotes), and the link by way of the (disambiguation) redirect lets those hardworking link-fixing editors know that they can ignore that one, because it's not broken.--ShelfSkewed Talk 18:33, 27 October 2010 (UTC)

There's nothing wrong with the hatnote displaying the (disambiguation) title, even when the (disambiguation) title is a redirect to a base-name disambiguation page. -- JHunterJ (talk) 19:42, 27 October 2010 (UTC)

What determines Primary Topic BEYOND outbound link usage?

Suppose we had dependable statistics on the use of links out of dab pages (and to articles from the search box) so we have accurate understanding of the relative use of the links to articles associated with a dab page. If the page has all the links it "should", should there be any other input into the primary topic determination? If so, what and why?—Preceding unsigned comment added by Jwy (talk • contribs) 18:26, 25 October 2010 (UTC)

In some cases one article possibly should be the primary topic no matter what the stats are like. Night is the first (and not exceptionally good) example I can think of, I'm sure people can think of better ones. If we assumed the traffic stats (18K for the time-of-day and 27k for the book in August) were instead the results of the "perfect" measure you describe and not the simple raw stats should we then follow these and make the time-of-day not the primary topic? Or should the fact that its the simple every day meaning take precedence even if its not the most popular search?--Nilf anion (talk) 22:01, 25 October 2010 (UTC)

I personally don't see that making the book primary makes Wikipedia worse. How important is handling the shock/surprise at finding an unexpected, but understandably "more popular" article when you enter a term? And if we must handle that shock, the bounds of "simple every day meaning" are going to be a tricky to define and easy to argue about. Any thoughts there? --John (User:Jwy/talk) 22:45, 25 October 2010 (UTC)

I agree with you, Jwy. It should be noted that PrimaryTopic does make an exception like this for so-called vital articles. That is, if Night was a vital article, then it would have claim to primary topic, regardless of normal primary topic criteria. But it's not vital, so it's subject to the same criteria and comparison as all other non-vital articles. --Born2cycle (talk) 22:57, 25 October 2010 (UTC)

It tries to make an exception for vital articles, but ends up with no exception. "An exception may be appropriate if only one of the ambiguous topics is a vital article. In such a case, consensus may determine that the vital article should be treated as the primary topic regardless of whether it is the article most sought by users." is no improvement on "Consensus may determine that an article should be treated as the primary topic regardless of whether it is the article most sought by users.", which is always true. I would like to see it strengthened so that it is a recommendation even without having to gain consensus on each page. -- JHunterJ (talk) 23:03, 25 October 2010 (UTC)

Wikipedia is first and foremost an encyclopedia that is collaboratively edited. It is not a web-site that arbitrarily renames articles based on traffic statistics. The judgement of editors is and always has been the critical factor in determining primary topic. older ≠ wiser 00:29, 26 October 2010 (UTC)

Again, I am asking a hypothetical question to capture what other things go into determining a primary topic beyond traffic statistics. Are there any other factors that you see editors generally using when they argue against the statistical choice? --John (User:Jwy/talk) 04:27, 26 October 2010 (UTC)

Yes, I find the "judgement of editors" answer to be most unhelpful. Even to justify something per WP:IAR you need to provide good reason that action is good for the encyclopedia. --Born2cycle (talk) 04:47, 26 October 2010 (UTC)

I agree with Bkonrad's views. This is an encyclopedia, not a popularity contest. The judgment of editors is key to many decisions about the content of the encyclopedia. I'm sorry that you, Born2cycle, think that is unhelpful, but it is nonetheless the case. Editorial judgment cannot be quantified, and because the facts of each situation are different, one cannot state in advance every consideration that may be taken into account in every decision about a primary topic. I do agree that editors need to provide good reason for their views; "because I said so" is not an exercise of editorial judgment. --R'n'B (call me Russ) 10:23, 26 October 2010 (UTC)

I'd also say we have a "moral obligation" to prevent linkrot. Constantly moving popular pages around means external links to Wikipedia becomes less reliable (even if the harm is just following the top link on a dab or clicking through a hatnote), if this is done too much it has the potential to do real harm to the project. Given that page moves should only be done for a good reason.--Nilf anion (talk) 10:48, 26 October 2010 (UTC)

Can you indicate what some of those "good reasons" might be? If there are some general ones (like weight being added for vital articles), it would be useful to have them listed somewhere. If there are some specific ones that you found especially compelling, I'd be interested in hearing those as well. --John (User:Jwy/talk) 16:26, 26 October 2010 (UTC)

Russ, you say this is an encyclopedia, not a popularity contest. I'm not sure what you mean by that in this context. There are two ways I can think of to interpret "popularity contest" here in this discussion about determining primary topic:

Determining which (if any) topic is primary based on popularity of usage of the name in question in reliable sources.
Determining primary topic based on which is most popular as primary topic (if any) among editors participating in a discussion about that.

(1) is what WP:PRIMARYTOPIC requires us to do. (2) is generally how we make the determination in (1).

I don't know what you guys mean by "editorial judgement". As Jwy says, "editorial judgement" based on what good reasons, exactly? How is that different from a popularity contest in sense (2)? --Born2cycle (talk) 16:45, 26 October 2010 (UTC)

The proposition at the start of this thread is "popularity" as in what is most wanted by readers? This may or may not be the same as your situation (1). In particular with some topics (ie memes) the relative interest of two topics to readers could change drastically, but this would not affect the reliable sources. Likewise the primary topic of a phrase can genuinely alter with time, eg on the historical timescale for cities. We should respond to the second sort of change, but not the first - as the second will affect reliable sources, but the first will not.

"Importance" is something that can be used as a predictive measure, and a numerical value can be attached to it may be easier traffic stats. The fact it makes more sense to the general community is a bonus. For example, reader interest in hurricanes can be predicted by how destructive they are and when they were. Reader interest and source coverage of settlements can be predicted by population - a city of 100,000 is more likely to be viewed and have stuff written about it than a village of 500.--Nilf anion (talk) 17:32, 26 October 2010 (UTC)

I can see make sure we don't react to short term changes in interest. A problem with the "importance" criteria as you discuss is that often the ambiguous articles don't have a common measuring criteria (one ambiguous article might be a city, another a movie, another a brand of underwear).

But it looks like you are suggesting we use "importance" as a way to get a better statistics to predict "interest" of those entering the dab term. That's a different question than what I am asking: Assuming we have a reasonable way to predict this kind of "interest" (I believe that's what the outlink statistics would do), what additional criteria should we look at. Put another way, if we had those statistics, what arguments should prevail against the primary topic determination they suggest. --John (User:Jwy/talk) 19:18, 26 October 2010 (UTC)

"Importance" is a test that can be applied in some cases, though you can't compare apples and oranges. However, I think there is a key difference between popularity in reliable sources and popularity in Wikipedia's readership. The first conforms to policy such as WP:NPOV. Using popularity with readers will inevitably run into the problems of systematic bias, which is an indication that going wholly down that route is in error.--Nilf anion (talk) 10:04, 27 October 2010 (UTC)

If the goal is to optimize navigation, a dab page necessarily has to cater to those who actually come to Wikipedia. We are not biasing the information in the articles by getting people to the articles they are looking for quickly. Some MAY interpret the selection of a primary topic as giving the topic special importance beyond this goal, but we are not doing so and we are not changing the information in the articles. Am I missing how systemic bias might creep in? --John (User:Jwy/talk) 15:57, 27 October 2010 (UTC)

Agreed. I have pointed out before that WP:BIAS and WP:RECENTISM can and should guide article content. Arranging complying articles in any particular fashion won't void their compliance though, and arranging them to best serve the readership will, ahem, best serve the readership. -- JHunterJ (talk) 16:23, 27 October 2010 (UTC)

Put me down as agreeing with Nilfanion. Let's say, for the sake of argument, that there are many more Wikipedia readers who are interested in popular jazz-fusion music than in American history. Then, the "it's just a navigation aid" argument would suggest that George Clinton (musician) should be the primary topic of George Clinton, despite the existence of several other articles about notable people by the same name, including a U.S. Vice President. That "arrangement" conveys to the minority who are reading about American history that their interests are less important to Wikipedia than the interests of music fans. Yes, I know it doesn't say that anywhere, but it still gives that impression. If there were a systematic effort to designate primary topics solely to satisfy the interests of a majority of our current readership, we would very likely be alienating many other readers and, perhaps more importantly, potential readers. --R'n'B (call me Russ) 18:23, 27 October 2010 (UTC)

Then there should never be any primary topic for any ambiguous title, lest we imply that any group of readers is less important than any other. I disagree with that conclusion, and would rather call the primary topic the primary topic and trust that the readers will not be "alienated" by the limitations of the encyclopedia (two articles can't have the same title). The explanations of the policies can be expanded to salve that wound, if it does exist. -- JHunterJ (talk) 19:37, 27 October 2010 (UTC)

If the determination of a primary topic is based upon usage in reliable sources, there is no implication of readership bias. For example, no one could reasonably contend that having Mao redirect to Mao Zedong suggests bias; any contention that, say, the Mao Restaurant in Lower East Armpit is an equally important topic would be ridiculous on its face, and could easily be refuted by reference to reliable sources. --R'n'B (call me Russ) 14:00, 29 October 2010 (UTC)

That tends to be my position as well, but I want to make sure I understand the other position. The assertion is that by choosing to have one article as primary instead of another, we appear to be conferring higher importance to that article - and we should consider this when choosing a primary. But, as JHJ points out, this argument could be used in all cases: Why should P-funk fans get alienated and not Mr. VP fans? How can we do so without introducing some bias? We handle it in an organized way with the vital articles concept, but doing so on a broader scale could prove difficult. Put another way, Russ, what guidance would you give to editors to implement your concept in the more general case? --John (User:Jwy/talk) 23:48, 27 October 2010 (UTC)

The bias issue I mention is probably overstated there. But there are certainly exceptions. The traffic for tea bag is 8K, the traffic for tea bag (sexual act) is 14 times greater. If we determine primary topic solely by navigational utility then clearly the wrong article is primary. Swapping the two would be wrong - the shock value for people looking for info on bags of tea would overwhelm the navigational aid for those with "dirty minds"... Is an IAR exception enough for that one instance or should the potential for similiar situations elsewhere be borne in mind?--Nilf anion (talk) 23:19, 27 October 2010 (UTC)

We have to be careful with the statistics. We don't know how much of the 8K and 14*8K hits are from someone entering "Tea bag", which is the subset of those numbers of interest ("Teabagging" redirects and many pages link directly to the sexual act, for example). But that's beside the point here: you are saying "reduce shock value" is one of the non-navigational attributes we should watch for. I'll think about that a bit. --John (User:Jwy/talk) 23:48, 27 October 2010 (UTC)

Exactly. Where it becomes contentious, I have created new redirects to be used exclusively on the disambiguation pages, so that the stats would more accurately reflect what we're trying to measure. Lincoln (disambiguation) and EA (disambiguation) have each used that approach in the past, with various acknowledgment of their results. -- JHunterJ (talk) 14:15, 28 October 2010 (UTC)

Yes. But if we DID measure it accurately and the results were as described, we MIGHT want to still avoid the extreme shock of someone searching for the more "subdued" of the articles - but suspect actual instances of this case are rare, unless our threshold for "shocking" is low. --John (User:Jwy/talk) 15:36, 28 October 2010 (UTC)

Agree, the stats I quoted are very naive (however those basic stats are what people use first). Agree also its hard to judge the shock/offence threshold. If we are going to go into the purely hypothetical: Imagine US music act that becomes more popular than Elvis (and becomes one of the top 10 articles by hit count), a purely navigational-based system would see this usurp just about anything. If that act was called Qur'an and that resulted in the page there moving to Qur'an (book) how would that go down? That may cause serious offence to a significant group of readers. At the other extreme are ones like Night/Night (book) - reversing that might cause confusion but it won't get a stronger emotional reaction. This is all subjective and depends on how the reader feels about the "wrong" target, we cannot assign a number to it. In general, I think readers are liable to understand the base page as being more "important" than disamiguated pages, even though that isn't the intent.--Nilf anion (talk) 21:14, 28 October 2010 (UTC)

OTOH, I'd rather make sure the system works well for plausible cases, even if that means it might break down under purely hypothetical ones. I think any liable misunderstanding that results can be addressed by assuming good faith and working out a consensus civilly. :-) -- JHunterJ (talk) 21:22, 28 October 2010 (UTC)

Yep of course. That's an extreme one. In realistic cases, the strongest reaction I can see is "WP is clearly biased to <topic, which I don't care for> because it favours that X over <topic, which I do> Y." That is a "negative reader experience" for want of a better phrase and may cause anger (possibly showing as rants on the talk pages) but shouldn't be too serious.--Nilf anion (talk) 21:27, 28 October 2010 (UTC)

I have argued that the WP:RM process has a built in bias since the discussion happens on the talk page of the article proposed for the move. This means that the people interested in the article are the predominate participants in the discussion. Naturally they are the ones that would be inclined to think that the article as is properly belongs at the main name space. Vegaswikian (talk) 01:34, 29 October 2010 (UTC)

I'm not looking to quantify the non-navigational reasons, just to identify any general categories that we might encounter and have a discussion about their importance without the immediate pressure and passion of a particular dab page. I feel comfortable that favoring vital articles and avoiding shock are understandable categories that we could explain clearly - although there would be some discussion about whether they apply in a particular case. I might identify "avoid being highly offensive" to cover cases like Qur'an example. I'm not sure how to characterize the Night issue. Maybe its the case there that the highly expected article is not the "navigationally correct" article and there is a surprise? --John (User:Jwy/talk) 02:00, 29 October 2010 (UTC)

Reinserting clarification about extended discussion indicating lack of primary topic

I just noticed that the following long-standing statement was removed on April 29, 2010 from WP:PRIMARYTOPIC:

If there is extended discussion about which article truly is the primary topic, that may be a sign that there is in fact no primary topic, and that the disambiguation page should be located at the plain title with no "(disambiguation)".

I've just restored it as I believe that consensus supports it and that it's an important consideration in any situation in which the issue of primary topic is raised repeatedly, especially if over a number of years. --Born2cycle (talk) 19:11, 25 October 2010 (UTC)

It was discussed at the time: Wikipedia talk:Disambiguation/Archive 30#"If there is extended discussion about which article truly is the primary topic...", and consensus supported its removal. -- JHunterJ (talk) 19:22, 25 October 2010 (UTC)

I don't see consensus support for removal there.

Propaniac proposes a slight rewording
Jwy supports that proposal with a slight change that Propaniac incorporates
Vegaswikian implies support with a small suggested change (users should be readers)
Lee∴V supports and suggests and other slight rewording "An extended discussion, with no consensus, may be a sign that there is no obvious primary topic."
Kotniski states that saying "may be" is important, which Lee adds back in
Propaniac suggests: "An extended discussion about which article truly is the most likely target, leading to no consensus, may be a sign that there is no clear primary topic."
RnB suggests another slight change to which Propaniac agrees
Note: still no discussion at all about removal of the phrase entirely.
Kotniski finally suggests deleting the phrase altogether, for reasons that frankly I don't fully understand, then seems to be back off a bit ("maybe it's not as strong as that...").
Then Russ complains about language as "amended by Kotniski", but it's unclear what language he's talking about.
Then Kotniski clarifies his point.
Then Lee makes a point indicating he's still thinking inclusion, which Kotniski asks about.
Propinksi disagrees with Kotniski (the statement can't be both meaningless and biased). He wants to add it back in, but is, at best, okay with it not in there.
Kotniski states his main concern: "But if it is to be put back in, then I don't agree that there should be any bias towards the "no primary topic" outcome in discussions where only one topic is seriously proposed as a primary topic."
Russ ends by indicating he's open to suggestions for a rewrite.

Again, I don't see clear consensus for deletion at all, nor even objection to the wording as it stood at the beginning of that discussion (which started as a suggestion to make a relatively minor change to the wording, not delete it). I would like to re-open this discussion, because, again, I think an important and logstanding criterion was removed here.

In particular, I want to hear Kotniski's take on this challenge to his argument: If there is repeated and extended discussion, such as when WP:RM proposals to move the article in question to a disambiguated name and put the dab page at the name, are repeated multiple times over several years, about whether even just one topic is the primary topic, then that in and of itself is strong (though not definitive) indication that it is not the primary topic (and, thus, there is no primary topic) for the term in question.

So, I hereby propose adding the following:

If there is repeated extended discussion about which article among two or more truly is the primary topic, or about whether one particular article is the primary topic, that may be a sign that there is in fact no primary topic, and that the disambiguation page should be located at the plain title with no "(disambiguation)".

I mean, if significant numbers repeatedly question the primacy of a topic for a given term, that alone surely puts that primacy in question. --Born2cycle (talk) 20:19, 25 October 2010 (UTC)

I would rather see it remain unrestored for the reasons given in the earlier discussion -- it is misused by editors who have no other position to fall back on in objecting to an arrangement that does reflect the actual primary usage. That there is repeated objection just means that Wikipedia editors are passionate (which I don't object to and is indeed a good thing), but not an indication that there is no actual primary topic. (There was consensus before -- the edit came out of the discussion and stood; that's consensus.) -- JHunterJ (talk) 20:31, 25 October 2010 (UTC)

Also, the proposed addition doesn't reflect "significant numbers". -- JHunterJ (talk) 20:37, 25 October 2010 (UTC)

I understand the concern, but let's not discount the meaning of passion. If there is a lot of passion about a certain topic not being primary, isn't that good indication that it is not primary? Also, this phrase has been in there a long time and was removed as a result of a discussion by a handful of people. There could be many like me who just didn't notice and assume it's still there. The real issue is whether the phrase or its removal better reflects community consensus (not just those participating in that or this discussion); I think the spirit of the removed phrase does reflect community consensus. To incorporate the significant numbers, how about this?

If there is repeated extended discussion about which article among two or more truly is the primary topic, or about whether one particular article is the primary topic, with significant numbers questioning the primacy of each topic in question, that may be a sign that there is in fact no primary topic, and that the disambiguation page should be located at the plain title with no "(disambiguation)".

--Born2cycle (talk) 21:07, 25 October 2010 (UTC)

No, passion is not a good indication of primary-ness or primary-lessness. Consensus is defined by the participants (and can be re-defined by later participants) -- there is no way to include non-participants. I still prefer the omission of the paragraph; I'll see who else weighs in. -- JHunterJ (talk)

I suggest that the consensus view has been that considerable and long-standing passion opposing the idea that a given use is primary is a good indicator of that topic not being primary is exactly why this phrase was added and supported by consensus for so long, and by removing this phrase this consensus view is not being represented here. --Born2cycle (talk) 23:30, 25 October 2010 (UTC)

I must agree with JHJ that any "passion" being displayed on a topic should be viewed with great suspicion - it's when people start getting passionate about things that rational arguments stop being heard. If there's discussion about whether A or B (or C) is the case, it may be a sign that A, or B, or C, is the case. That's all. We can't say that it may be a sign that A is the case to a greater extent than for B or C. The deleted passage gave the appearance of doing that (i.e. saying that all the anti-primary-topic side have to do is make a lot of fuss about the matter, and they automatically gain the upper hand in the argument purely on the basis that they're making the fuss). I think we're well rid of it.--Kotniski (talk) 07:58, 26 October 2010 (UTC)

I guess I've just seen too many cases of repeated efforts to move something being shot down year, after year, until, finally, the move occurs, sometimes because of this reasoning, and then stability and peace set in. One example that springs to mind is the move of Cork to Cork (city). I don't think the phrase implied that passion alone should be a determining factor; just that it's something to be considered (because, frankly, there is always a reason for the passion - you just have to determine if the reason is a good one). --Born2cycle (talk) 16:26, 26 October 2010 (UTC)

Perhaps "passion" is a bit of a red herring. A primary topic is the "obvious" meaning of a term. If there is disagreement in good faith as to what that obvious meaning is, it seems reasonable to decide that there is no primary topic. The consequence in practice is that all users go via a dab page, which is no burden if it is clearly laid out, while the alternative would be that a substantial minority are taken to the wrong page, then have to follow the hatnote to find what they want. Overall user-friendliness is often better served by having no primary topic. SamuelTheGhost (talk) 11:53, 8 November 2010 (UTC)

Again, all this applies equally well in reverse. If there are people arguing in good faith that there is an obvious meaning (and no-one arguing that any other meaning is obvious, usually just that there is no obvious meaning) then it seems reasonable to decide that the meaning really is obvious (anyway, no less reasonable than to decide that it isn't). Then ,the consequence in practice is that most users go straight to the page they want, saving them a potentially awkward navigation step, while only a minority are required to take an extra step (and even that is often not necessary, if the only other significant meaning(s) can be included in the hatnote) - so user-friendliness is also often better served by having a primary topic. Users are best served if we get these decisions right, whichever way "right" might be in a particular case, not by trying to pre-bias the outcome of the discussion.--Kotniski (talk) 12:42, 8 November 2010 (UTC)

Helping p-i-g-(s) searchers get where they want to go

This is not a move request, so please do not instruct me in making move requests. What I want to establish now is simply that it is highly likely that a person who goes to Wikipedia and types in p-i-g(s) is highly likely to be trying to find an article about regular, ordinary, domestic pigs. You may say that to do so would in effect be a move request, and that may (or may not) be the case, but exactly which one or series of moves turned out to be a complicated question with several possible solutions and a cascade of ramifications, all of which would have to be thought through before a move request that can create concensous can be properly written. The best way to go about it is one step at a time. First, we must establish what a person who types in "p-i-g-(s)" is most likely trying to do. Then we can get into the sticky question of what do about the problem. Please instruct me in the best way to get concensous that it is safe to assume that they are looking for an article about plain, ordinary pigs. Then we can move on to the next question. Chrisrus (talk) 18:35, 31 October 2010 (UTC)

The best way to find out what the consensus for the target of "pig" is is discussion at Talk:Pig; the best way to find out what the consensus for the target of "pigs" is is discussion at Talk:Pigs. The tools from WP:PRIMARYTOPIC may be useful in those discussions. -- JHunterJ (talk) 18:45, 31 October 2010 (UTC)

Thank you for your kind reply. I have been and am discussing this matter there at Pig, Domestic Pig, Pig (disambiguation), and here. Domestic Pig is the obvious most likely target referent of a searcher, so WP:PRIMARYTOPIC is not being followed. Ok; that's fine, I suppose; it's only a guideline and long established consensus may trump a general guideline. Reasoning includes but is not limited to the idea that even though and despite the fact that users are clearly most likely to be looking for domestic pig, we assume ignorance of everything but the English language, so they are sent elsewhere because one of the first things to learn about pigs is that there are also other these other pigs apart from familiar pigs. There are probably other reasons, too: inertia; the fact that there's no clear fix....trust me, it turns out to be more complicated than you probably imagine, and this question is very simple. At this point, my sights are set very, very low. If I first establish domestic pig as the most likely intended target, and then we can all re-read WP:PRIMARYTOPIC and make sure we all agree on what it says.

So I wonder if it's not too much bother, please if you could just go on record that you see the same obvious likely search target that I do and go no further for the moment. Chrisrus (talk) 04:00, 1 November 2010 (UTC)

I don't mean to seem cold about it. I've seen my own obvious primary topics also ignored at various places, like Lincoln. Sometimes the local majority at a particular page will tilt against the guidelines. I will go read the discussion at Talk:Pig. -- JHunterJ (talk) 11:23, 1 November 2010 (UTC)

External links on DAB pages

I have removed the odd external link from dab pages, as they had no WP article and seemed, possibly, to have been put there for promotional purposes. However I can not find any 'rule' on "Wikipedia:Disambiguation" page against this. Perhaps one should be added?
I suggest: "External links are not to be out on Disambiguation Pages, they are for disambiguation of articles only."
- 220.101 talk^\Contribs 13:12, 7 November 2010 (UTC)

"Never include external links, either as entries or in descriptions. Disambiguation pages disambiguate Wikipedia articles, not the World-Wide Web. To note URLs that might be helpful in the future, include them on the talk page." WP:MOSDAB -- JHunterJ (talk) 13:16, 7 November 2010 (UTC)

Looking in the wrong place! Thanks JHunterJ! At least I know I was correct to remove the ext. links. Bu,t I think the guideline may need to be given a bit more prominence perhaps? - 220.101 talk^\Contribs 15:04, 7 November 2010 (UTC)

The guideline is also stated at WP:EL#Links normally to be avoided.--ShelfSkewed Talk 18:19, 7 November 2010 (UTC)

I have copied that same prohibition to this project page's "what not to include" section. -- JHunterJ (talk) 22:37, 7 November 2010 (UTC)

WP:PT

Why in the article is WP:PT given as a shortcut when it leads to a disambiguation page? Surely a better shortcut is appropriate? pgr94 (talk) 13:30, 7 November 2010 (UTC)

Done I have removed the shortcut. I didn't replace it since there is a valid one already in place, but if someone wants to create a shortcut with a shorter name, say WP:PRIMTOP, feel free.--ShelfSkewed Talk 18:16, 7 November 2010 (UTC)

"Bosnia" primary topic

I believe that Bosnia should redirect to Bosnia and Herzegovina as the primary topic, but this was reverted by another user. Discussion initiated at Talk:Bosnia#Primary topic. --R'n'B (call me Russ) 19:39, 8 November 2010 (UTC)

use of references

I twice attempted this change and was immediately reverted. My reasoning is as follow:
As a matter of principle, all factual material in wikipedia is subject to WP:RS. No guideline can override that. In this case, the only way to avoid the issue is to guarantee that no factual information is inserted into dab pages. Dab pages usually have an introductory line, followed by a bulleted list of entries. It is quite reasonable to stipulate that there should be no references in the individual entry lines, since all factual information in them should immediately be duplicated in the target article. The introductory line, however, which sometimes grows to an introductory paragraph, may contain factual information which is common to all the following entries, and if it does so WP:RS should apply. A simple example of where this might arise is in giving the pronunciation of an ambiguous word. Are people suggesting that this should be banned? or that it should be allowed but required to be unreferenced?
I've been coming across this sort of problem frequently when creating disambiguation/name pages for Arabic names. Some editors have insisted that where these pages contain the meaning of the name, no reference should be allowed. I've tried the alternative of categorising them just as name pages, even though disambiguation is their main purpose. There, however, I've encountered editors who object to the inclusion in those pages of place names which are also personal names, as one would do in a disambiguation page. Splitting into two pages would in many cases lead either to confusing duplication, or to one of the pages being very short, and in any case to a situation which was less helpful to the reader. I'd be grateful for helpful constructive comments on this issue. SamuelTheGhost (talk) 17:16, 10 November 2010 (UTC)

I have thought of this concern before. Usually, there is not really a conflict as much of the information in the dab page falls under "unlikely to be challenged." Secondly, where it might be challenged, I believe the thinking is that information is supported in the lead of the respective article. And finally, more specifically in your case, unless the pronunciation information is necessary to distinguish between articles, it is not to be included according to WP:MOSDAB.

More abstractly, the dab pages are intended for navigation, not information. Thus, references should be unnecessary. --John (User:Jwy/talk) 17:32, 10 November 2010 (UTC)

Also, search for references in the archives above for older discussions. --John (User:Jwy/talk) 17:37, 10 November 2010 (UTC)

Generally I agree with the current practice of excluding references from dab pages, as they produce needless clutter when the information (which should be uncontroversial anyway) is sourced at the target page. However we should probably recognize that in certain situations there exist hybrid pages, like the name pages you refer to, which serve partly as articles and partly as dab pages - in that case the disambiguation guidelines (e.g. no references) should apply only to the dab-page part, not to the article part.--Kotniski (talk) 17:48, 10 November 2010 (UTC)

I tend to agree with Kotniski. In a perfect world, the information about the name (including the pronunciation) would be in a separate name article, not on the disambiguation page; but in the real world, we often don't have a separate name article and there may not be much interest in writing one. Also, although it may not strictly comply with the guidelines, it is not uncommon for a disambig page to have an introductory sentence or two. In most cases, there should be no reason to include in that intro anything of potential controversy that might require a reference, but I can imagine an occasional IAR exception where a reference could be useful. Having said that, I don't favor changing the guidance on WP:D, because any such references should be the exceptional case, not the general rule. --R'n'B (call me Russ) 17:58, 10 November 2010 (UTC)

For those who don't know, surname and given name pages are considered WP:Set index articles, not disambiguation pages, and should therefore contain references in any passages regarding the name. –Schmloof (talk · contribs) 18:02, 10 November 2010 (UTC)

I'm very grateful for the discussion above. In summary, I think it says that I can produce the sort of name/dab hybrid pages which I have been producing, and that I should categorise them using {{sia}} (which I previously was entirely unaware of). If that's right, I'll introduce that template into the pages concerned and if other editors question what I'm doing I'll point them at this discussion. SamuelTheGhost (talk) 18:48, 10 November 2010 (UTC)

No, I don't think so. There might be some exceptional cases where such mixes make sense, but I don't think it should be regularly used. Are you aware of Wikipedia:WikiProject Anthroponymy which covers the name aspects of these pages? Duplication of some articles between a dab page (when the target is commonly known by just the one name that is the dab term) and a name article is fine. If someone is likely to search for someone by a term, their article should be easily accessible from the dab page for that term. --John (User:Jwy/talk) 19:32, 10 November 2010 (UTC)

To be specific, the main, though not the only examples I have in mind are the 57 articles listed under ad-Din and about the same number listed at List of Arabic theophoric names. I'd like to fill out some of the introductions a bit, giving sourced information about the meaning of the name in all cases (as I already do for most of the "Abdul" pages). Are you seriously suggesting I must split each of them into two? SamuelTheGhost (talk) 20:17, 10 November 2010 (UTC)

For DAB pages, the question is what articles will someone be looking for when they enter the dab term in the search box. I don't know enough about the name domain you are looking at, but an example: On an "Elvis (disambiguation)" page, we would likely have "Elvis Presley", but not "Elvis Sina". "Elvis Sina" is not usually known as just "Elvis." Both would appear on a "Elvis (name)" page. The name page could have references (I think). The DAB not. I understand the instructions here to be consistent with that. What do you suggest we change? --John (User:Jwy/talk) 21:50, 10 November 2010 (UTC)

<edit conflict>I took a look at some of those. I notice several that are categorized as human name dab pages that probably shouldn't be. Human name dab pages are for people whose names are more the same than the ones on some of those pages. An example is James Smith: everyone on that page is named "James Smith" (although most have different middle names or initials). I'm not as familiar with Arab names, but it seems to me that the names on, for example, Rashid al-Din, aren't the same that way.

You mentioned categorzing these pages as SIA pages using {{sia}}: I don't believe these would qualify for that. Although given name and surname pages are a type of SIA, they don't use that tag directly, they use {{given name}} and/or {{surname}}.
As for splitting the pages, it sounds like that's what's needed here. Dab pages shouldn't include information about a name, or long lists of people who have the name. Shorter lists of people are OK on dab pages, often separated by given name, surname, and people known by the name alone. When you start getting into information about the name itself, or including references or external links, then you need a non-dab page. I know that sounds like a pain, but dab pages are only for pointing people to other articles. If you decide to go with splitting, I'd be willing to help -- I've done several of those splits before. --Auntof6 (talk) 21:55, 10 November 2010 (UTC)

If you have info about a family name or given name, an anthroponymy article should be created, and references are very useful therein. If you have a list of people with the name (as part of their full name), that list should go on such an anthroponymy article, if it exists. (If you have a list of people with the same full name, that's an human-name disambiguation page, as Auntof6 mentions.) If there is no anthroponymy article and the list is relatively short, it can be added as a section of a disambiguation page (if one exists) for the title, but without the references. If you've got references, they'll need to go on an article. -- JHunterJ (talk) 22:11, 10 November 2010 (UTC)

Before taking this further I'll mention that I've been having a discussion with User:MegaSloth at User talk:SamuelTheGhost#Your merge of Abdul Halim and Abdul Halim (name), as a result of which he has created a new page Abdur Rahim (disambiguation) to go with the page Abdur Rahim. We haven't yet talked through the appraisal of how well this works, but I'm dubious, particularly when you look at the implications for the alternative spellings which bedevil these names, and the corresponding many redirects. SamuelTheGhost (talk) 22:33, 10 November 2010 (UTC)

That example looks good to me! --John (User:Jwy/talk) 22:37, 10 November 2010 (UTC)

The reasons it is not so good are the following:

"Abdul Rahim" is a single given name. Nobody is called just Abdul Rahim alone, just as nobody is called just "James". What happens is that when the name is used in a context where it is adequately distinctive, it gets used alone. English-speaking journalists and others go along with this because they don't realise what a common name it is, while the holder may foster the same illusion for reasons of vanity. So in all the cases where brackets have been added to the name, there is in fact one or more further names, but wikipedia has not discovered them. Furthermore if the holder lives in a culture that is not English-speaking, there will in general not be a unique transliteration of his name, and he may equally be refered to as Abdur-Rahim or Abd Al-Rahim or 'Abdul Rahiem etc etc. These are not different names; they are all legitimate transliterations of Arabic or Urdu into the English alphabet.
the half dozen articles in the "dab" page have been selected on the basis of the form of the name which has been used as the title of the wikipedia article, which is essentially arbitrary. The way Megasloth has left it, Abd Al-Rahim (disambiguation) redirects to Abdur Rahim (disambiguation), where the spelling Abd Al-Rahim does not appear, which seems calculated to confuse the user. Furthermore if a user is loking for someone of this name and does not find it in the dab page, he/she may well conclude that wikipedia does not have an article on the person sought, so that the dab page will have positively misled them.
duplicating some of the name page articles on the dab page means that maintenance is a problem. Editors making new modifications to the pages will probably update only one of them
In fact although it may be judged that the example "looks good" in conforming to WP guidelines, it seems to me that it is definitely inferior from the point of view of wikipedia users, and editors for that matter. SamuelTheGhost (talk) 23:18, 10 November 2010 (UTC)

For the dab page, the question is not whether their names really are longer, its whether someone coming to this encyclopedia will enter simply "Abdul Rahim" to find him (like my Elvis example above). If journalists have not been using the full name and the subject is notable, its likely English speakers will enter "Abdul Rahim" and that is enough to have it on the page. There ARE people just called James (several kings come to mind). The selection of the articles for the dab page should be based on that. I'm not sure what maintenance problems you would have. A page moved would include a redirect from the old item. Can you be more specific? My "looks good" assumed that the dab page entries were appropriately selected as described above. It "looks good" in that the approach appears to be right. The detail as to whether the entries are correct I don't know. . . --John (User:Jwy/talk) 23:35, 10 November 2010 (UTC)

There are about 50 entries at Abdul Rahim. We must assume they are all notable, otherwise they should not have articles. I can't prioritise them as more or less notable and wouldn't wish to attempt that. Any such attempt would be both time-consuming and contentious. Any of them might legitimately be referred to just as Abdul Rahim, whether or not further names for them are given there. Except where they live in cultures which use the Latin alphabet, they could be referred to by any of the dozen or so legitimate transliterations. My point remains: a user could search using the name Abdur Rahim or some variant for a person whose article was not in the dab list, and falsely conclude there was no article.

The maintenance problem I had in mind was the adding of the names of new articles.

My challenge to MegaSloth was to get an example where the detail was right. That way, alone, I might be convinced that the approach is right.

It remains my view that the only viable way of splitting is to put all the names in one article and none at all in the other. In cases where the discussion part is substantial, as with William and William (name), that's fine, and that's what is done there. That doesn't make much sense where the discussion part is just a paragraph, and then the sensble way forward is to allow a hybrid. Whether it's classed as a name page or as a dab I don't mind. SamuelTheGhost (talk) 00:33, 11 November 2010 (UTC)

An all inclusive list neither filtering for notability nor assessment of readers tendency to use just the partial name should be a name page - and can use references all you like. If you or someone takes a shot at doing the appropriate analysis and filtering, a dab page can be created and consensus for what should be included there discussed on the talk page. --John (User:Jwy/talk) 01:04, 11 November 2010 (UTC)

Maybe we should remind ourselves what a dab page is for. The general principle is set out in the first sentence of WP:MOSDAB that "disambiguation pages ("dab pages") are designed to help a reader find Wikipedia articles on different topics that could be referenced by the same search term." The assertion at MOS:DABNAME that "pages only listing persons with a certain given name or surname (unless they are very frequently referred to by that name alone) are not disambiguation pages" seems to conflict with the general principle. "Could" somehow seems to have been interpreted as "are very frequently". The Abdul Rahim case seems to be a clear case where a dab page is needed - relative notability of individual entries should not come into it, unless one is a primary topic. --Mhockey (talk) 03:53, 11 November 2010 (UTC)

The DAB page is a navigational aid. To optimize it as such, we need to estimate the behavior of the readers and put more likely (not necessarily the same as "notable") targets "closer" to the start of the search.

Are you suggesting the reverse of what I've said: That the unfiltered list of everyone with the word X in the name should be on the dab page? --John (User:Jwy/talk) 04:10, 11 November 2010 (UTC)

No, but the dab page should include everyone who could be referred to by term x. Call that "unfiltered" if you like, but that is what dab pages are for, and that is what WP:MOSDAB tells us. "Could" is not a high threshhold, a lot lower than "very frequently". I agree that it is sensible to order entries on a dab page to take account of users' searching behaviour, but to exclude from a dab page names of people who could be referred to by that term or name is just unhelpful.--Mhockey (talk) 14:03, 11 November 2010 (UTC)

I'm afraid I find "we need to estimate the behavior of the readers and put more likely ... targets 'closer' to the start of the search" to be profoundly mistaken. Actually doing that estimation, by whatever means, would be extremely onerous, necessarily involving subjective judgments, and almost always contentious. Furthermore it's not even what users want. On encountering a dab page users want to understand what they see. This means there should be some visible order in the list. Recognising that order, users should be able to find what they want without too much trouble. The order I've been using mostly is chronological, by date of birth. There are certainly reasonable alternatives, for example by nationality or by area of notability (sports, politics, arts, terrorism etc). But an ordering by what some editor has judged to be "more likely" will appear to a user as random. SamuelTheGhost (talk) 11:57, 11 November 2010 (UTC)

I agree with the formulation "it is sensible to order entries on a dab page to take account of users' searching behaviour". When I'm a user I try and see whether what is there is in some logical order, and if so I use that order to locate what I want. SamuelTheGhost (talk) 14:57, 11 November 2010 (UTC)

I used the phrase '"closer" to the start of the search" not just to mean "higher on the dab page," but also to include pushing the least likely off to another page. This, I believe, is standard practice on name related dab pages. A dab page with hundreds of entries is not efficient or useful. If the lesser names have to be found by navigating to the name page, that's okay. Yes, on some pages this determination can get contentious, but that's what talk pages are for. If "filtering" proves impossible, don't have a dab page. --John (User:Jwy/talk) 15:55, 11 November 2010 (UTC)

I started this discussion with a request for helpful constructive comments, and have indeed had some, especially early on. They seem to have dried up, however. I notice that Auntof6 offered to help, for which I am grateful. I don't want to behave as WP:OWNER of the Arabic name pages we're talking about, but on the other hand I asked for persuasive advice, not to be told what I must do. So in that spirit I invite Auntof6 to restructure one or more of these pages as she thinks right, preferably without completely dumping good information, and I shall comment. SamuelTheGhost (talk) 17:16, 11 November 2010 (UTC)

Auntof6's notes on trying to split

The first thing I notice is that some of these pages (such as Abdul Bari) contain only human name entries. On those, I think all you need to do is remove the dab tag (whether it be {{disambig|hn}} or {{hndis}} or whatever, and turn them into given name, surname, or human name (category "Human names") pages.

Some of the pages (such as Abdul Ghaffar) contain mostly name entries with only one non-name entry. On those, I think you could remove the dab tag and put the non-name entry in a "See also" section.

As for pages with multiple non-name entries, I gave it a shot with Abdul Samad, splitting the name entries off to the new page Abdul Samad (name). See what you think. If it doesn't work, we can revert Abdul Samad to what it used to be and delete the new page.

Question: Are these two-part names considered single names, or name combinations? To me, who knows pretty much nothing about Arabic names, it would seem more accurate to call them name combinations; otherwise, you'd have a page for (for example) Abdul and one for Samad, and they could both list all the name (plus other combinations). You need to guide me there, though. --Auntof6 (talk) 06:05, 13 November 2010 (UTC)

Thanks very much for your constructive response. I think we may be getting there. The answer to your question, is that they should be considered single names, although consisting of three Arabic words, like Abd-ul-samad, and actually in Arabic script divided as Abd ulsamad. Traditionally that could be abbreviated to Samad, or whatever. Only in modern times and I think only in English-speaking environments has there been some use of Abdul as if it was a separate name. This is very like supposing that a Scotsman called Macdonald has first name Mac and surname Donald, and of course "Mac" does get used on its own, but it's not usually correct to do so. SamuelTheGhost (talk) 12:52, 13 November 2010 (UTC)

Football

There is a rather long discussion at Talk:Football#RFC:_Association_football to which this question relates and I can seem to find any guidance on this. The issue is, is it acceptable to use football or any other ambiguous phrase for that matter and have the user figure out for the rest of the article the sport in question is Association football

(example: Manchester United Football Club is an English professional football club, based in Old Trafford, Greater Manchester, that plays in the Premier League.

or should it'd be explicitly made clear some how.

(example;Manchester United Football Club is an English professional association football club, based in Old Trafford, Greater Manchester, that plays in the Premier League.

Gnevin (talk) 11:58, 12 November 2010 (UTC)

IME, the first style is prevalent and is clear enough. Only if you mention two or more different footballs in the same section or article does it become necessary to explicate inline. --Joy [shallot] (talk) 14:40, 12 November 2010 (UTC)

I'm sorry but how is it clear? Gnevin (talk) 15:50, 12 November 2010 (UTC)

BTW, this sounds like a style question. Disambiguation is the process of resolving ambiguity in Wikipedia article titles for articles that might otherwise have the same title. If you're asking about phrasing on a disambiguation page entry description, it would be "* Manchester United Football Club, an English professional football club" (or simply "* Manchester United Football Club", with no description). If you're asking about the phrasing within the article, Talk:Manchester United F.C. or Wikipedia talk:WikiProject Football/Clubs would probably need to determine. -- JHunterJ (talk) 14:41, 12 November 2010 (UTC)

So you think this is an issue for WP:FOOTY . Ok thanks so Gnevin (talk) 15:50, 12 November 2010 (UTC)

How to link to a disambiguation page

The "link to the title that includes the text "(disambiguation)", even if that's a redirect" part of this guideline is entirely asinine and unnecessary. Why should we bother to use a redirect? Redirects are for common misspellings or alternate names for articles. Not for dabs or lists.—Ryūlóng (竜龙) 21:23, 17 November 2010 (UTC)

It's useful for fixing unintentional links to disambiguation pages. I'd explain further, but since you lead with "asinine", I'd rather make sure you were interested in discovery first. -- JHunterJ (talk) 21:25, 17 November 2010 (UTC)

Why do we have 124k pages like this? Why should we point out links to disambiguation pages? Is it all that necessary?—Ryūlóng (竜龙) 21:28, 17 November 2010 (UTC)

It's useful for fixing unintentional links to disambiguation pages, because the intentional links are supposed to use the (disambiguation) redirect. -- JHunterJ (talk) 21:37, 17 November 2010 (UTC)

That does not make any sense. If I want to link to America I will link to America and not to America (disambiguation).—Ryūlóng (竜龙) 22:41, 17 November 2010 (UTC)

Okay. And if I'm fixing incoming links to America, I'll change the link to [[America (disambiguation)|America]] so I (or another editor) won't have to check it again (and again, and again...).--ShelfSkewed Talk 22:52, 17 November 2010 (UTC)

You're essentially claiming that consensus doesn't apply to you. That is not a good approach. -- JHunterJ (talk) 23:23, 17 November 2010 (UTC)

I'm not saying that the consensus doesn't apply to me. I just don't agree with the current consensus. It's unnecessarily circular to have a disambiguated redirect to a disambiguation page that is not disambiguated. Why not just disambiguate every single disambiguation page to include "(disambiguation)" in the title and not use a non-disambiguated title for a disambiguation page?—Ryūlóng (竜龙) 23:45, 17 November 2010 (UTC)

It isn't circular -- the (disambiguation) title either is the disambiguation page (0 hops) or redirects (1 hop) to one, and ends there instead of circling around. I do not understand what you mean by "a disambiguation page that is not disambiguated", unless you just mean that the title has no primary topic and so the disambiguation page is at the base name. We could move all the base-name disambiguation pages to the (disambiguation) title, but so far it has not appeared to be an improvement over the current arrangement. And linking to the redirect is necessary (or at least useful) to distinguish intentional links to disambiguation pages, even those for titles with no primary topic. What problem is that creating? -- JHunterJ (talk) 23:53, 17 November 2010 (UTC)

Having a redirect from America (disambiguation) to America is unnecessary. Why not have it the other way around and just make it part of policy that all disambiguation pages need "(disambiguation)" in the title?—Ryūlóng (竜龙) 00:25, 18 November 2010 (UTC)

That would be a big effort(124K page moves), and wouldn't accomplish anything. All the intentional links would still (be supposed to) use "America (disambiguation)", all the links that need to be investigated would still be pointing to "America". It could make it easier for editors to go against consensus and change a base-name redirect rather than having to move a base-name disambiguation page (unless all of them were protected). Can you answer the question "What problem is it creating?" -- JHunterJ (talk) 00:30, 18 November 2010 (UTC)

The "problem" is that it's incredibly stupid and it obviously wastes people's time. The amount of work people do to check on DABs like this, when the link is intentional, is pointless. I've reverted a user twice for unnecessarily adding a piped link to the redirect when a direct link to the page is perfectly fine.—Ryūlóng (竜龙) 01:10, 18 November 2010 (UTC)

Have you read the answers you've been given? It doesn't waste people's time, it saves them time, by helping them to distinguish links that need fixing from those that don't. --Kotniski (talk) 07:05, 18 November 2010 (UTC)

They really help with de-orphaning too. But we shouldn't expect anyone who hasn't done this kind of maintenance work to immediately understand their purpose. -- œ ^™ 08:26, 18 November 2010 (UTC)

"The "problem" is that it's incredibly stupid and it obviously wastes people's time." isn't a problem. That would be the outcome of a problem, if there were a problem. Since there isn't a problem, it isn't incredibly stupid. Since it saves people time, obviously it doesn't waste their time. If you are not interested in learning the guidelines or discussing actual improvements to them (instead of just throwing about pronouncements of how things that you disagree with only at a gut-feeling level are stupid), please at least stop reverting users for adding the pipe links in accordance with the consensus guidelines. -- JHunterJ (talk) 12:51, 18 November 2010 (UTC)

Please have a look at my explanation in the section Total confusion above; these redirects have a function.--ShelfSkewed Talk 21:41, 17 November 2010 (UTC)

I really am puzzled by this discussion. I understand, Ryulong, that you don't like these redirects and consider them unnecessary, not to mention some of the less civil terms you've used. That's fine, everyone is entitled to their opinion. But, I know that you are an experienced and valued long-time contributor to Wikipedia, so surely you are familiar with the principle that "Redirects are cheap." We have redirects all over the place, and as long as they are not misleading or harmful, they are encouraged. A link like [[America (disambiguation)|America]] looks to the reader exactly the same as [[America]], and it takes the reader to the same place, so what is the harm? Even if you don't like it, I don't understand why it gets you so upset. It is not hurting anyone or anything other than your sense of aesthetics. --R'n'B (call me Russ) 11:25, 18 November 2010 (UTC)

I am going to resurrect long lost plans to write a page describing why we are doing this. Being able to point to a clearly reasoned explanation should help reduce the rancor around this issue. --John (User:Jwy/talk) 23:21, 18 November 2010 (UTC)

Am I on the right track here: User_talk:Jwy/Intentional_DAB_Links? Feel free to edit and discuss on the talk page there. --John (User:Jwy/talk) 03:31, 19 November 2010 (UTC)

Dabfix tool

Hi all, I happen to have Never Say Never on my watchlist, and noticed this diff. I was a little surprised to see how much text was added to some of the entries. Do you see this as reasonable, or a bit much? --AndrewHowse (talk) 23:05, 18 November 2010 (UTC)

More than a bit much, yes. And certainly not a minor edit. -- JHunterJ (talk) 00:48, 19 November 2010 (UTC)

When to split a list into sections

Could the article make some recommendation as to the max nr of entries in the list before it is worth splitting the list into say 3 or more sections ? I would suggest 15 or 20. Rod57 (talk) 19:34, 22 November 2010 (UTC)

Alphabetical or by notability

Some, perhaps many, DAB lists are alphabetical. If this is policy could the article confirm that or say if it is ok or prefered to have the most notable/likely uses near the top (my preference) ? Rod57 (talk) 19:37, 22 November 2010 (UTC)

Where likelihood can be used and useful, it should be used. For long lists, or for short lists where likelihood is undetermined (and so can't be used), readers need to be able to find what they're looking for: sectioning, grouping, and sorting by alpha/chrono/geography can help. But for short lists (or sections/groups with short lists), sorting entries by likelihood is useful and should be used. -- JHunterJ (talk) 19:44, 22 November 2010 (UTC)

Proposal for non-Roman Redirects and Disambiguation

Discussion on a proposal to change parts of this guideline is taking place on the Redirect talk page. Handschuh-^{talk to me} 23:17, 23 November 2010 (UTC)

Should dab pages have infoboxes?

See Feaster. I removed the broken template but the editor who placed it there (COI editor I'd say), replaced it, along with Feasterville, South Carolina, which I'd removed as it doesn't seem to be a legal entity (if it had any legal status I would have left it on the basis there could be an article for it), and an entry for "John Feaster, very prominent South Carolina Planter and Businessman of whom founded the Feasterville Church along with the Feasterville Male and Female Acadademy, of which is listed on the National Register of Historic Places" - this person seems only to appear in a history of Fairfield County [1] where it says about him only "John Feaster, son of Andrew Feaster, was the founder of Feasterville Academy, and donated 7^ acres of land to Liberty Church, and ^^ acres to the Academy. Tradition says that John Feaster had the first glass windows in the township." Dougweller (talk) 14:56, 24 November 2010 (UTC)

No, dab pages should not have infoboxes. If a dab page is temporarily housing a short list of name holders that should really be a separate anthroponymy article, and an editor would like to expand that short list by adding an infobox or other "real article"-style information, the split of the disambiguation page from the anthroponymy article should be made first, and then the article expanded. -- JHunterJ (talk) 15:16, 24 November 2010 (UTC)

Proposal for non-Roman redirects and DABs

To participate in that discussion, go to this link. Thanks. --HXL 's Roundtable, and Record 22:32, 24 November 2010 (UTC)

Should WP:TWODABS not apply at Sephiroth?

Am I misinterpreting WP:TWODABS, or should it not apply in the case of Sephiroth (currently a dab page with two entries) for reasons that I do not understand? See Sephiroth (Final Fantasy)#Requested move for the discussion and reply there please... Thanks. --Born2cycle (talk) 02:41, 24 November 2010 (UTC)

Somewhat more generally, I've always admired the phrasing "should not be created" in TWODABS; it doesn't read "is a burden on the servers", "should be deleted" or anything of that sort. It simply says that if it doesn't exist, then don't bother creating it. Equally, the phrasing used also implies that it's not worth the bother of convincing anyone that it should be deleted - in other words, it's unnecessary, but largely harmless.

I suppose you could take it to MfD, but it doesn't really seem that it's worth the trouble, does it? --AndrewHowse (talk) 03:15, 24 November 2010 (UTC)

Good point, but, yes, I think eliminating 2 entry dab pages is a worthy goal consistent with improving this encyclopedia. The dab page itself is not the problem, especially if it's at [[Plainname (disambiguation)]]. The problem is when the dab page is at [[Plainname]] which means users entering "Plainname" are taken to the dab page instead of to the more likely subject to be sought of the two uses of "Plainname". --Born2cycle (talk) 06:23, 24 November 2010 (UTC)

That could be a move (boldly, or using a move request if controversial) of the dab from the base name, redirecting the base name to the primary topic or moving the primary topic to the base name, and tagging the new-name dab page with {{db-disambig}}. -- JHunterJ (talk) 12:09, 24 November 2010 (UTC)

But I just closed the move request without moving the pages. There was no consensus on a primary topic, different tools given at WP:PRIMARYTOPIC indicated different possible primary topics, so there appear to be no primary topic for "Sephiroth", and the two-entry disambiguation page can continue to disambiguate those two non-primary topics. -- JHunterJ (talk) 12:40, 24 November 2010 (UTC)

By my reading, WP:TWODABS doesn't apply. TWODABS means that if you have a primary topic article at [[Article]] and another at [[Article (about something else)]], you have a hat note at the top of the former pointing to the latter, rather than a hat note pointing to [[Article (disambiguation)]], which in turn points to both articles. —me_and 16:38, 24 November 2010 (UTC)

That's right. WP:TWODABS is for cases with a primary topic. If there's ambiguity and no primary topic, then a disambiguation page is needed even if there are only two ambiguous topics. -- JHunterJ (talk) 17:07, 24 November 2010 (UTC)

I thought the whole point of twodabs is that dab pages with two entries are inherently unhelpful, even if neither topic is primary (yes, I know it doesn't say that, I'm just talking about the underlying reasoning).

The whole point of having a dab page is to help readers find the subject they are seeking with minimal hassle, is it not?

Even if two topics are each perfectly equally likely to be the one being sought when a given term is entered in the Search box, if you toss a coin and put one of the two topics at that term, then half the time the reader will be taken directly to the desired article, while the other half will be one hatnote link click away from their intended destination. So compared to having a dab page, half are clearly better off, while the other half is worse off. But if you put a dab page at that term, then none of the readers searching for that term will be taken directly to the article they are seeking; everyone searching for that term is guaranteed to be a search plus a click away from the desired article.

Frankly, even if you have only three topics and none are primary, if you put the article most likely of the three to be the one being sought at that name, and hatnote links to the other two at the top of that article, then at least 1/3 of the readers will be taken directly to the desired article, while the others will again be only one click away.

The benefit of avoiding a dab page once you have 4 or more articles associated with the term, and no primary topic, starts to diminish rapidly for two reasons. First, because the percentage of readers who get to the desired article becomes increasingly insignificant as the number of entries increases. Second, because the number of hatnote links at the top of the article becomes unwieldy. But none of those problems apply when we have only two or even three articles with topics that are called by the term in question. --Born2cycle (talk) 19:21, 24 November 2010 (UTC)

The underlying reasoning (as I read it) is if there's a primary topic and all of the navigational assistance can be rendered through hatnotes, then there's no need for a disambiguation page, since no additional navigational assistance is needed. The unstated counterpoint is that if there's no primary topic, then there's no place to put navigational hatnotes, so a disambiguation page is needed to render navigational assistance. -- JHunterJ (talk) 19:40, 24 November 2010 (UTC)

Yes, my reading is the same as yours. I'm asking about the underlying reasoning.

I don't understand why the absence of a primary topic means there is no place to put a hatnote. If there are two topics the article of one can be at the name in question (just as well as a dab page can be there), and it can have the hatnote to the other one, even though it is not the primary topic (not primary because it is more likely than the other, but not much more likely, to be the one being sought). Wouldn't that be more desirable for the navigational advantages I noted above... no extra clicks for at least half, probably significantly more in most cases, of those searching for the term, while the others are still only one click away? --Born2cycle (talk) 21:01, 24 November 2010 (UTC)

I was answering about the underlying reasoning. If one of the articles is at the name in question then there is a primary topic. Your premise of no primary topic but one at the base name is in error. Yes, randomly assigning a primary topic would reduce the click count. That is not the underlying reasoning, though. Picking one of two non-primary topics to be primary would be (and will continue to be) contentious. -- JHunterJ (talk) 14:14, 25 November 2010 (UTC)

Not to argue semantics, but it is search likelihood, not article placement, that determines whether a topic is primary or not. If someone changed Einstein to be a redirect to Einstein Bros. Bagels that would not make the bagel company be the primary topic for "Einstein"; the scientist would remain the primary topic and the redirect would be incorrect.

Similarly, there is no primary topic for Cork, which is a dab page. If someone moved Cork (city) to Cork, that would not make it the primary topic. In fact, if placement determined primary topic, then no one could ever argue that an article should be disambiguated on the grounds that it's not the primary topic, because, by your reasoning, it is the primary topic, by definition, simply by being at the base name.

What I'm suggesting is that in the case where a term has only two uses is special. That that is why we have WP:TWODABS. Obviously if one of the two topics is much more likely to be searched for with that term than is the other, then it is the primary topic, by definition. There is no reason to make WP:TWODABS a special case for that situation; it's standard primary topic determination. Therefore, the only situation in which it makes sense to have a special WP:TWODABS case is when neither use is much more likely than the other to be searched for by the term in question. And, yet, in that case, because there are only two uses, it still benefits the reader for us to treat one of the two uses as if it is primary. That'sthe only I reason I can see to make WP:TWODABS be a special case. Am I missing something?

As to the point that deciding which of two uses should be at the base name would be (and will continue to be) contentious for editors, perhaps, but that's not a good reason to make the encyclopedia less convenient to use for readers (by forcing everyone to go through a dab page even though there are only two uses). It is a reason to make the guideline more clear about this, which, if done correctly, should reduce if not all but eliminate the contention. In fact, we might even say that unless one of the two uses is clearly the primary topic, that article movement (between one being at the base name and the other disambiguated) is not justified (status quo prevails). --Born2cycle (talk) 22:10, 26 November 2010 (UTC)

Dab request

Is there someone who loves to create Dabs who could make one of Chatton for all of these? --Kleopatra (talk) 07:33, 27 November 2010 (UTC)

I've added a hatnote and See Also section to Chatton. A dab page doesn't seem to be necessary. Station1 (talk) 07:51, 27 November 2010 (UTC)

Usually, however, you can find some way to get to a surname page. If I don't know a person's first name, I can often find it through the dab for the last name. In this case, I get two placename choices, and no option for the surname. This seems to be exactly the purpose of a dab, that you can readily access all articles. --Kleopatra (talk) 07:56, 27 November 2010 (UTC)

I never saw the benefit of a surname page over using the search function. The latter is usually more up to date and inclusive. Dab pages were never intended to be search indexes, but rather to disambiguate article titles. Station1 (talk) 08:06, 27 November 2010 (UTC)

When I enter "Chatton" in the search box, I am taken to the Chatton page. The search box does not return a page of choices, unless I enter a misspelling that is not on en.wiki, like Chattonstsadfa, then delete the "tsadfa," and I get choices for "Chatton." While the dab pages may have not been intended to be search indexes, the search box is also not intended to be a search index, it is intended to take you to the one page named "Chatton."

So, the dab page isn't a search engine and the search page isn't a search engine, and this means there can only be one article on all of wikipedia titled "Chatton?"

Maybe more experienced editors know how to use the search box to get a choice of pages to click on, but the audience of this encyclopedia includes readers and the less experienced users. There is no way to get to anything with "Chatton" in it through the search box, except for this one placename article.

Again, great, the dab isn't meant to be a search engine, but if the search engine isn't a search engine either, then we might as well warn users that they have to know the full name of their topic before they can find it. Not very encyclopedia helpful. --Kleopatra (talk) 08:22, 27 November 2010 (UTC)

I'm not really sure if all browsers show the same thing, but when I type Chatton in the search box I get a drop-down menu showing all articles starting with "Chatton" and then a separate box below "containing Chatton". There's also a magnifying glass on the right that performs the same search function. I had assumed that's how you got the list of "these" in your original query. I agree that the search function is not as intuitive as it could be, but if another article is added to WP about another person named Chatton, Chatton (surname) will be out of date and therefore unintentionally misleading. Station1 (talk) 08:38, 27 November 2010 (UTC)

No, if I type in Chatton I get drop down choices of Chatton, Chatton transmitting station, and Chatton, Illinois. If I hit the search button, the magnifying glass, I get taken to the Chatton page. How I got the search results is as I stated above, I entered nonsense in the search window, then I got taken to another search page, and I replaced the nonsense with "Chatton," and I got the search results. That's not non-intuitve, that's "doesn't work."

"The search box does not return a page of choices, unless I enter a misspelling that is not on en.wiki, like Chattonstsadfa, then delete the "tsadfa," and I get choices for 'Chatton.'" (From above.)

I won't be able to get to the other person name "Chatton" any more than I could find Édouard Chatton earlier, without knowing his first name, so it really doesn't matter that the surname page is out of date, as there was no means to get to the other Chatton before. --Kleopatra (talk) 08:46, 27 November 2010 (UTC)

And browsers, it does it on both the latest Firefox and on the latest IE. Two very common browsers. --Kleopatra (talk) 08:51, 27 November 2010 (UTC)

Well, you have a good point then. I'm using IE, so I don't understand why you don't get the drop-down at the bottom "containing Chatton", but if some people don't I agree it's a problem. Maybe someone else knows the technicalities of what's happening. Station1 (talk) 09:03, 27 November 2010 (UTC)

I'm puzzled: using Firefox, if I input "Chatton" I get a drop down set of choices of which the bottom one, below a line, is "containing Chatton", and that one leads me to the full list of articles containing the word, which seems to be different from Kleopatra's experience.

That said, it has always seemed to be inadequate that we have no easy way for readers to find all the people with a specific surname. There are circumstances in which people are referred to by surname only ("until Chatton's work in the 1920s" or "the row when Chatton was fired from the team" etc). For living people, those of us who know how to work Wikipedia can look at Category:Living people with its browse bar, but it would be wonderful if there was some similar way to find items on all people, sorted by their sort key. It would save the perceived need for masses of surname pages or immense surname lists in dab pages, all of which are, as stated above, prone to be or become misleadingly incomplete. But I suppose it's been looked into and rejected in the past, somewhere. PamD (talk) 10:01, 27 November 2010 (UTC)

Until someone pointed out that last entry on the search function in a different discussion recently, I had never noticed it -- it isn't exactly intuitive. It is one reason I find value in using {{intitle}} and {{lookfrom}} on dab pages of "common" words. While common words is somewhat subjective, it seems somewhat more user-friendly to offer an option of last resort on a disambiguation page. So long as people are creating articles, creating new redirects, or renaming existing articles, there will always be a possibility that any given disambiguation page will be incomplete. older ≠ wiser 13:35, 27 November 2010 (UTC)

Primary topics with other titles

Trying to get more into the primary topic section, without making it a legalese brain-hurty mess. Possible approaches:

When determining a primary topic for a given term, the question is the likelihood of each topic being the one sought by the reader who searched for the ambiguous title. Whether the topic article has that title is not a consideration. The questions of "which of several topics is most likely sought from this title?" is separate from "which of several possible titles is best for this topic?"

When determining the primary topic for a given term, the significant likelihood of any topic which might be called by that term being the one sought by a reader searching with that term must be considered. Such a topic being at a different title which is not in question is no reason to exclude the likelihood of that topic being the one being sought from the process of determining whether there is a primary topic.

The second leaves me scratching my head, but the first is one I mostly wrote, so I have no doubt that it makes others scratch their heads. Further ideas sought. -- JHunterJ (talk) 02:43, 25 November 2010 (UTC)

The first one makes more sense to me, though I think "The questions of" should be replaced by "The question". In the second version, I have no idea what "the significant likelihood" means. PamD (talk) 14:31, 25 November 2010 (UTC)

Likelihoods are a measure of probability. A likelihood of 0% means it never happens, 100% means it always happens. A very small likelihood is insignificant. A likelihood large enough to be significant is a "significant likelihood", or "likelihood sufficiently greater than zero to be significant". The point is that unless the likelihood of a term being used to search for a topic is practically zero, that likelihood needs to be considered in determining whether that term has a primary topic.

The 1st and 2nd paragraphs don't mean the same thing. I don't understand the point of the last sentence in the first paragraph, or how it relates to what the 2nd paragraph is trying to say. --Born2cycle (talk) 16:14, 25 November 2010 (UTC)

Kotniski, JHunterJ, PamD, et. al, the point I'm trying to introduce and explain which keeps getting deleted is not exemplified by either Einstein or Danzig. The point is that when you're determining whether a given use of a term is the primary use, you have to consider all uses of the term to refer to other topics in terms of their respective likelihoods to be used to search for those topics. Uses are not to be dismissed or discounted in primary topic determinations just because they are secondary uses. --Born2cycle (talk) 17:02, 25 November 2010 (UTC)

Hopefully the "defamation" example does what you're seeking. If not, I'm still not clear on what it is. I'm assuming this is being driven by a current or recent discussion -- which one? -- JHunterJ (talk) 18:01, 25 November 2010 (UTC)

Maybe something like (needs wordsmithing):

When determining a primary topic for a given term, the question is the likelihood of each topic being the one sought by the reader who searched for the ambiguous title. Editors should consider the article the reader is looking for may not incorporate the search term in its title, but still may have a high likelihood of being the article a reader is looking for - and thus may be a primary target.

--John (User:Jwy/talk) 18:05, 25 November 2010 (UTC)

No, Defamation does not exemplify the point either because slander/libel is the primary use of "Defamation", and the term defamation is the name the film would use if the legal use was not primary.

Here is a link to a discussion in which the problem is exemplified. See the comment under Important Note and my reply. The term in question there is "Fergie" and the argument made is that the use by Sarah Ferguson (whose nickname is Fergie) is not relevant since the article about her would never be at Fergie no matter what. I want something in here that explains clearly why that's not true. Just because that article would never be at "Fergie" doesn't mean readers will never search for it using "Fergie", and, so, in deciding whether the singer is the primary use of "Fergie", we need to consider the likelihood that that term will be used to search for the Sarah. I run into this flawed reasoning quite often in WP:RM discussions. --Born2cycle (talk) 16:04, 26 November 2010 (UTC)

It's not entirely flawed, since the fact that her article would never be at "Fergie" (and not just because of Wikipedia's private and perverse naming conventions, but in any serious encyclopedia anywhere) means that readers are less likely to search for it using "Fergie". (How much less likely is hard to measure, though I think we should err on the side of helping readers who type in viable article titles for the topics they're looking for.) --Kotniski (talk) 16:21, 26 November 2010 (UTC)

Yes, the reasoning that the usage should not have to be considered is entirely flawed. The term being less likely to be used for searching for the Sarah Ferguson article because that article would never be at "Fergie" is automatically incorporated in the likelihood estimate. Besides, we have countless redirects under titles that would never be the title of the actual article (but are there because those terms are used to refer to the subjects to which they redirect), and readers expect that. So it's not unreasonable for readers to expect to find the article about Sarah Ferguson by searching for "Fergie".

Besides, I, for one, could only remember the nickname, so if I was looking her up I would have to search for her with "Fergie" (certainly not by the title at which her article is actually at, which I can't remember even after seeing it).

Here's another way to look at it. When considering whether a given term (e.g., "Fergie") is the primary use of a given topic A (e.g., the singer), if the term would be a redirect to a topic B (e.g., Sarah Ferguson) if topic B was the only use of that term, then the likelihood of users using that term to search for topic B relative to the likelihood of using that term to search for topic A needs to be considered. That is, if the only use of "Fergie" was the nickname for Sarah Ferguson then Fergie would of course be a redirect to her article, so that's why we need to consider the use of Fergie to refer to Sarah in deciding whether the singer is the primary use.

So, the idea such a likelihood should not have to be considered is flawed. How do we make this clear? --Born2cycle (talk) 17:05, 26 November 2010 (UTC)

The article at "Albert Einstein" would never be at "Einstein" no matter what -- we use first & last names for articles, and ol' Al never published under just his surname, so that example would seem to be appropriate for the "Fergie" discussion. If people are ignoring the guidelines, they will also ignore expansions of those guidelines. OTOH, I thought the bit about The question of "which of several topics is most likely sought from this title?" is separate from "which of several possible titles is best for this topic?" would have also helped make that clear. -- JHunterJ (talk) 17:42, 26 November 2010 (UTC)

Albert Einstein is the primary topic for Einstein, so of course Einstein has to redirect to Albert Einstein; Sarah Ferguson is not the primary topic for Fergie, but "Fergie" is used to refer to Sarah Ferguson, so in deciding whether some other topic is the primary topic for Fergie we need to take that under consideration. I don't think people are ignoring the guideline when they say it should not have to be considered; I think they don't understand how that's an integral part of deciding primary topic, and that's why I want the guideline to be clear about this point.

I agree the question of "which of several topics is most likely sought from this title?" is separate from "which of several possible titles is best for this topic?", but that is not relevant to this point because primary topic is not merely about "which of several topics is most likely sought from this title?", but whether that likelihood is sufficiently high relative to the likelihood of other topics being sought when that term is entered into the Search box, and this point is about that likelihood. --Born2cycle (talk) 19:22, 26 November 2010 (UTC)

If "Sarah Ferguson" is not the primary topic but is still ambiguous, then the current guidelines already take this into account. In order for the singer to be primary (for instance), the article on the singer would have to be the one most used by readers searching on "Fergie" -- much more than any other (much more than Sarah Ferguson, for example), and more than all the others combined (more than the DJ, Sarah Ferguson, and whatever else combined). -- JHunterJ (talk) 19:33, 26 November 2010 (UTC)

Right, yet never-the-less the argument posted at the Fergie discussion is repeated surprisingly often. Even Kotniski gave that argument some credence above and here (the fact that "Fergie" is "only" a nickname does not in and of itself matter as to how we determine primary topic for it - all that matters are the various likelihoods of it being used to search for the associated topics). I think this points needs to be explained more plainly and explicitily, because it's clearly not as obvious to many as it is to you and me. --Born2cycle (talk) 19:48, 26 November 2010 (UTC)

Another example of this problem is at Talk:Freston,_Suffolk#Requested_move. The entire basis for proposing the move of Freston, Suffolk to Freston is that "The other entries on this disambiguation page just have the word Freston in them, not as the main title."

What can we write in this guideline to make it clear that just because another "candidate" (or entry on a dab page) for a given term doesn't just have that term alone as its main name does not mean we ignore, or discount in any way, how likely that topic is to be searched for with that term when determining whether another use is the primary topic for that term? --Born2cycle (talk) 21:26, 26 November 2010 (UTC)

That's a good basis. The other articles have no indication of being commonly referred to as "Freston". Surname holders, in general, are partial title matches, not ambiguous entries. In the unusual cases where a person is referred to by a single name (without clarification), then they become ambiguous with the title. Elvis, Madonna, Patton, and Lincoln, for examples, (and Fergie, for that matter) are different than Freston and Freston (who?). -- JHunterJ (talk) 22:16, 26 November 2010 (UTC)

No. Arguing that "Freston" is unlikely to be used to search for either of those people would be good basis, if there was reason to believe that was the case. But in the case of articles about people, especially people with relatively unusual surnames like "Freston", we have to presume that they are often searched for by only surname (often, it's all that can be remembered, and it's less typing). In fact, isn't the presumption that people are often searched for by only surname why we have dab pages for surnames?

The main point here is the stated basis -- the term is only part of another use's article title -- is no reason to discount the likelihood of a reader using that term to search for that other use. We can agree to disagree on how likely readers are to be searching for those topics with the term, and that would be relevant, but to simply say that they can essentially be ignored because the term in question is only part of the name used in the title is not correct. --Born2cycle (talk) 22:36, 26 November 2010 (UTC)

I'm with JHJ on this one. Partial title matches should never usurp primary topic status when only one article has a full title match. Wikipedia has a tendency to over-disambiguate, and this is one of those cases. --JaGa^talk 00:04, 27 November 2010 (UTC)

Yes (to Born2cycle's No). Arguing that "Freston" is likely to be used to search for either of those people is the part that needs a reason to be believed. We do not have to presume that articles about people, especially obscure people who are not commonly known at all, let alone commonly known by a single name, are going to be searched for by those names. Instead, we presume that anyone looking for one of them would be likely to use the full name (since any reliable source that mentions them will use the full name) And, as we've clarified many, many times now, we don't have dab pages for surnames. We have WP:Anthroponymy list articles for surnames, and includes lists of people by their surnames there. Only in the cases where no surname article exists and the partial-title match list is short do we take the shortcut of tacking-on the "People with the surname" section after the list of actually-ambiguous entries. We can say that partial title matches can essentially be ignored. -- JHunterJ (talk) 01:12, 27 November 2010 (UTC)

I don't think any of this can be regulated with hard-and-fast rules - we have no way of knowing (do we?) how many readers are searching for articles under particular search terms; we have to use a certain amount of common sense, and I think we should assume there's going to be a certain tendency among readers to enter real titles (particularly now the search box produces a drop-down list), but exactly how much comes down to our judgement. My personal take is that people are going to search for Albert Einstein under "Einstein" quite a lot, relative to any other use of Einstein they might be searching for, whereas they're going to search for Sarah or Alex Ferguson quite rarely under "Fergie", relative to the singer who's "properly" called Fergie. But it's a judgement call rather than something we can lay down firm rules about. In any case primary topic decisions are sometimes made based on other factors, such as the (in)convenience of disambiguating, rather than strictly on the basis of WP:PRIMARYTOPIC - New York probably being the most notable example. --Kotniski (talk) 10:12, 27 November 2010 (UTC)

The hard-and-fast rule is that only an ambiguous topic can be the primary topic -- the primary topic is one of the ambiguous topics. You can find reliable sources that talk about Einstein that don't use "Albert", and about either Fergie without using "Sarah" or "Alex", so they're ambiguous with those single names. You cannot find reliable sources that talk about Tom Freston that don't use "Tom". -- JHunterJ (talk) 14:41, 27 November 2010 (UTC)

I"m not sure what you mean by that - just click on the reference given under Tom Freston, and you'll see him referred to as just "Freston" several times (that's after he's been introduced as Tom Freston, of course, but I think any truly reliable source that referred to Sarah/Alex Ferguson as "Fergie" would only do so after introducing them by their real name).--Kotniski (talk) 18:29, 27 November 2010 (UTC)

I hope you find the New York Times truly reliable. http://tmagazine.blogs.nytimes.com/2010/09/30/using-her-head/ (singer), http://www.nytimes.com/2010/08/08/books/review/Light-t.html (Duchess). Using the surname as shorthand is one thing, but you may as well claim that they are also commonly referred as "He" or "She" then as well. The Frestons are not commonly known by the single name, and are partial title matches. The Fergies are ambiguous with the single name and are not just partial title matches. -- JHunterJ (talk) 19:10, 27 November 2010 (UTC)

I honestly don't see your point, or the difference. I'm not disputing that the singer is referred to as just Fergie, that's the whole point - but the source that calls the duchess "Fergie" puts it in quotes, with an explanation of who they mean right after it - I don't see that's any more or less a sign of ambiguity than the use of "Freston" in the reference I looked at earlier. --Kotniski (talk) 19:16, 27 November 2010 (UTC)

The difference is whether the topic is a partial title match (most name-holders) or an ambiguous entry (very few name holders). If some editor started a surname article on Freston and included the list of notable Frestons there, would they also be on the disambiguation page? No, because they are partial title matches, not ambiguous entries. While someone might have used an {{R from surname}}, that doesn't make it ambiguous. -- JHunterJ (talk) 21:49, 27 November 2010 (UTC)

This argument seems rather circular to me - surnames are not regarded as ambiguous because Wikipedia doesn't always list surname holders on disambiguation pages; Wikipedia doesn't always list surname holders on disambiguation pages because surnames are not regarded as ambiguous. But of course they are ambiguous in the real world - people are often referred to by their surname alone - why should we treat them as different from nicknames? (Someone could equally well start a page called "Fergie (nickname)" and move all the people with that nickname from the dab page to the new page - it wouldn't be any different from the surname case.)--Kotniski (talk) 10:35, 28 November 2010 (UTC)

I don't see the circularity. Some topics are ambiguous. Some are partial title matches. That's a fork, not a circle. And yes, consensus is that surname-holders are partial title matches, not ambiguous. -- JHunterJ (talk) 12:39, 28 November 2010 (UTC)

Can consensus make something unambiguous? Even if it can, though, surely there are exceptions - even Consensus must surely admit that "Einstein" and "Hitler" have Albert and Adolf as their primary meanings? --Kotniski (talk) 14:00, 28 November 2010 (UTC)

In general, people are sometimes referenced by last name only, although quality publications will usually give the full name at some point. And many people will search for a person by last name. Both of these generalizations are reasons for including people by surname on disambiguation pages, even though strictly speaking they are not ambiguous with the simple term. In cases where such a list grows unwieldy, it is common to shift the name-holders from the disambiguation page to a separate page. I think what JHunterJ is getting at is that apart from the generalizations above, there are cases where a person has become so commonly identified with the single name (either the surname, e.g., Einstein, or a given name, e.g., Elvis) that these are considered as primary topic for those terms, even though the article title may use a more formal name. Consensus can determine when a person is commonly known by only surname (in which case they should be listed on the disambiguation page, regardless of the existence of any separate surname page) and when persons are the primary topic for the term that is their surname. older ≠ wiser 14:29, 28 November 2010 (UTC)

Right. Einstein, Hitler, Patton, Madonna, Elvis, Cher, Fergie & Fergie, Beyonce, Shakespeare, and such are the exceptions -- they are commonly referred to by the single name only. Most name holders might use the first name or last name just as a pronoun -- as a placeholder for the common name. The consensus is that name holders are partial title matches in general, and that in some exceptional cases are actually ambiguous. If they are not actually ambiguous, then they are not the primary topic. The article on the surname (with a list of name holders) might be primary, if that anthroponymy article exists, or it might not be. -- JHunterJ (talk) 17:55, 28 November 2010 (UTC)

I have proposed Bungay, Suffolk, as the google search clearly shows Bungay, Suffolk is the prime topic, even if Freston or Cliburn isn't in this move Crouch, Swale talk to me My contribs 19:58, 2 December 2010 (UTC)

HLB

The Disambiguation page called "HLB" does not refer to the Citrus disease "Huanglongbing". The disease was previously called "Greening" but it was first described in China as "Huanglongbing". This name is not easy to pronounce, so we are used to say "HLB"

This disease is very important. It causes the decay of millions of orange trees in Brazil and Florida since five years. It is spreading through the US and has reached Georgia, Louisiana and South Carolina. The whole US orange production is threatened. There are many scientific references. —Preceding unsigned comment added by 147.100.149.242 (talk) 08:29, 29 November 2010 (UTC)

It wasn't on the dab page because there was no mention of this abbreviation in the article at Huanglongbing. I found a source which used the abbreviation, added it to the article, and added the abbreviation to the dab page. All done and dusted. PamD (talk) 10:08, 29 November 2010 (UTC)

Civil parishes, villages and hamlets

When we disambiguate two hamlets in the same district, we disambiguate by the civil parish that the hamlet is in, eg Rose Green, Lindsey and Rose Green, Assington, but what happens when we have a hamlet and a civil parish in the same district, eg Soulby, Kirkby Stephen and Soulby, Dacre, they are both in the Eden district, but Soulby, Kirkby Stephen is a civil parish near the town of Kirkby Stephen, while Soulby, Dacre is a hamlet in the civil parish of Dacre, Cumbria, there is also a similar situation with Newbiggin, Temple Sowerby and Newbiggin, Dacre. Crouch, Swale talk to me My contribs 13:30, 2 December 2010 (UTC)

We typically rely on the content-project's naming conventions (if any) to guide the selection of disambiguating titles (either in the "real" title or in a parenthetical qualifier), so Wikipedia:WikiProject Lancashire and Cumbria or its mother projects. I'm not sure what the problem here is exactly, though. It appears all the topics have been given unique names. Is there a problem with having Soulby, Kirkby Stephen and Soulby, Dacre? If so, perhaps Soulby, Eden (hamlet) and Soulby, Eden (civil parish), or Soulby (hamlet), Eden and Soulby (civil parish), Eden? BTW, I find the latter pairing ugly, but that's the pattern used in some other geographic projects. -- JHunterJ (talk) 13:59, 2 December 2010 (UTC)

Surname holders and partial title matches and disambiguation

General issue from Talk:Freston, Suffolk#Move request. There is only one article that could have the title "Freston", so (IMO) it's the primary or only topic. Surname holders (excepting those who are commonly referred to by the surname alone) are partial title matches per consensus at the disambiguation and anthroponymy projects. Partial title matches cannot be the primary topic for the partial elements of their titles (if they are actually ambiguous with the partial elements, they are not partial title matches, but again, this is not the case for surname holders in general). Instead, the anthroponymy article or list article (if one exists) might be primary, one of the ambiguous entries might be primary, or there might be no primary. If there is only one article that is not a partial title match, there is no ambiguity. A surname that does not also have ambiguous topics and no anthroponymy article or list article might redirect to the only notable holder as an {{R from surname}}, but if there are multiple notable holders, an anthroponymy list article (at least) would be needed. If instead there are topics that could have the name, the title should lead (directly or redirectly) to one of them (if there's only one, that one; if there's more than one, the primary one) or lead to a disambiguation page if there a ambiguous topics (possibly including an anthroponymy list article) and none are primary.

This is my understanding of the current consensus and guidelines of the disambiguation project and anthroponymy project. Other views on the general case are welcome here, or on the Freston case there in the move request. -- JHunterJ (talk) 13:46, 30 November 2010 (UTC)

As I just stated at the Freston discussion, I don't understand the basis for this at all. In particular: "Partial title matches cannot be the primary topic for the partial elements of their titles". Really? Why not? Why can't partial title matches be the primary topic for the partial elements of their titles? Are they not the primary topic by definition if readers use the partial elements of the title to seek the topic with the partial title match sufficiently often to meet the primary topic criteria? And what if readers don't search for such a topic with the term that is the partial title match often enough to make that topic primary, but often enough for the topic with the full title match to not meet the primary topic criteria? Why should partial title in general (or surname in particular) even be a consideration when determining primary topic? Should we be looking exclusively at likelihood of search, per the wording at WP:PRIMARYTOPIC? --Born2cycle (talk) 15:15, 30 November 2010 (UTC)

Because they aren't ambiguous. WP:PTM. -- JHunterJ (talk) 15:30, 30 November 2010 (UTC)

Potential outcome if we switch to "whatever the user is searching for with the term" vs. "whichever ambiguous topic is the most likely one": If the users searching for the automobile who misspell it "lexis" outnumber the users searching for either of the ambiguous topics actually ambiguous with "lexis", should we redirect Lexis to Lexus and put a hatnote there sending the smaller group to the automobile page, then the dab page, and then the sought page? My answer would be "No", we should leave the disambiguation page at the page name, and leave the non-ambiguous "Lexus" in the See also section of it. -- JHunterJ (talk) 15:43, 30 November 2010 (UTC)

Yes, the Lexus/Lexis example is a hypothetical problem with the way primary topic is defined, but I don't know that that problem has ever actually manifested itself. Are you suggesting the wording should be changed to address this hypothetical problem? What exactly do you propose?

I'm quite baffled by this argument. You just referenced WP:PTM, as if that is relevant here. Well, it does say, "Do not add a link that merely contains part of the page title, or a link that includes the page title in a longer proper name", but adds, "where there is no significant risk of confusion". First, that's talking about links on dab pages, not titles of articles or redirects. Second, I'm not sure what it means... so, the 22 persons with surname "Landis" linked at the Landis dab page should be removed because "there is no significant risk of confusion"?

Anyway, even with respect to links, it also states, "add a link only if the article's subject (or the relevant subtopic thereof) could plausibly be referred to by essentially the same name as the disambiguated term in a sufficiently generic context. For instance, the Mississippi River article could not feasibly be titled Mississippi, but it is included at Mississippi (disambiguation) because its subject is often called "the Mississippi".. Since any article whose subject is a person "could plausibly be referred to by essentially the same name as the disambiguated term" where "the disambiguated term" is the surname of the person, according to this a link should be added on the dab page for the term matching the surname, as is the case for everyone with surname "Landis" at Landis.

WP:PRIMARYTOPIC could be more clear, but I think it's sufficiently clear to show no basis whatsoever in considering partial title matches (including surnames) a factor in determining primary topic. First, it states how primary topic is determined for a given term: "[when] one of these topics [to which the term may refer] is highly likely – much more likely than any other, and more likely than all the others combined – to be the subject being sought when a reader enters that term in the Search box. If there is such a topic, then it is called the primary topic for that term. " Note that this definition makes no mention of title, much less states or even implies that somehow partial title matches should be a factor to consider.

Then, once primary topic is established per this definition, WP:PRIMARYTOPIC goes on to state, "the term should be the title of (or redirect to) the article on that topic." I highlight "(or redirect to)" because this clearly indicates that primary topic determination applies just as much to redirects to the topic, including redirects that are partial elements of the full title (as in how McNealy redirects to Scott McNealy), as it does to the article's actual full title. It even goes further to state, "The title of the primary topic article may be different from the ambiguous term being considered. This may happen when the topic is primary for more than one term, when the article covers a wider topical scope, or when it is titled differently according to the naming conventions [note: by convention, articles about persons are at [[Firstname Surname]]]. When this is the case, the term should redirect to the article (or a section of it). The fact that an article has a different title is not a factor in determining whether a topic is primary."

To interpret that last statement in terms of the Freston discussion, the fact that the article Tom Freston has a different title from "Freston" is not a factor in determining whether the topic at Freston, Suffolk is primary. --Born2cycle (talk) 17:07, 30 November 2010 (UTC)

I have to agree with User:Born2cycle on this. A reader coming to Wikipedia looking for the term "Freston" is as likely to be looking for Kathy or Tom (or more so) than the tiny village. That is ambiguous usage and disambiguation is appropriate. older ≠ wiser 23:59, 30 November 2010 (UTC)

This is arguing that the term is ambiguous, which is a reasonable argument. But just saying "that's their last name, they should be there" is a different argument. I don't know any of the three Frestons here (people or town), so I can't help decide whether what is said is true about Kathy and Tom being searched for as much as the town. In the general case, we want to keep the list manageable so we can't go with "its there last name, put them on the dab page"). Practicality would indicate - at some point - creating a name page and linking to it from the dab page and include on the dab page only those with the last name that are prominently known by that one name. --John (User:Jwy/talk) 00:10, 1 December 2010 (UTC)

You're quite right, but so much depends on context. In this case there are two relatively well-known authors, by no means mega-stars and neither have any claim on primary topic -- but in comparison with the tiny village where the primary claim to any notability other than bare existence appears to be a peculiar tower of some age (which arguably is more likely target for the term than the village). So far as I'm concerned this is a pretty clear-cut case that there is no primary topic. older ≠ wiser 00:27, 1 December 2010 (UTC)

Jwy wrote, "I don't know any of the three Frestons here (people or town), so I can't help decide whether what is said is true about Kathy and Tom being searched for as much as the town. ". We should not be determining primary topic based on personal experience or opinion. We need to look at the relevant facts... page view counts, ghits, incoming links, and common sense, like recognizing that searching for persons by surname, especially those with relatively unusual surnames, is a common practice. It's not a completely objective process, of course, but reasonable people should usually reach the same conclusion on these matters. When in doubt, there probably is no primary topic.
And of course the name is ambiguous. That's the point. It's unclear, ambiguous, whether "Freston" refers to Tom, Kathy or the village. --Born2cycle (talk) 04:07, 1 December 2010 (UTC)

Just for the record, by "I don't know any of the three..." meant that I didn't have any knowledge about any of the three subjects to have an opinion as to which, if any, should be primary. In the end, the choice is going to be an informed opinion - or consensus. --John (User:Jwy/talk) 06:50, 2 December 2010 (UTC)

If those really are the only three topics, I would suggest a good solution would be to title the village article Freston, and have a hatnote with links to the two people. That way everyone wins (or at least doesn't lose). In general, though, I would say that the fact that a topic isn't going to have its article titled using a particular term is a valid (though weak) reason for preferring some other topic as the primary topic for that term; but it becomes a stronger reason if no-one could even reasonably expect the article to be so titled (which I think is the case with both "Freston" for Tom, and "Fergie" for the Duchess). --Kotniski (talk) 07:14, 1 December 2010 (UTC)

There are references for the Duchess that refer to her as Fergie without any other names. There are no references for Tom that refer to him as Freston without any other names. One is commonly referred to by the single name alone (ambiguous), the other isn't (partial title match). OTOH, the fact that a topic isn't going to have its article titled using a particular term is not a valid reason for preferring some other topic as the primary topic; this is what we have the set of {{Redirect}} hatnotes for, so that Usa (Germany) isn't at usa. The determination is handled by the existing criteria, without regard for the titling, as long as the topics are ambiguous with the title. -- JHunterJ (talk) 12:29, 1 December 2010 (UTC)

I still don't see this difference that you so insist on between Freston and Fergie. They're both alternative ways in which the people in question can be referred to; I don't expect any serious reliable source to call either of them by those names without explaining who they mean (and if they do, then it's because one is much more famous than the other, rather than because one name is a surname and one is a nickname). I agree that Usa (Germany) shouldn't be at Usa, but that's because the difference in prominence between it and the USA is so overwhelming. If it were not so clear-cut, then other factors like the one I mentioned might well come into play.--Kotniski (talk) 13:32, 1 December 2010 (UTC)

If they do (and they do, or the New York Times isn't serious), you're right, it's because one is actually well known by the single name (ambiguous) and the other isn't (partial title match, not ambiguous, but as a convenience sometimes listed after the ambiguous entries on a disambiguation page in the absence of an anthroponymy article). People who are not so famous have a much harder time being known by a single name, true, so they are much less likely to be ambiguous with the single word. The other factors that you mention come into play regardless, and we can leave out the proposed factor of actual titling, since it has no impact on the outcome. -- JHunterJ (talk) 14:39, 1 December 2010 (UTC)

Is this a new New York Times link you've found? (The last one you showed me didn't meet the criterion - it was accompanied by an explanation of who they meant.) But I'm not so sure the factor of actual titling can be said to have no impact (isn't the English city allowed to be at plain Plymouth at least partly because our weird naming conventions would not allow Plymouth, Massachusetts to be there in any case? And isn't the case for making the singer the primary topic for Fergie at least somewhat strengthened by the fact that "Fergie" wouldn't be a conceivable encyclopedia article title for Sarah or Alex?) --Kotniski (talk) 15:12, 1 December 2010 (UTC)

Yeah, it met the criterion. Talking about the subject, like identifying an occupation or title, is different than using a full name. If you can find a corresponding "executive Freston" (without Tom) to go alongside "duchess Fergie" (without Sarah or Ferguson), then it would also meet the criterion. Our weird naming conventions still flex where needed -- Boston, Lincolnshire is not at Boston, and Boise and Greensboro redirect to their primary topics. If the primary topic of a placename in the UK and the US is the US place, but that place is not one of the ones that our weird naming conventions place at the bare name, then the bare name redirects to the place (like Boise or Greensboro) and navigational tools lead the other readers where they want to go. -- JHunterJ (talk) 16:17, 1 December 2010 (UTC)

J & K, honestly, what are you two going on about? Back and forth posts with no references to policy or guidelines whatsoever in any of your arguments amounts to I like it/don't like it rambling. Let's raise the bar, shall we?

The definition of primary topic is quite clear: "one of these topics [that are referred to by the term] is highly likely – much more likely than any other, and more likely than all the others combined – to be the subject being sought when a reader enters that term in the Search box.". The term Freston ambiguously refers to one of three topics in Wikipedia, the village, Tom and Kathy, and, based on page view counts, only Tom comes close to meeting the criteria. The idea that even a modicum of consideration should be given to the fact that "Freston" is a "partial title match" for "Tom Freston" is not only completely unsupported by WP:PRIMARYTOPIC or any policy, guideline or style guide, it is expressly contra-indicated in at least three statements of this guideline:

"... the term should be the title of (or redirect to) the article on that topic."
"The title of the primary topic article may be different from the ambiguous term being considered."
"The fact that an article has a different title is not a factor in determining whether a topic is primary."

Why are you ignoring all this?

As to the idea of putting one of the three ambiguous topics at Freston despite none meeting primary topic criteria, that might be worth considering, except that the topic meeting the criteria most closely is the one which should either be at Freston, or to which Freston should redirect, and that of course is Tom Freston. --Born2cycle (talk) 16:24, 1 December 2010 (UTC)

But how do we know? We can only know that if we know how many people are typing in "Freston" when looking for Tom Freston. And the reason we're talking without reference to the guidelines is (I thought) that we were vaguely considering the possibility that the guidelines might be improved. My opinion is that the fact that "Freston" could not be expected by anyone to be the title of Tom Freston's article is a factor. I mean that (this is going to be awkward to express) the value of n such that the knowledge that "n% of readers typing in "Freston" are looking for Tom" would preclude Freston, Suffolk from being the primary topic is (or should be) possibly greater than it would be if we had another place called Freston rather than a person called Freston. (uff, will anyone understand that? but I think they do intuitively apply it) --Kotniski (talk) 16:38, 1 December 2010 (UTC)

We know from the all-knowing oracle. Google searches are for more intelligent than doing simple pattern matches. Google results are sorted by information that is compiled based on how countless others search and what they ultimately click on. When we search for "Freston", especially with &pws=0 in the url to turn off individual and regional bias, the results reflect what others who searched with "Freston" ultimately clicked on, in order of popularity. Interestingly, that (with -wikipedia to remove WP bias) indicates that Kathy Freston is the primary topic. WP page view counts conflicting with google search results like that indicates a lack of primary topic; in any case the primary topic for "Freston" is definitely not the village, by any measure.

And I don't understand why you think "that the fact that 'Freston' could not be expected by anyone to be the title of Tom Freston's article is a factor". Are people even thinking about titles when they search for something? I sure don't. --Born2cycle (talk) 17:22, 1 December 2010 (UTC)

I think they are when on Wikipedia, at least, more than they are on Google, certainly. And if they're not, then they can't really expect to go straight to their target page, can they? I'm beginning to feel there's not a lot of point in discussing this at this kind of length - it doesn't matter a huge deal that we get every primary topic decision "right", or that different editors may have differing interpretations of "right" and consequently different factors may prove to have weight in different move discussions - it's not something we can legislate for in tiny detail. --Kotniski (talk) 18:01, 1 December 2010 (UTC)

So, you're saying that because people know neither Tom nor Kathy Freston will be at the title Freston that they are unlikely to search for either, at least within Wikipedia, with just "Freston"? I really doubt most people who use Wikipedia are anywhere near sufficiently familiar with all the nuances of how WP titles are constructed to take them into account when searching. You and I are rare wonks who know this stuff basically inside and out, but those of us with that kind of knowledge and understanding about titles make up a tiny percentage of all Wikipedia users. And even among us wonks, I, for one, can't recall ever thinking about titles when I construct my searches.

Readers are not supposed to figure out how WP titles work and construct their searches accordingly. Rather, editors are supposed to title articles consistent with how readers tend to do searches.

I agree we can have a philosophical discussion about what is "right" that would ultimately be pointless, but that's exactly why I've been trying to get us to talk about what the guideline says and means, and just follow that. I mean, it explicitly states, "The fact that an article has a different title is not a factor in determining whether a topic is primary."

Not a factor. That seems very clear. There's nothing to talk about, except whether to follow that, or ignore it for some good reason. I vote follow. You? --Born2cycle (talk) 23:29, 1 December 2010 (UTC)

Born2cycle, your elevation of the Google test to absolute authority isn't borne out. Note that Amazon is a disambiguation page and Hunter redirects to Hunting, despite the Google test's results that would make Amazon redirect to Amazon.com and Hunter a disambiguation page. I understand what you have put forth. You need to understand that it isn't the consensus. -- JHunterJ (talk) 01:52, 2 December 2010 (UTC)

By characterizing my position as elevating the Google test to "absolute authority" you indicate you most certainly do not understand what I have put forth. Either that, or you do understand but won't characterize it without using hyperbole. Either way, it's not helpful to finding consensus. --Born2cycle (talk) 02:11, 2 December 2010 (UTC)

Comment There is a similar move discussion going on at Cliburn (see Talk:Cliburn#Requested move), I have created Cliburn (disambiguation) so that Cliburn does not need to be moved, also Tony Blair isn't refered to as Blair, is he Crouch, Swale talk to me My contribs 13:19, 2 December 2010 (UTC)
- There are so many other uses of "Blair" among which at least one must receive enough traffic, or all of them combined receive enough traffic, to make Tony Blair not be the primary use of "Blair", per the criteria at WP:PRIMARYTOPIC. That same criteria indicates that the pianist Van Cliburn is the primary topic for "Cliburn", and, so, per that guideline Cliburn should redirect to Van Cliburn, just like, for example, McNealy redirects to Scott McNealy. --Born2cycle (talk) 16:31, 2 December 2010 (UTC)
  - I have proposed Bungay, Suffolk, as the google search shows it is clearly the prime topic, even if Cliburn and Freston aren't. Crouch, Swale talk to me My contribs 19:48, 2 December 2010 (UTC)

I've lost track what the global discussion is here. If I'm the only one, let me know and I'll just shut up. Are we trying to refine the global policy about whether/when surnames show up on dab pages or are we discussing specific instances? If specifics, it doesn't belong here but on the particular talk pages. If we are trying to refine the policy, could we summarize the positions? One extreme would seem to be "All people with the dab term in their name should appear on the dab page" and the other extreme is "Since the surname is a partial match, it should never appear on the dab page." I don't think anyone is arguing for these extremes, but I'm not sure where people are in the middle. A summary of the positions might be useful, perhaps in a new section.

An interesting and possibly productive exercise would be for JHunterJ and born2cycle attempt good faith summaries of each others position? --John (User:Jwy/talk) 16:49, 2 December 2010 (UTC)

As I understand it, it's about primary topic decisions rather than dab pages, and the question is whether the idea now expressed at WP:PRIMARYTOPIC by the words "[likely] to be the subject being sought when a reader enters that term in the Search box" should cover (a) any string which a reader might type into the search box in his endeavour to find an article on a subject (B2C's position), (b) any string which is commonly used as a "self-contained"(?) name referring to the sought subject (JHJ's position), or perhaps (c) any string which could be reasonably conceived as the title of an encyclopedia article on the sought subject (my position). Though I don't think any of us holds any one of these positions absolutely categorically.--Kotniski (talk) 11:10, 3 December 2010 (UTC)

(b), if my position, is that the primary topic must actually be ambiguous with the title for which it will be primary. This is, IMO, the same as (c), in that Albert Einstein could be the primary topic of "Einstein" since he is actually ambiguous with "Einstein"; Van Cliburn could not be the primary topic of "Cliburn" (nor of "Kilgore prodigy pianist"), no matter what those bare Google searches say, since he is not actually ambiguous with "Cliburn" or "Kilgore prodigy pianist"; and since there is only one "ambiguous" topic for "Freston", it is the primary topic of "Freston". In other words, Cliburn could not be reasonably conceived as the title for the article on Van Cliburn. My position does lean on the earlier consensus (which has be called into question by at least Bkonrad) that lead to the formation of the Anthroponymy project, that name holders in general are partial title matches, therefore not mixed with the disambiguation entries and preferably split from the dab page to an anthroponymy list article about the name if an editor will create it. The list article then might or might not be the primary topic, but none of the general name-holders could be -- only exceptional cases where the person is commonly referred to only by just the single name. Then he'd be ambiguous and possibly primary. -- JHunterJ (talk) 14:14, 3 December 2010 (UTC)

So my position is effectively very similar to that, except that we seem to disagree about cases like "Fergie", which I believe could not reasonably be conceived as the title for an encyclopedia article about Sarah or Alex Ferguson any more than the plain "Ferguson" could.--Kotniski (talk) 14:31, 3 December 2010 (UTC)

So we might disagree on some of the case-by-case application of the guidelines, but not on the guidelines themselves. And we don't have to hash out the Fergie case here. :-) -- JHunterJ (talk) 14:51, 3 December 2010 (UTC)

I agree with Kotniski that this discussion is about how primary topics are determined (both what the guideline says and what it should say) and not about what is linked or not on dab pages. I also generally agree with Kotniski's formulation of the positions above except he has his terms and topics reversed.

Primary topic is about which one, if any, among several topics is primary for a single given term. Primary topic determination is never about "any string which a reader might type", but about only one specific string: the term in question (like "Fergie" or "Cliburn"), and how likely it is that a reader might type that term to search for each of the topics to which that term might refer.

So instead of:

The question is whether the idea now expressed at WP:PRIMARYTOPIC by the words "[likely] to be the subject being sought when a reader enters that term in the Search box" should cover (a) any string which a reader might type into the search box in his endeavour to find an article on a subject (B2C's position), (b) any string which is commonly used as a "self-contained"(?) name referring to the sought subject (JHJ's position), or perhaps (c) any string which could be reasonably conceived as the title of an encyclopedia article on the sought subject (my position).

I would say:

The question is whether the idea now expressed at WP:PRIMARYTOPIC by the words "[likely] to be the subject being sought when a reader enters that term in the Search box" should cover (a) any topic to which the term may refer (B2C's position), (b) any topic for which the term is commonly used as a "self-contained"(?) name (JHJ's position), or perhaps (c) any topic for which the term could be reasonably conceived as the title of an encyclopedia article on the sought subject (Kotniski's position).

Further, I suggest that (a) is the only reasonable interpretation of those words on this page, is supported by other statements on this page (like, "The fact that an article has a different title [different from the term in question] is not a factor in determining whether a topic is primary"), and challenge positions (b) and (c) as having no basis whatsoever on this page or any other policy or guideline (except WP:IAR, perhaps, but even that requires "good reason" to make the encyclopedia better, which I have not seen here), and so amount to being rationalizations of personal preference. I mean, the qualifications in (b) and (c) are, as near as I can tell, pulled out of thin air (not from any guideline). Because editors of Kotniski's and JHJ's stature are not clear about this, I seek wording on this page that makes it even more clear that what is meant here is (a).

Without bringing clarity on this point to WP:PRIMARYTOPIC the encyclopedia is worse off in two ways:

Readers are not served as well as they can be in terms of how often searches take readers directly to the topics they are seeking (the whole point of WP:PRIMARYTOPIC).
Discussion involving primary topic are more contentious than they need to be due to being turned into popularity contests (in terms of popularity of topics as being "primary" for the editors involved, whatever "primary" means to each editor, rather than about how popular the topics are in reader searches with that term) mostly characterized by expressions of personal preference based on who-knows-what. I don't mean to imply that being more clear about (a) in this guideline is going to completely end debates about primary topic in requested move discussions, but I suggest it will move us a significant distance in that direction.

I might be wrong, so I'm also open to a change in wording that clarifies (b) or (c) as the actual guideline, but I really think those interpretations are so inherently vague that any such change would make primary topic discussions only more contentious, and can only lead to poorer search results for readers relative to (a). But, again, I might be wrong, so if there is a way to convey (b) or (c) that will make primary topic determinations less contentious, and will improve search results for readers, I'm open to it. --Born2cycle (talk) 16:42, 3 December 2010 (UTC)

Let's also be clear: I'm quite clear about this, and I've been clear about it, and my clearly stated view happens to disagree with yours. That does not mean the "JHJ is not clear about this." The qualifications in (b) and (c) are pulled from the guidelines. OTOH, the view that the guidelines about disambiguating ambiguous topics switch from "ambiguous topics" to "any topics that might be sought by a search term" for the purposes of determining primary topic appears to be pulled from thin air, and is contrary to the guideline. -- JHunterJ (talk) 19:05, 3 December 2010 (UTC)

No, you haven't been clear. This isn't about opinions or disagreements. It's about whether the positions expressed are supported by quotable words from guidelines or not. My position is so supported (which I redo again below); yours is not (at least you haven't provided such supportive guideline wording for your position - you are free to prove me wrong by citing the diffs).

"The qualifications in (b) and (c) are pulled from the guidelines." Excuse me, but I've asked repeatedly for this and you have not provided any support for this claim whatsoever. I ask again... Please indicate in which guideline and with which specific words the guideline indicates that the topics considered for WP:PRIMARYTOPIC should be limited to only those topics "for which the term is commonly used as a "self-contained"(?) name" (qualification (b)) or "for which the term could be reasonably conceived as the title of an encyclopedia article on the sought subject" (qualification (c)). If you can't do that, then please stop claiming these qualifications are "pulled from the guidelines."

Now, let's look at what you claim is pulled out of thin air: "the view that the guidelines about disambiguating ambiguous topics switch from "ambiguous topics" to "any topics that might be sought by a search term" for the purposes of determining primary topic appears to be pulled from thin air. Thin air? Modeling the kind of answer I expect from you, here is specific wording from actual guidelines supporting this view.

When talking about article title, WP:TITLE states that "article titles are based on what reliable English-language sources call the subject of the article". Limiting the scope to what the subject is called is a characteristic of title determination; it's not a characteristic of disambiguation in general or determining primary topic in particular.
When talking about disambiguation: Wikipedia:Disambiguation#Deciding_to_disambiguate states: "Disambiguation is required whenever, for a given word or phrase on which a reader might search, there is more than one existing Wikipedia article to which that word or phrase might be expected to lead.". Note the lack of qualification, explicit or implied, with respect to "for a given word or phrase on which a reader might search".
Also at WP:Disambiguation, WP:PRIMARYTOPIC specifically defines determining primary topic exclusively in terms of search likelihood: "the subject being sought when a reader enters that term in the Search box." Note again that there is no qualification on "the subject being sought"; in particular, no qualification implying much less clearly stating that the toipc needs to be an "ambiguous topic" narrowly defined as "a topic for which the term is commonly used as a self-contained name".
Again distinguishing from title determination (which, again, is based on what subjects are normally called), WP:PRIMARYTOPIC also states: "If a primary topic exists, the term should be the title of (or redirect to) the article on that topic.". Since titles reflect what subjects are normally, the recognition that the term in question for a primary topic could be a redirect clearly rests on the assumption that the term for which a topic is primary is not necessarily "commonly used as a self-contained name" for the topic.
WP:PRIMARYTOPIC also states: "There are no absolute rules for determining how likely a given topic is to be sought by readers entering a given term;" Note how no qualifications for "a given topic" are stated or even implied, except that it might be sought be readers entering the given term.
WP:PRIMARYTOPIC even states: "The title of the primary topic article may be different from the ambiguous term being considered. ... When this is the case, the term should redirect to the article". Again, the recognition that the term in question for a primary topic could be a redirect clearly rests on the assumption that the term for which a topic is primary is not necessarily "commonly used as a self-contained name" for the topic.
And finally, "The fact that an article has a different title is not a factor in determining whether a topic is primary". How much clear can "not a factor" be?

These plain words from actual guidelines clearly support the view that "any topic that might be sought by a given search term" is the primary topic if it meets the criteria are not pulled out of thin air; whether the topic also meets qualifications (b) or (c) is not relevant.

Now, can you provide specific wording from guidelines at all, much less this much, which provides support for your view that the the topics considered for primary topic (not for consideration as a title or for listing on a dab page - those are separate issues) should be limited to only those topics subject to qualification (b)? --Born2cycle (talk) 20:04, 3 December 2010 (UTC)

And the other part of the disconnect is your approach that treats the guidelines as legal ordinances, seeking chapter and verse to justify positions. They are Wikipedia guidelines to assist Wikipedia editors (not lawyers). The guidelines in question are specific to ambiguous topics, which is my clear support. -- JHunterJ (talk) 20:10, 3 December 2010 (UTC)

I'm as big a supporter of WP:IAR as anyone I've encountered. Characterizing my view as treating "guidelines as legal ordinances" apparently because I quote from them in discussions about what they say is ludicrous.

Anyway, These are your words: "The qualifications in (b) and (c) are pulled from the guidelines". Back that up, or retract it. --Born2cycle (talk) 20:29, 3 December 2010 (UTC)

Please quit barking orders. You are not in charge here. -- JHunterJ (talk) 20:41, 3 December 2010 (UTC)

I'm sorry. I thought it was understood that I'm not in charge and that nothing I say could possibly be an order. Why would you interpret it that way? It should be obvious that the point is that you're making claims about ideas being "pulled from the guidelines" when they're actually pulled from thin air.

Anyway, since the only way to address that is to concede the point and that seems to be highly unlikely, let's go on to your most recent claim. Even if we accept the view that "the guidelines in question are specific to ambiguous topics", given that...

persons are regularly referred to by surname (for example, WP:SURNAME actually requires persons to be referred to by surname only throughout an article, except at the initial mention, as is common practice in English),
WP:D defines "disambiguation" as "the process of resolving the conflicts that arise when a single term is ambiguous, and so may refer to more than one topic which Wikipedia covers",
ambiguous means "can be understood or interpreted in more than one way"
one of the ways a term which is also a surname "can be understood or interpreted" is as a reference to a person with that surname.

... how do you define "ambiguous topic" such that a person "which Wikipedia covers" is not an "ambiguous topic" for the ambiguous term that is that person's surname? I mean, you keep declaring that persons are not "ambiguous topics" for their surnames as if it is obvious, but what is the source of that definition of "ambiguous topic"? What exactly is that definition? --Born2cycle (talk) 21:37, 3 December 2010 (UTC)

I interpret it the way it is written. "Back that up, or retract it." is an order. If you understand that you're not in charge, the question isn't "Why would you interpret it that way?" but rather "Why would you write it that way?". The "only" way to address the disconnect is through consensus, which is why I'm going out of my way to avoid repeating myself over and over again, so that other editors can read what we've each written and also make their opinions heard. I feel comfortable in my clarity, since all but one editor appears to understand what I've said, whether or not they agree with it. Once we find the new consensus, which may or may not be the same as (my understanding of) the old consensus, we can try to make the guidelines clearer so that they reflect the consensus. -- JHunterJ (talk) 23:10, 3 December 2010 (UTC)

In retrospect it wasn't an ideal choice of words, which is why I apologized. But the reason I wrote it that way is because it's something I would have said had we been having a conversation. The reason it isn't ideal in this forum is because I couldn't convey with voice intonation and facial expression what I would have had we been having a conversation.

I seriously doubt anyone is reading your words as closely as I am, and I'm really trying to understand. But your use of "ambiguous topic" remains undefined, your claim that qualifications (b) and (c) are "pulled from the guidelines" remains unsubstantiated, and your assertion that persons in general are not "ambiguous topics" for their surnames remains unexplained. --Born2cycle (talk) 23:23, 3 December 2010 (UTC)

Lest you think I'm conceding, I disagree that they remain undefined, unsubstantiated, or unexplained. I also seriously doubt that many are slogging through this vast expanse of text (here and elsewhere), which is why I am not repeating it every time you re-ask the question. -- JHunterJ (talk) 23:45, 3 December 2010 (UTC)

Yeah, a vast expanse of you claiming you've already answered questions you have never answered. If you had actually already answered, you could easily link to your supposed answer (as you did when you corrected me for incorrectly claiming goalpost movement). Just to take one example since the entire discussion is immediately above this, Kotniski's a, b, c characterization of our positions is new today and just above, and so is your claim that qualifications b and c are "pulled from the guidelines", as are my questions about that assertion. Still, no where between Kotniski's post and here is there an answer to my request for substantiation in guideline wording of that claim. --Born2cycle (talk) 00:11, 4 December 2010 (UTC)

A vast expanse of you claiming I've never answered questions I have answered. Kotniski was able to synthesize the a, b, c based on the answers you claim I never gave. And my response to him apparently resolved any lingering lack of clarity he may have had. And, to repeat myself, the guidelines are the guidelines for disambiguating ambiguous topics, therefore the guidelines on determining a primary topic from among the ambiguous topics are for determining a primary topic from among the ambiguous topics, not from among the search result leaders. -- JHunterJ (talk) 01:39, 4 December 2010 (UTC)

I just reread your response to Kotniski and have the following comments/questions.

You say, "the primary topic must actually be ambiguous with the title for which it will be primary." I presume that what you mean by "topic must actually be ambiguous with the title" is that "the topic must be one of several topics to which the ambiguous term which is the title might refer." Is that what you mean? If not, what do you mean?
If that is what you mean, then how is a person not a topic which is one of several topics to which the ambiguous term which is the person's surname might refer? Is not the last U.S. President ambiguous with Bush? Is not the American inventor of the lightbulb ambiguous with Edison? Is not the founder of General Motors ambiguous with Durant?
You say that Van Cliburn is not ambiguous with Cliburn. Do you not agree that Cliburn might refer to Van Cliburn? WP:D states in the first sentence that it applies to "[ambiguous terms that] may refer to more than one topic which Wikipedia covers." We know from Cliburn (disambiguation) that it "may refer to more than one topic which Wikipedia covers", and, we know from all the topics listed on that page that one of those topics is Van Cliburn. How is Van Cliburn not ambiguous with Cliburn?
You also say that "Cliburn could not be reasonably conceived as the title for the article on Van Cliburn." Fine, but what does that have to do with whether Van Cliburn is ambiguous with Cliburn, and, more importantly, what does it have to do with whether determining the primary topic for Cliburn, especially considering WP:PRIMARYTOPIC explicitly states: "The fact that an article has a different title is not a factor in determining whether a topic is primary."
You also say "that name holders in general are partial title matches, therefore not mixed with the disambiguation entries"... Well that just means that names of persons who are ambiguous topics with the term of that dab page are listed in a separate section on the dab page, not omitted from the dab page, unless there are so many of them that they are sub-articled out to special surname-only article. Just being listed on the dab page of a term (or sub-articled from it) clearly indicates the name holders are indeed ambiguous topics for that term in every manner that could possibly be relevant here. No?
You also say, "The list article then might or might not be the primary topic". No a list article can never be a primary topic - a list article has no topic. If a list or dab article is at the base name -- not disambiguated with (disambiguation) or (surname) or whatever -- that means there is no primary topic, not that the surname or dab page is the primary topic. I think you're again conflating the notion of a primary topic with the article that is located at the base name.
You also say "none of the general name-holders could be [the primary topic] ... -- only exceptional cases where the person is commonly referred to only by just the single name [are ambiguous]". This again seems to presume, inexplicably, that the term in question not being reasonably conceived as the title for the topic in question means that topic cannot be an ambiguous topic for that term (which means that term cannot refer to that topic which is absurd).

In short, if an "ambiguous topic" is any topic to which WP:D applies, then persons are "ambiguous topics" for their surnames whenever the surname is an ambiguous term since the first line of WP:D states that it applies to terms "that may refer to more than one topic", and surnames definitely are used to refer to persons who have those surnames. --Born2cycle (talk) 02:38, 4 December 2010 (UTC)

Confusingly similar titles

Question; is it useful to proactively add hatnotes to linking articles with potentially confusingly similar titles to each other? For example, below are listed pairs of similarly title U.S geographical articles. Is it of general benefit to addTemplate:About hatnotes to them? - TB (talk) 16:09, 1 December 2010 (UTC)

I'd use {{distinguish}} in those cases. -- JHunterJ (talk) 16:18, 1 December 2010 (UTC)

Thanks. I've listed the 1500 closest matching titles that do not contain links to each other here. Many look like good candidates fort his treatment. - TB (talk) 22:26, 2 December 2010 (UTC)

Ah, one more possibly useful template: for cases like Linguistic Imperialism vs. Linguistic imperialism (that differ by caps or punctuation, rather than by spelling), I'd probably use {{for}} rather than {{distinguish}}. -- JHunterJ (talk) 14:18, 3 December 2010 (UTC)

Ambiguous topics

Can someone explain exactly what ambiguous topics are and how they relate to disambiguation? Now if we are talking about ambiguous page names it could make sense. Vegaswikian (talk) 21:23, 4 December 2010 (UTC)

I've searched for all uses of the nonsensical phrase in the talk pages. It's funny what a large percentage are attributed to one particular editor. Leaving those uses aside, most other people either use it to mean "ambiguous topic name" (with ambiguous modifying "topic name" rather than "topic" - which makes sense), or to mean any one of multiple topics to which an ambiguous term may refer. That's not English, but it seems to be used by a small number of editors to mean that, and everyone seems to understand it well enough. But it's not English, and semantic errors like that should not be in our guidelines like WP:PRIMARYTOPIC. I've already fixed it once, but it has been reverted back to non-English. --Born2cycle (talk) 21:50, 4 December 2010 (UTC)

I don't mean to engage in an edit war, but once an English error like that is identified it can't stay in the guideline. --Born2cycle (talk) 21:55, 4 December 2010 (UTC)

In most cases ambiguous topic is equivalent to ambiguous page name. But not all ambiguous topics with articles use the ambiguous term as their article titles (for example, Lysergic acid diethylamide is not itself ambiguous, but it can be the intended target for an ambiguous use of the term Acid, and is listed at Acid (disambiguation) under the redirect Acid (drug)). And not all topics have separate articles (for example, a book mentioned in the author's article, or a TV episode discussed in a list article). So the term ambiguous topics is used to encompass all ambiguous uses that might be covered, whether or not they have ambiguous article titles or articles at all.--ShelfSkewed Talk 22:04, 4 December 2010 (UTC)

When you say, "an ambiguous use of the term Acid", don't you mean "a use of the ambiguous term Acid"?

Merriam-Webster defines "ambiguous" as "capable of being understood in two or more senses or ways", so "ambiguous topics" is a misuse of the term "ambiguous". It's the term "Acid" that is capable of being understood, not the topic of the article at Lysergic acid diethylamide, and so the topic cannot be an "ambiguous topic" of "acid" or anything else. "Ambiguous topic" is nonsensical. That topic is, however, one of the uses of the ambiguous term acid. --Born2cycle (talk) 22:30, 4 December 2010 (UTC)

Okay, "a use of the ambiguous term Acid", if you wish. Or I might have said, more to the point, "an ambiguous linking of the term Acid", i.e. a link that goes to a disambiguation page or to the primary topic. I wasn't trying to argue against your phrasing of the guidelines; I was just trying to clarify for Vegaswikian why a more encompassing term than ambiguous page name was used. If you prefer ambiguous topic name, I have no problem with that.--ShelfSkewed Talk 22:46, 4 December 2010 (UTC)

While ambiguous topic name could be an improvement over what has been in the guideline for a while, I still view it as problematic. Page names for articles should not be ambiguous. Yes we need some exceptions, but the guideline needs to make it crystal clear that unambiguous is the right way with limited exceptions. Vegaswikian (talk) 23:02, 4 December 2010 (UTC)

Just to clarify... those only exceptions being when one of the meanings of the ambiguous term in question meets primary topic criteria, right? --Born2cycle (talk) 23:17, 4 December 2010 (UTC)

I'd expand that to be a vital article as someone pointed out today or the article clearly meets the criteria in primary topic. Vegaswikian (talk) 23:31, 4 December 2010 (UTC)

Okay, right. There are so view vital articles I forget about them. but yeah, an ambiguous article name implies that use meets primary topic criteria, or it's a vital article.

By the way, that's another wording change I made today. I changed:

In such a case, consensus may determine that the vital article should be treated as the primary topic regardless of whether it is the article most sought by users.

to:

In such a case, consensus may determine that the vital article should be treated as if it is the primary topic (the term is made the title of, or redirect to, the vital article) for the term regardless of whether it is the article most sought by users.

The point I'm trying to stress is placement at the basename does not mean it is the primary topic. For example, if some rock star who went by the name "Art" became super famous, that topic might arguably meeting primary topic criteria and thus be the primary topic, but the vital article, not the primary topic, would remain at Art. So the vital article would be treated (in terms of being at the basename) as if it is the primary topic. But that wouldn't make it the primary topic. --Born2cycle (talk) 23:46, 4 December 2010 (UTC)

I agree some of the terminology is sloppy, but not sure I agree with your solutions. Its possible that we could define "primary topic" as the article that gets treated such - i.e. that the article on the page the disambiguation term takes the reader to. Then are discussion are not "which article is primary topic" but "which article shall we choose as primary topic." In that case, "Art" would be the primary topic in any event. But in practice, I have not found this subtle issue to be enough of a problem to get too worked up about it. For some of the other terms, we might want to work out consistent terminology for the following concepts (and what I generally call the concepts):

The term being disambiguated - the Foo of Foo (disambiguation) for example. I use DAB term'.
Articles that are linked to from the dab page (I think this is what ambiguous topics is intended to mean). I think I use target articles.

There are probably other concepts it would be useful to have a common term for. Perhaps a small glossary? --John (User:Jwy/talk) 02:33, 5 December 2010 (UTC)

Ambiguous topics are topics of articles that might have titles that could also be the titles of articles of the other ambiguous topic(s). -- JHunterJ (talk) 02:36, 5 December 2010 (UTC)

I don't think we should be talking about "ambiguous topics" - that would mean topics that have more than one meaning, and topics don't have meanings, so the phrase seems entirely illogical. Let's keep it clear and talk about "ambiguous terms" - it's terms that have meanings.--Kotniski (talk) 11:04, 5 December 2010 (UTC)

While I agree we should probably avoid ambiguous topics, I don't think ambiguous terms is good here as I think terms would mean the title of the article, not the content. I think when ambiguous topic has been used (as in HJH's comment 02:36, 5 Dec above) it is the articles that is being talked about, not the title. The title is changeable and, I think, irrelevant to the disambiguation process. That's why I use target article(s). I use "target articles" for the articles that we are trying to help the readers find when they enter something ambiguous. No matter what the title, if its likely the reader will enter the disambiguation term to find a given article, it doesn't matter what the title is. --John (User:Jwy/talk) 16:25, 5 December 2010 (UTC)

I agree with everything you say, except that "ambiguous terms" should be avoided because terms would mean the title. Term is far more general in meaning than that. It could be a title, it could be the location of a redirect, or it could be neither... just whatever someone enters in a search. So I think "ambiguous term" is absolutely correct, as "term" is used to mean in the initial phrase of WP:PRIMARYTOPIC: "Although a term may refer to more than one topic". That is, a term that may refer to more than one topic is an ambiguous term, and any one of those topics may be the primary topic for that term (or there may be no primary topic for that term).

Whether that term is a title or redirect is a separate matter. The main point is: first we determine whether there is a primary topic for a given term, only when we have established a primary topic do we decide if the article about that primary topic should be at that term as its title, or that that term is a redirect to that primary topic article.

A related point is that when we determine which if any of the topics to which a given term may refer is primary, that we look at the relative likelihoods of readers entering that term to seek each of these topics.

Another point is that when determining primary topic we have many factors to consider when deciding whether a given topic meets the criteria ("much more likely than .."), and, there seems to be some disagreement about how much weight should be given to readers taking into account whether a given term is likely to be the Wikipedia article title of the topic's article they seek when deciding what to enter in the search box. --Born2cycle (talk) 17:05, 5 December 2010 (UTC)

(ec) I think perhaps there isn't any disagreement here - Jwy may just have been pointing out that "ambiguous terms" is not synonymous with "ambiguous topics", which I carelessly implied in what I wrote (I wanted to say that the adjective "ambiguous" should be applied to terms rather than topics, not that the words "ambiguous topics" should be directly replaced by "ambiguous terms").--Kotniski (talk) 17:20, 5 December 2010 (UTC)

Oh, I didn't get that implication from what you wrote, but I see now how Jwy might have. Yes, for each ambiguous term there are multiple topics to which that ambiguous term may refer, and it's sloppy/illogical/nonsensical to refer to those topics as "ambiguous topics", but of course "ambiguous topic" is not a reference to the term itself. After the last week or so I'm not accustomed to all this agreement to what I've been trying to say. --Born2cycle (talk) 17:30, 5 December 2010 (UTC)

I just boldly reworded two sections and the intro of this page using the term homograph. The intended meaning of the words has not changed, but I think using homograph conveys it more clearly. Thoughts? --Born2cycle (talk) 19:36, 6 December 2010 (UTC)

I don't like it. Who knows what a homograph is without having to click through? Its hardly standard everyday English. Terms, titles, topics, whatever, but not homographs. Tassedethe (talk) 22:07, 8 December 2010 (UTC)

I know, and it concerns me too. However, it easy to understand once you click on it, and is exactly and specifically what we're dealing with on this page. "Terms, titles, topics, whatever" are all different or are insufficiently precise in meaning, and I think can lead to a lot of confusion and misunderstanding about what the words using this terminology mean. I realize the changes I've made could benefit from more wordsmithing, but I also think they are already less likely to be misunderstood. For example, I often see people, even experienced editors and admins, conflate "topic" and "title" with "the ambiguous term". No one will confuse "homograph" with either of these.

The problem with using more familiar but less specific terminology is everyone thinks they understand what it means, but many understand the same words differently. That's why more specific jargon has to be used in every field I can think of, from sports to engineering, from law to fashion, from medicine to acting. The same is true here, this is just a new one to this context, and so initial ambivalence or worse about its use is to be expected (frankly, I'm surprised it's taken someone so long to object). But please give it a chance, I think you'll find it works quite well. --Born2cycle (talk) 22:33, 8 December 2010 (UTC)

Jargon is often used in specific fields but in an endeavour such as this, which will include many people who's first language is not English, or who have varying levels of education, it should be avoided. Medicine is a good example of a field where precision is often deliberately lost for the sake of clarity and understanding (by non-medics). In all my time looking at dab pages I have yet to meet anyone who has added an item to a dab page based on a misunderstanding of term or topic or title. And certainly noone who would have been enlightened if I started talking about homographs. Tassedethe (talk) 23:27, 8 December 2010 (UTC)

Certainly noone who would have been enlightened if you started talking about homographs? How can you be certain? --Born2cycle (talk) 23:46, 8 December 2010 (UTC)

The wisdom of my years, or the arrogance of youth, depending on whether I am older or younger than you. :) Tassedethe (talk) 00:03, 9 December 2010 (UTC)

Can one of the meanings of an ambiguous term be a person whose surname is that term?

A specific example is being discussed here. Basically, I've added the following hatnote to Cliburn to help readers who search for the pianist Van Cliburn by just his surname:

This article is about a village in Cumbria County, England. For the pianist, see Van Cliburn. For other uses, see Cliburn (disambiguation).

This hatnote keeps getting replaced with a more generic one:

For other uses, see Cliburn (disambiguation).

The two reverts were accompanied by the following comments in edit summaries:

there is no pianist "Cliburn", only one "Van Cliburn" [2]
name holders not ambiguous, see MOS:DABNAME and this Talk page [3]

Since persons are commonly referred to by surname, the assertion that "there is no pianist 'Cliburn'" is absurd. Google for "pianist Cliburn" and see what you get! I suppose it could mean, "There is no pianist with a claim on the title 'Cliburn'", but if there were no other uses of the term "Cliburn", Cliburn could very well be a redirect to the article about the pianist, just as McNealy is a redirect the only use of that name in WP, a person with that surname.

The assertion that "name holders are not ambiguous" is also absurd. I've asked for explanation here, but it has been refused unless someone else asks.

Prior to the relatively recent creation of Cliburn (currently an article about a small village that gets a few hundred views per month), there was no article there, so anyone searching for the pianist by surname would immediately find the link to Van Cliburn (an article with thousands of views per month) and could click on it. Now, the same user will find himself at the article about the village with the hatnote to the dab page, so he will have to click on that, and then click on the appropriate link on the dab page, before getting to the wanted article.

I see nothing at MOS:DABNAME (which addresses how to format dab pages) or Talk:Cliburn which explains why a popular article about a person should not be linked in a hatnote of the article whose title is the person's ambiguous surname.

It seems blatantly obvious to me that one of the meanings of an ambiguous term can be a person whose surname is that term, but this is the very point being disputed, even in the context of one the most famous American pianists whose surname is an ambiguous term that does not mean anything else remotely as popular (in terms of page views and ghits).

Any help to resolve this would be much appreciated. --Born2cycle (talk) 23:10, 4 December 2010 (UTC)

How many more splinters of this thread are you going to start? Most name holders are not ambiguous with either their given name(s) or their surname. Exceptions for people who are widely known by just a single name only are made (e.g., Einstein, Shakespeare, Patton, Madonna, Elvis). In the absence of and until the creation of an anthroponymy article for the name, short lists of name holders might be included on disambiguation pages, after the list of ambiguous topics (MOS:DABNAME). So, yes, one of the topics (not meanings) for an ambiguous title might be a person whose surname is that title, but only in the exceptional cases where the person is referred to by just that name only. -- JHunterJ (talk) 02:33, 5 December 2010 (UTC)

It has been over a week since I first started pointing out to you, eventually in multiple locations and multiple discussions, that the phrase "ambiguous topic" is nonsensical. You ignored every word I said about that. Not until I edited the guideline accordingly (and you immediately reverted) and others jumped in (see section above) did you finally seem to concede.

However, you are still even with this post missing the larger point... "ambiguous" is an adjective that can only apply to concepts that are interpreted, like terms, words, phrases, facial expressions, tones, etc., yet above you say, again, "name holders are not ambiguous". Name holders are not interpreted. Name holders cannot be ambiguous. That phrase is just as nonsensical as is "ambiguous topic". Is it going to take another week for you to get that too?

Can you please figure out how to state your position without using the word "ambiguous" in a nonsensical context in which it modifies something that is not interpreted? If you did that in our first discussion about Freston there would have been no need for all these splinters. Since you seem to be ignoring what I say no matter how reasonable I am, you leave me with little choice but to enlist assistance, which is why I start these splinters.

Now, I realize that this might seem like a pointless semantic exercise, but it has important consequences. For example, now that the nonsensical phrase "ambiguous topics" has been replaced with "these topics" at WP:PRIMARYTOPIC, you can no longer claim that primary topic determination applies only to topics that meet some poorly conceived notion of what qualifies as an "ambiguous topic". Instead, "these topics" can only reasonably refer to the topics referenced in the previous paragraph. That paragraph, the first one at WP:PRIMARYTOPIC, starts with, "Although a term may refer to more than one topic, ...", therefore "these topics" can only mean "the topics to which the term may refer".

This distinction gets to the heart of our disagreement. You want it to say the nonsensical "ambiguous topic" so you can define it as excluding persons whose surname is the term in question, but when it says "these topics" you are forced into a position of saying that surnames are not used to refer to persons with that surname, which is absurd, and you know it. And please don't bring up titles again, because determining primary topic for a given term has nothing to do with titles, something else I've been pointing out repeatedly and you've been ignoring. --Born2cycle (talk) 16:42, 5 December 2010 (UTC)

It does seem that in this case (whether or not this situation is "exceptional") the single word "Cliburn" is very often used to refer to the pianist, so I think the explicit mention in the hatnote is justified. --Kotniski (talk) 17:14, 5 December 2010 (UTC)

Thank you for the confirmation of my point and showing why I have to start these splinters. --Born2cycle (talk) 17:33, 5 December 2010 (UTC)

If you had been clearly and directly arguing '"Cliburn" "is" very often used to refer to the pianist' and convinced us that was true, you would have gotten agreement quicker and in a single thread: "only exceptional cases where the person is commonly referred to only by just the single name. Then he'd be ambiguous and possibly primary." -- JHJ above. The semantic discussions are important, but I think not a major issue with the Cliburn discussion. --John (User:Jwy/talk) 18:41, 5 December 2010 (UTC)

First, I have presumed that anyone familiar with usage of the English language knows that persons in general are often referred to by surname, and anyone familiar with computerized searching knows that people are often searched for by surname. That point has nothing to do with Cliburn in particular.

Second, the article about Cliburn has by far the highest page view statistics and ghits of any use of the term "Cliburn".

I thought the implications of these two points with respect to whether a hatnote link to the pianist at Cliburn would be useful were rather obvious. --Born2cycle (talk) 04:19, 6 December 2010 (UTC)

Actually, the opening sentence of WP:MOSDAB tells us that "dab pages are designed to help a reader find Wikipedia articles on different topics that could be referenced by the same search term". Nothing about "only exceptional cases". The interpretation in MOS:DABNAME that "could" means "are very frequently" in the case of given names and surnames seems to me to be too restrictive and inconsistent with the general principle. Mhockey (talk) 19:40, 5 December 2010 (UTC)

Please read the linked discussions from Wikipedia:WikiProject Anthroponymy#Background reading, where this was hashed out. -- JHunterJ (talk) 21:55, 5 December 2010 (UTC)

Except that I read the same discussions and I don't quite see the that the discussions reached any strong consensus regarding the treatment of surnames on disambiguation pages, apart from agreement that surnames are ambiguous, but that they could be split off into separate pages when the lists became unwieldy. older ≠ wiser 23:01, 5 December 2010 (UTC)

I agree. In any case, the discussions referred to are, as far as I can see, over 3 years old. Consensus changes. Editors use dab pages for their intended purpose, to assist users to find topics which could be (not are very frequently) referenced by the same search term - including cases where the search term is a surname. They have consensus on their side (the opening sentence of WP:MOSDAB). The assertion that editors should not do so except in "exceptional cases" detracts from the helpfulness of dab pages. Mhockey (talk) 00:00, 6 December 2010 (UTC)

I have just nominated for deletion here the recently created surname articles Cliburn (surname) and Simister (surname). --Born2cycle (talk) 04:19, 6 December 2010 (UTC)

Yes, consensus can change. If consensus has changed, the guidelines should be changed (with consensus) before trying to apply the hypothetical new consensus to the articles. I'd like to think that the disambiguation pages will continue to disambiguate only those "ambiguous topics" -- topics that are need of disambiguation, rather than partial title matches, and I'd rather not see long lists of surname-holders cluttering up disambiguation pages, which is what we were getting away from with the consensus three years ago. -- JHunterJ (talk) 18:10, 8 December 2010 (UTC)

There is no need to change the guidelines except maybe to make it clear that your interpretation is based on a misreading. In fact, when you tried to change the wording to be more in line with your understanding, it was reverted, and consensus apparently supports that revert.

No where do the guidelines intend to say or imply that links to articles with partial titles matches, including to articles about persons with surnames that match the homograph being disambiguated, should not be on the dab page, unless there is "no significant risk of confusion". The plethora of examples of surname redirects to articles about persons, as in McNealy, and the countless dab pages that list links to persons on them due to surname "partial matches" (see Armstrong), clearly indicates consensus about the "significant risk of confusion" that people with a given homograph as surname have with other article topics to which that homograph might refer. But of course when the list becomes so long as to warrant a split out to a list article, that's fine, but that has nothing to do with articles about the topic of the surname in question, which, as you like to point out, are not dab pages (and therefore should not list the persons that need to be disambiguated - those belong on the dab page). --Born2cycle (talk) 20:45, 8 December 2010 (UTC)

I find that a rather Pointy move given that the topic is still in dicussion here. But if you are going ahead with it, I suggest you also notify the Wikipedia:WikiProject Anthroponymy project as they have a lot of experience and interest in such pages. --John (User:Jwy/talk) 05:38, 6 December 2010 (UTC)

Actually, I thought their recent creation was the rather Pointy move, especially since their existence is contrary to what the relevant guidance suggests (the basis of the deletion nomination). Besides, AfD is probably a good way to get more people involved to weigh in the various related issues. Good suggestion about the Wikipedia:WikiProject Anthroponymy project notification to get even more input. Thanks. --Born2cycle (talk) 05:51, 6 December 2010 (UTC)

Anthroponymy

Over at Wikipedia:WikiProject Anthroponymy I noticed a list of resources at the top, including a link to Wikipedia:Deletion policy/names and surnames which is a discussion about surnames that reached, among others, the following conclusion relevant to this discussion:

2. Articles on surnames are useful as disambiguation between articles between people with that surname (or as a redirect to one such article).

Isn't the statement that "articles on surnames are useful as disambiguation" based on the premise that people are likely to search for articles about persons using just surname? Note this comment from one participant:

[All names] may also serve as useful disambiguation pages to those who have that name ("What was he called... something Robertson...").

and this from another:

Names should basically be kept as disambigs only.

I see no qualification here (or anywhere else) that this applies only to articles about persons that meet the "Einstein" criteria (widely known to be referred only by surname). As far as I can tell, there has never been any support, much less consensus support, for JHJ's claim above that "Most name holders are not ambiguous [sic] with either their given name(s) or their surname." --Born2cycle (talk) 06:26, 6 December 2010 (UTC)

Disambig templates

There's a discussion regarding disambig templates over at Wikipedia:Templates_for_discussion/Log/2010_December_7#Disambig_templates. Mhiji (talk) 02:32, 7 December 2010 (UTC)

Too much emphasis on "search term", to little emphasis on "refer to"

The recent changes by Born2cycle put too much emphasis on "being a search term" at the expense of "being referred to as". I have disagreed with this clearly and repeatedly over the recent explosion of topics, move requests, deletion requests, and other talking points. But I am taking a break at least for the Christmas season -- I'll check back to see what the new direction of "disambiguation" is to be on Wikipedia, whether it will include more unexceptional name holders, other partial title matches, other search destinations without risk of confusion, etc. -- JHunterJ (talk) 19:17, 9 December 2010 (UTC)

Well, at least we can agree that this is at the crux of our disagreement. It is my understanding that the importance of "refers" is derived from the importance of "search". After all, what is the reason we even care about what refers to what if not to make "search" work better for users? That is, the reason Einstein redirects to Albert Einstein is because "Einstein" is often used to refer to "Albert Einstein", but the reason that's relevant is because that makes it likely for people searching for the article about him to be searching with just "Einstein". The first two of three aspects of disambiguation are mostly unaffected by this disagreement. Disambiguating article titles to avoid conflicts in titles and fixing links in articles to point to the correct article title has little to do "refers" or "search". After all, disambiguating Madonna by putting the entertainer at Madonna (entertainer) is not about "refers" or "search". But with respect to the third aspect, redirects and links on dab pages and in hat notes, that is all about searching. --Born2cycle (talk) 20:27, 9 December 2010 (UTC)

I think I have to agree with JHunterJ on this. The search term language brings us uncomfortably close to opening the door to disambiguating the Internet. This policy/guideline was originally and always has been an adjunct to policies about naming articles. I don't agree with JHJ about the treatment of surname-holders on disambiguation pages and primary topics, but I don't think that needs changing this to be based primarily on "search strings" rather than article titles. Seems there is are two different aspects that need to be considered, 1) inclusion on a dab page (and I think likely search terms should be included on dab pages and which is more the purview of WP:MOSDAB than this page) and 2) consideration for primary topic -- examples like Einstein indicate that surname holders can be the primary topic for a term; and so far as I'm concerned, surname-holders should also be considered when determining whether some other term is a primary topic. I don't think this page needs to be based on search terms instead of article titles for that. 21:17, 9 December 2010 (UTC)

Above comment is from User:Bkonrad [4]. --Born2cycle (talk) 23:32, 9 December 2010 (UTC)

R to Bkonrad: Examples like Einstein (examples for which you can find reliable sources that refer to the subject without ever using the given name, or using the surname for examples like Madonna) are the exceptional cases of people whose topics are actually in need of disambiguation from the single name title (what I would have called "ambiguous topics" before). The unexceptional cases, the normal cases (Freston or Cliburn, for example), since they are not ambiguous, have been separated to the anthroponymy project. In the cases where both the list is short and also there is no anthroponymy article yet created, MOS:DABNAME allows the list to remain on the disambiguation page but still separate from the list of topics in need of disambiguation ("ambiguous topics"). Only the topics in need of disambiguation should be considered to determine the primary topic for an ambiguous title; topics that might show up in a set of search results for the title, but that are not in need of disambiguation, would not be considered for the primary topic. -- JHunterJ (talk) 12:56, 10 December 2010 (UTC)

This is precisely where I disagree with you. Surname-holders are ambiguous under conventions of English usage. This guideline allows such lists of surname holders to be split off when the list becomes unwieldy OR when an anthroponymy article exists (and although I have done it on occasion myself, it is extremely lame to call a list of surname-holders an anthroponymy article). It is (or at least has been) common practice to create a redirect from the surname to a bio article where there are no other uses of the surname. In exceptional cases, one individual might be designated as the primary topic for a term that is the person's surname. That should not preclude consideration of persons with the surname when determining whether some other term is a primary topic or if the term should be a disambiguation page. older ≠ wiser 13:48, 10 December 2010 (UTC)

Then you disagree with the consensus that lead to the anthroponymy project (however lame or not lame the project's list articles may be), not just with me. -- JHunterJ (talk) 13:58, 10 December 2010 (UTC)

Well, no. I think what I have said represents consensus. I think most serious participants in the anthroponymy project would agree that a bare list of surname-holders is a lame excuse for an anthroponymy article. older ≠ wiser 14:11, 10 December 2010 (UTC)

I also believed they are lame lists, and should be deleted. My opinion was also not the consensus that resulted in the anthroponymy project, so at least the lists aren't masquerading as disambiguation pages any more. I don't believe any of the participants, serious or otherwise, in the anthroponymy project, have identified the problem you see or have an approach to clean them up. -- JHunterJ (talk) 14:16, 10 December 2010 (UTC)

But surely if you have a bare list, it shouldn't be deleted as such, but made back into a disambiguation page - so I don't understand your comment about "masquerading" - if it's a list with no historical or etymological information, then it is a disambiguation page (or part of one) rather than an article. Or do you really think Wikipedia should not allow readers to search for biographies by browsing a lightly annotated list of people with a given surname? --Kotniski (talk) 14:23, 10 December 2010 (UTC)

Not all lists without information are disambiguation pages. If it's not a list of topics in need of disambiguation, no, it surely shouldn't be made into a disambiguation page. That would go for set index articles, lists of name holders, or any other lists that editors may find lame. If they are problem lists, then they should be cleaned up to meet the encyclopedia's criteria, not mislabeled to avoid the clean up. And lists of people who are not ambiguous with a title, even if the list lacks historical or etymological information, are not disambiguation pages. I was a supporter of and contributor to the non-disambiguation list of people by name, but Wikipedia:Miscellany for deletion/List of people by name did not agree with me. But we have the lame anthroponymy list articles to allow readers to search for biographies by browsing, and (unlike disambiguation pages) can be annotated as editors have time and inclination.-- JHunterJ (talk) 14:38, 10 December 2010 (UTC)

I would prefer to see the page rolled back to 23:33, 5 December 2010 per WP:BRD, and the subsequent edits discussed to gauge whether consensus supports their inclusion. The edits in question seem to me to alter the guideline in non-trivial ways (e.g., replacing references to ambiguous terms with "homograph"; shifting emphasis from article titles to search strings) and as such I feel prior talk page discussion to establish consensus is warranted. --Muchness (talk) 22:48, 9 December 2010 (UTC)

I've tried to be careful about not changing what the words mean in all of my edits, but only make the intended (or only reasonable) meaning more clear, so I'd like to understand why you believe, for example, changing "references to ambiguous terms with 'homograph'" amounts to a non-trivial alteration. After all, the definition of homograph is: "a word or a group of words that share the same written form but have different meanings" (Homograph) or "A term with the same spelling as, but a different meaning from, another term" [5]. Isn't that just a more precise definition of what was really meant by "ambiguous term"?

Where I think I might have come closest to actually making a non-trivial change is with the nutshell summary. It used to say:

"When an article title could refer to several things, it is necessary to provide links or a disambiguation page so that readers typing in that title can quickly navigate to the article that interests them."

It now says:

"When a search string could refer to any of several topics covered in Wikipedia, it is necessary to provide links or a disambiguation page so that readers searching with that string can quickly navigate to the article being sought."

At a quick glance the original version seems to be about titles since it starts by referring to titles. However, a closer reading reveals that it's really talking about what readers are likely to type in order to find an article, and practically speaking we know that may or may not be a title. Note that the latter part of the original nutshell wording -- "provide links or a disambiguation page " -- refers exclusively to what ~~the unidentified commenter~~ User:Bkonrad above says is "more the purview of WP:MOSDAB" (what is included on dab pages). I'm not saying the current wording is perfect, but I do think it's more clear about what the original wording could have meant. I suggest that the use of "title" in the original wording was a bit sloppy but understandable, considering how often the string used to search for a topic is a title.

But if anyone thinks this or any other specific edit of mine does amount to a non-trivial alteration, rather than merely a clarification in meaning which is what I was trying to achieve, I'd like to know what that is, and why it's believed to be non-trivial. If I did err and create a change in meaning, I would rather we fix that than rollback all the work that has been done (believe me, I put a lot of time and thought into understanding the original meaning, and how to make it more readily understandable). --Born2cycle (talk) 23:20, 9 December 2010 (UTC)

As to Bkonrad's (a.k.a Older/Wiser) concern about "bringing us uncomfortably close to opening the door to disambiguating the Internet", I believe the WP:PTM section continues to protect us from that as much as it did before. What it used to say:

A disambiguation page is not a search index. Do not add a link that merely contains part of the page title, or a link that includes the page title in a longer proper name, where there is no significant risk of confusion.

What it now says:

A disambiguation page for a given homograph is not intended to be a comprehensive search index for that homograph. Do not add a link that merely contains part of the homograph, or a link that includes the homograph as part of a longer proper name, where there is no significant likelihood that anyone would search for that link's topic with that homograph.

Is that really different in meaning? Again, if so, then I failed, because that's not what I was trying to do. Frankly, I was confused about about "confusion" was referring to in the original wording. But after I thought about it some, I realized it could only mean confusing one meaning to which a homograph could refer with another, and that confusion is only relevant to the decision of whether to put links to the meanings on the dab page in the context of search. No? Anyway, that's why I changed this wording as I did. --Born2cycle (talk) 23:32, 9 December 2010 (UTC)

I know that homograph looks, well, unusual or something, but I think it allows for much clearer explanation. The best alternative I can think of is "ambiguous term or phrase", and that's much more cumbersome, and doesn't as clearly convey "a term or phrase that may refer to one of several distinct meanings", especially considering how often the concept needs to be referenced here. Another problem with "ambiguous term" is that it is often conflated with "vague term" because "ambiguous" and "vague" are synonyms (i.e., "ambiguous" is itself a homograph that can refer to either "imprecise" or "one of several" - we mean the latter), but "vague" cannot be the intended meaning in how "ambiguous" is used on this page, but I know it's often (mis)interpreted that way.
If anyone can suggest an alternative to "homograph" that is better than "ambiguous term" and at least as precise as "homograph", I'm open to it. --Born2cycle (talk) 00:02, 10 December 2010 (UTC)

It might be helpful here to consider the deleted contributions of SpringSummerAutumn, an editor on Australian wine topics who, in their first days of editing, created numerous redirects to Howard Park Wines. For example, since the article discusses the joint venture Marchand & Burch Wines, SpringSummerAutumn created redirects from "Machand and burch wines", "Machand and burch", "Machand burch", "Marcharnd burch", "Marchand burch", "Marchand and burch wine", "Marchand wine", "Machandburch", "Marchburch", and even "Mandb". These are legitimate search strings, but not legitimate names.

Perhaps even more relevant to this discussion, SpringSummerAutumn also created redirects from search terms like "Biodynamic wine great southern", "Burgundian australian wine" and "Riesling western australia". These were problematic because various companies offer products related to these search terms, so it is completely inappropriate for Wikipedia to redirect them to a specific one. My question, then, is whether, rather than deleting the redirect from Riesling western australia, we should have converted it into a disambiguation page listing all the topics relevant to that search string? It seems to me that the old version of this policy would firmly answer "No", whereas B2c's new version offers a "Yes". Hesperian 02:30, 10 December 2010 (UTC)

Ah, thank you very much. This is why I asked you to review. I now understand the problem and agree 100% that is not what it used to mean, nor is it what we now want it to mean. It seems like what we want is to mean something in between "title" and "search string". Let me think about it.

Anyway, do you think the problem is inherent in all/most of the edits I made, or only in the nutshell summary and maybe a couple of other spots? Without checking the text I'm expecting to find the latter to be true. Thanks very much. --Born2cycle (talk) 02:47, 10 December 2010 (UTC)

You could argue that when someone employs the search string "riesling western australia", they aren't specifically searching for Howard Park Wines. Possibly they doesn't know precisely what they are looking for; possibly they want multiple results; possibly they are after more general discussion on the overarching topic. This may be a solution to your dilemma: disambiguation pages list all of the topics that users of a search string are specifically looking for when they employ that string, as opposed to all relevant topics. I'm not saying I agree with that, but it's an idea. Hesperian 10:06, 10 December 2010 (UTC)

BTW, this problem was brought up above in the example of ""Kilgore prodigy pianist". -- JHunterJ (talk) 13:54, 10 December 2010 (UTC)

"Kilgore prodigy pianist" illustrates why Hesperian's suggestion is not water tight, because use of this search indicates a specific search for Cliburn. Below I've proposed the wording "reasonably likely to be an article title" as a hopefully more robust solution to this problem. --Born2cycle (talk) 17:36, 10 December 2010 (UTC)

And even though the use of that search indicates a specific search not for Cliburn but for Van Cliburn, we would not create Kilgore prodigy pianist and redirect it to Van Cliburn. So it illustrates why your search term language is to be avoided -- what was Hesperian's suggestion that you think this illustrates a problem with? -- JHunterJ (talk) 21:27, 10 December 2010 (UTC)

If unexceptional surname holders are ambiguous, they should be listed on disambiguation pages regardless of the existence of an anthroponymy article (lame or full) -- the two pages would serve different functions. If unexceptional surname holders are not ambiguous, they should not factor in to the determination of the primary topic for an ambiguous title. So far, since the formation of the anthroponymy project, unexceptional name holders have not been considered ambiguous. -- JHunterJ (talk) 15:23, 10 December 2010 (UTC)

I suggest that surname-holders should be treated similarly to other borderline partial title matches. For example, a disambiguation page for Foo would typically include articles titled Foo Township or Foo University because these entities are frequently know as simply "Foo". In cases where there are many such entities sharing those names, there would be separate disambiguation pages for those terms that are linked to from the main Foo disambiguation page. I think it is much the same with surname holders -- they are ambiguous by conventions of English usage where people are commonly referenced by their surname -- but when the list of such surname holders grows unwieldy they can be split off into separate pages, which currently are perhaps misleading classified as anthroponymy pages. They might be more accurately described as surname set index pages. older ≠ wiser 15:58, 10 December 2010 (UTC)

Yes, I think I and most people agree with that approach, and from my experience I believe it's what generally happens. I don't think it's possible to divide people into two distinct groups - the Einsteins that can be referred to by their surnames alone, and the Frestons who can't - the great majority of notable people are in a big grey area in between, and it's reasonable to expect some readers (though not all, of course) to be searching for those people by surname alone. --Kotniski (talk) 16:06, 10 December 2010 (UTC)

I agree with treating name holders as other partial title matches, and including them when they are frequently known as simply "Foo" (this is the exceptional case of Elvis or Shakespeare) and not when they aren't (the usual case, as with Freston and Cliburn). It's easy enough to divide them into the two groups through reliable sources that refer to the people without using more than the ambiguous title (Einstein, Elvis, Shakespeare) and those that have to use another name to introduce them (Freston, Cliburn). And this is what is generally happening. The hypothetical readers who search for Van Cliburn by surname alone will be able to find him regardless. -- JHunterJ (talk) 16:13, 10 December 2010 (UTC)

Well, they will if we include him. What do you mean by "including...and not"? Are you still arguing that most biographical articles should not be listed on the dab pages for their surnames even if the list is short and unobtrusive, as you did at Cliburn? If so, then I don't think we yet agree at all on that point, nor have I seen any argument as to why your approach might conceivably help the reader. But if you mean listing them but under a separate section of the dab page ("People with the surname...") then that (except in the case of the very short dab pages, where subsections look silly) is fine by me. --Kotniski (talk) 16:20, 10 December 2010 (UTC)

The problem I had with Cliburn was the errant conclusion that a topic that wasn't in danger of confusion (Van Cliburn) could be the primary topic for the title (Cliburn). Including on the disambiguation page a list of name holders not in need of disambiguation as a convenience is fine, as long as editors realize that it is a convenience, not a creation of ambiguity where none exists. Since there was confusion on that point, I tried to solve it (within the guidelines) by taking on the inconvenience of separating out the anthroponymy list to the surname list article. -- JHunterJ (talk) 16:40, 10 December 2010 (UTC)

This is precisely the disagreement, I think. Surname holders ARE ambiguous by conventions of English language usage. There are some exceptional cases where the surname holder might be the primary topic for the term (e.g., Einstein), or might merit specific mention on the disambiguation page even though there is a separate page with list of surname-holders (e.g., Isaac Newton on Newton). In cases such as Cliburn, if over time evidence shows that a significant proportion of readers are going to the disambiguation page from the village article, that would be a strong indication that the village is not the primary topic. In a case like Freston, even leaving aside the likelihood of the two minor celebrities with the surname being the desired target of readers, I don't think the village article even begins to rate as primary topic -- Freston Tower, which appears to be the only thing notable about the place is arguably more a likely target for readers searching for "Freston" than the village and it is appropriate for the disambiguation page to be at Freston. older ≠ wiser 17:26, 10 December 2010 (UTC)

(edit conflict)JHJ, by "errant conclusion that a topic that wasn't in danger of confusion (Van Cliburn) could be the primary topic for the title (Cliburn)", are you suggesting that if there were no other uses for "Cliburn", or the only other uses were other lesser known persons with that surname, that it would be inappropriate for Cliburn to redirect to Van Cliburn? --Born2cycle (talk) 17:36, 10 December 2010 (UTC)

Bkronrad, yes that's the disagreement. If you are correct about the conventions of English language use, the conventions of Wikipedia currently disagree with it (which sometimes happens -- Wikipedia consensus is not English language consensus, for instance). I don't think that you are correct about the conventions of English language use, though, which is why persons that would not be in the Wikipedia disambiguation are introduced in English language sources with more than the surname, while person that are "Wikipedia ambiguous" are introduced without more than the ambiguous title. B2C, if there were no topic that could be titled "Cliburn", yes, "Cliburn" could redirect as a {{R from surname}} (which is not a disambiguation redirect); this is similar to the case where if there is no topic for a title that could be a misspelling for another topic, it can be an {{R from misspelling}}, but if there is a correctly-spelled topic, then the title goes to it, even if the other spelling far outstrips it in hits, and we use {{distinguish}} to link them. -- JHunterJ (talk) 21:16, 10 December 2010 (UTC)

Except that I think what I have described is current practice on Wikipedia. I'm not able to parse what you mean by which is why persons that would not be in the Wikipedia disambiguation are introduced in English language sources with more than the surname, while person that are "Wikipedia ambiguous" are introduced without more than the ambiguous title. Do you seriously dispute the notion it is a convention of the English language that persons are commonly referenced by surname? Really? older ≠ wiser 21:31, 10 December 2010 (UTC)

Note the subtle switch in his wording from merely "referred" to "introduced". The implication is that because non-exceptional persons are virtually never introduced in sources with surname only, that means they are not "commonly referenced" by surname only (never mind that is how they are referred once they are introduced with full name in most of those sources!).

And I don't see how any of the wording at {{R from surname}} supports JHJ's argument that his position reflects policy. No one is suggesting anything other than what that template calls for. Of course a redirect from a surname to an article about a person is not a disambiguation redirect; disambiguation redirects take you to disambiguation pages. Not sure what the point is there either. --Born2cycle (talk) 21:49, 10 December 2010 (UTC)

Evidence this guideline needs wording improved

This incident discussed on Bkonrad's talk page is specific to the wording at thje WP:INTDABLINK section of this page, but I suggest the problem of being unclear is sprinkled throughout this entire guideline page. I tried to improve it, and believe I did, but all that work was reverted for what I believe to be insufficient basis (improving what I had, especially with respect to the "search string" problem discussed above, would have been the better way to go but maybe we can still do that).

Note that Bkonrad's interpretation (that intentional internal links to dab pages that are at [[Name]] rather than [[Name (disambiguation)]] should go to the redirect at [[Name (disambiguation)]]) was correct, but this was not clear to the user in this discussion until I changed the wording (which has now been reverted). So now it and other sections are less clear again. --Born2cycle (talk) 22:42, 10 December 2010 (UTC)

Proposed nutshell rewording

Great discussion, folks. The nutshell summary (now reverted to the version prior to my changes) currently states:

This page in a nutshell: When an article title could refer to several things, it is necessary to provide links or a disambiguation page so that readers typing in that title can quickly navigate to the article that interests them.

Since we "provide links or a disambiguation page" (and redirects) not only when an actual article title could refer to several things, but when a topic name or any reasonable conceivable title could refer to several things, I propose the following wording to convey this:

This page in a nutshell: When a name or description of a topic reasonably likely to be the title of an article may be used to refer to any of several topics covered in Wikipedia, it is necessary to provide redirects, links or a disambiguation page so that readers searching for that topic can quickly navigate to the article being sought.

I suggest this solves several problems without introducing the problems of the "search string" nomenclature I used before, and hopefully without introducing any other problems. I think we need to say "or description of a topic" because many of our articles titles are not names, but descriptions (the family of "List of ..." articles, is one obvious example, but there are myriads more).

I also contend this version reflects the consensus view about how surnames of unexceptional people and other partial title matches are handled: the hurdle that needs to be met is, "reasonably likely to be the title of an article".

Thoughts/comments? --Born2cycle (talk) 17:22, 10 December 2010 (UTC)

I think the proposed version is unnecessarily hard to read because of the added convolutions. We can incorporate the "topic" idea without adding more twists:

This page in a nutshell: When a possible topic title may be used to refer to several topics covered in Wikipedia, it is necessary to provide links or a disambiguation page so that readers typing in that title can quickly navigate to the article that interests them.

-- JHunterJ (talk) 21:22, 10 December 2010 (UTC)

Okay, but title is very specific, and "possible" is too open-ended; ultimately not that most specific in meaning than "search string". Mercury (planet) is a title, but disambiguation is about expecting users to enter "Mercury", not the title "Mercury (planet)", and getting them to Mercury (planet) quickly never-the-less. So, how about:

This page in a nutshell: When a reasonably possible topic name may be used to refer to several topics covered in Wikipedia, it is necessary to provide links or a disambiguation page so that readers typing in that name can quickly navigate to the article they seek.

--Born2cycle (talk) 21:36, 10 December 2010 (UTC)

Certainly the current nutshell isn't quite right, but can we reword it without making it more complicated? Nutshells are supposed to be short and sweet. Or if we can't manage that, then why not just ditch the nutshell? People can read the first sentence of the guideline if they want to know what it's about. I don't really know why people insist on putting nutshells on project pages anyway - they can oversimplify to the point of being highly misleading, and at best just duplicate the function of a well-written leading sentence/paragraph. --Kotniski (talk) 12:32, 11 December 2010 (UTC)

This page in a nutshell: Where two or more articles have similar titles, disambiguation links and pages help readers find what are looking for.

- is as concise as I can formulate. - TB (talk) 15:23, 11 December 2010 (UTC)

But that has the same problem as the existing nutshell had - it implies that disambiguation is only between article titles, when in fact it is between names of topics (which are not necessarily also article titles). It also fails to tell anyone anything unless they already know what disambiguation links and pages are, in which case they almost certainly already know what purpose they serve as well.--Kotniski (talk) 17:20, 11 December 2010 (UTC)

I think a nutshell summary is important. In order to not be "oversimplified to the point of being highly misleading" it needs to be "more complicated". I suggest the modified version of what JHJ proposed strikes a pretty good balance, is better than what was there before, and certainly better than what is there now (nothing). --Born2cycle (talk) 19:15, 11 December 2010 (UTC)

I'm still firmly in favour of nothing ;) The version you refer to is not a helpful summary of the whole page - just a (not particularly informative) expression of one aspect of disambiguation. The page covers a multitude of aspects; I really don't see how it can be summed up in one sentence any better than is already done by the first sentence of the text (or if we can improve on that first sentence, then why not modify that sentence rather than adding another one above it). --Kotniski (talk) 10:44, 12 December 2010 (UTC)

Dictionary definitions?

I've always thought that including dictionary definitions was frowned on -- in the "Dos and Don'ts", it's pretty clear about this: "Don't add dictionary definitions." Big red X and everything.

However, if you follow the link, you find the sentence: "A short description of the common general meaning of a word can be appropriate for helping the reader determine context." Which pretty much undercuts the policy, to my mind.

This is followed by "Otherwise, there are templates for linking the reader to Wiktionary", but this is not the same as explicitly saying "but don't write one in the page itself; link to Wiktionary."

In WP:MOSDAB, it's a little more straightforward and helpful: "When a dictionary definition should be included (see What not to include), then rather than writing a text entry, create a cross-link to Wiktionary"

I would suggest we change the wording here (specifically, here) to align with the wording in MOSDAB. Any objections?--NapoliRoma (talk) 00:28, 19 December 2010 (UTC)

Proposed new wording:

A disambiguation page is not a list of dictionary definitions. If a description of the common general meaning of a word is considered appropriate for helping the reader determine context, add a link to Wiktionary, the wiki dictionary; see Template:Wiktionary.

--NapoliRoma (talk) 00:35, 19 December 2010 (UTC)

I'd rather tone down the prohibitive language. A brief dictionary definition can be helpful in some cases (if only to tell readers that we don't have an article about that meaning). These pages are above all supposed to be helpful to readers - it isn't helpful to shunt them off to wiktionary to find information that we could just as easily give them ourselves. (Though it depends on the situation - sometimes we would be oversimplifying by summarizing the dictionary meanings in one sentence.)--Kotniski (talk) 10:51, 19 December 2010 (UTC)

...but dab pages are there precisely to shunt. It's what they do. Making them do more reduces their functionality as a switchboard to where the content really lives.

Can you give an example of where a dictionary definition makes a dab page a better dab page?--NapoliRoma (talk) 21:16, 19 December 2010 (UTC)

Probably, though I don't remember any examples at the moment - will let you know when I next come across one. --Kotniski (talk) 07:47, 20 December 2010 (UTC)

Disambiguation of Foreign Language terms and inclusion of information not supported by existing articles

Pursuant to the discussion here, I'm wondering if we could add some language either on this page or WP:MOSDAB to clarify when translations of foreign language terms are appropriate for inclusion on a disambiguation page. Brief summary: an editor added an entry to Pine Valley as

Pine Valley, russian term "Kedrovaya Pad", nature reserve in Primorsky Kray, Russia

I removed the entry, as at the time there was no article for Kedrovaya Pad and neither of the linked articles made any mention of either "Kedrovaya Pad" or "Pine Valley". A series of reverts took place. Evidently, the name "Kedrovaya Pad" can be translated as either Cedar Valley or Pine Valley (the plant is the Korean Pine). An editor created a stub for Kedrovaya Pad Nature Reserve and included a reference to calling it Pine Valley. While it still is questionable to me whether the place is commonly known in English as Pine Valley, the reference seems enough to justify a mention in the see also section of the disambiguation page Pine Valley.

I think this is already implied by the content of WP:DAB and WP:MOSDAB, but what I'd like to suggest is clearer language that disambiguation pages should not introduce information that is not supported existing articles. A further question is should disambiguation pages include entries for translations of foreign language terms where there is no indication in any article that the subject is known in English by the translated title? older ≠ wiser 19:53, 21 December 2010 (UTC)

Who is helped by twodabs?

Okay, let's talk about this edit which changed the wording of the twodabs section from this:

If there are two topics to which a given title might refer, and one is the primary topic, putting the primary topic article at the title with a hatnote pointing to the other article is usually preferable. Those looking for the primary topic are taken directly to the topic they seek, and those seeking the secondary topic are only one click away. A disambiguation page at that title would also get readers looking for the secondary usage to their article in one click, but it would cause those looking for the primary topic, the topic most readers are likely to be looking for, to have to click once as well.

to this:

If there are just two topics to which a given title might refer, and one is the primary topic, then no disambiguation page is needed – it is sufficient to use a hatnote on the primary topic article, pointing to the other article. (This means that readers looking for the second topic are spared the extra navigational step of going through the disambiguation page.)

The edit summary is: "this wasn't right - once we have a primary topic we know that people looking for that topic are going straight to where they want to be, it's the others this helps"

Let's take an example, say Birmingham Airport. The primary topic is Birmingham Airport, West Midlands and so that is at Birmingham Airport (actually Birmingham Airport redirects to Birmingham Airport, West Midlands, but that's moot here). The second use is Birmingham-Shuttlesworth International Airport for which there is a hatnote at the top of the primary topic article.

So, of those entering "Birmingham Airport" those seeking the primary topic are taken straight to their article, and those seeking the second topic are one click of a hatnote link away.

Now, if we make Birmingham Airport a dab page, everyone entering "Birmingham Airport" is taken to the dab page, not to any article. From the dab page, those seeking the second topic are still one click (of a dab page link this time) away from their article, but now the primary topic users are also one click away from the article they seek.

So, it seems to me that, going backwards, by getting rid of the dab page we are helping the primary topic seekers; it's a wash (one click either way) for the seekers of the second topic, contrary to what the edit summary and changed wording above says. That is, if we have a dab page then even primary topic seekers are not being taken straight to where they want to be (they're taken to the dab page first). --Born2cycle (talk) 15:31, 27 December 2010 (UTC)

I think the first version above unnecessarily confuses things. The immediately preceding paragraph leads into this as follows: "However, if there is a primary topic, then the question arises whether to create a disambiguation page, or merely to link to all the other meanings from a hatnote on the primary topic article." Thus this paragraph is precisely about the situation in which there IS a primary topic. There is no need to explain the mechanics of what happens when there is no primary topic. This section is corollary with the first bullet in {{db-disambig}}. older ≠ wiser 15:49, 27 December 2010 (UTC)

I think the editor of the second example above is saying that the suggested set up is better than a hatnote link to a two entry dab page on the primary topic article. I like the brevity of the second example, but the parenthetical sentence might be improved to something like "A dab page would add unnecessary navigational clicks in this circumstance." --John (User:Jwy/talk) 16:04, 27 December 2010 (UTC)

Okay, I get it now. In the case of a primary topic there is no option to put the dab page at the title. But, I daresay, the first paragraph explains why readers benefit from hatnote links even when neither of the two actually meets primary topic criteria. This is why I always thought twodabs applied even when there is no primary topic. I mean, what's the downside in randomly picking one to be "primary" and the other to be linked by hatnote? The upside is that at least the readers seeking one of the two topics get there directly, while the others are still only one click away. All upside no downside seems like an obvious improvement to me. --Born2cycle (talk) 16:34, 27 December 2010 (UTC)

That is often what happens -- the first article created becomes primary by default and unless someone is motivated to move pages about, it is often easier to just set up a second article with disambiguated title and add a hatnote in the first. But I'd rather not try to write language to codify that -- the result would likely be perceived as watering down the criteria for primary topic. older ≠ wiser 17:12, 27 December 2010 (UTC)

Observation of move discussions reveals that quite a lot of people think that the shortcut WP:TWODABS means something like "if there are only two topics, make one of them primary" (or, equivalently, "we don't like two-line dab pages"), where in fact all it says is "we don't like two-line dab pages when one of them's a primary topic"). In fact this thinking - that we are more inclined to pick a primary topic when there are only two to choose from - seems to be so common that it might well be documented in the guideline (and when that possibility has been raised at this page in the past, I don't think there's been any objection, though it never actually got done). It belongs in the preceding section, though, the one about "Is there a primary topic?", not in TWODABS (a rather confusing shortcut anyway), which deals with the separate question of whether to use a dab page or just hatnotes.--Kotniski (talk) 18:29, 27 December 2010 (UTC)

The reason I thought that we didn't make non-primary topics primary is because we have enough contention identifying primary topics when one of them actually is primary without having to worry about the contention that may arise from choosing between two obviously non-primary topics. If codified, we'd need to say that if there is consensus for one of the two to be placed at the base name even though it doesn't meet the criteria for primary topic, but rather the editors would rather have one "side" skip the need for an extra click, then we can. I wouldn't want to use the word "randomly". The downsides would be that contention and, possibly, readers on slow connections having to load a large article with a hatnote rather than loading a small dab page. Not killer downsides, but not "0" either. -- JHunterJ (talk) 18:49, 27 December 2010 (UTC)

Yeah, I only said "randomly" to point out that regardless of which one is chosen to be treated as if it's primary, even if it's the one less likely to be sought, no users are worse off as compared to having a dab page at the plain name, and at least some readers are better off. But I wasn't suggesting we use that term in the guideline.

I think what really happens is when there are only two uses the criteria for primary topic is looser. In other words, when there are only two topics to which a given term may refer, then the normal criteria for primary topic, "much more likely than any other, and more likely than all the others combined " often seems to be compressed and softened to simply be, "more likely than the other".

But instead of having two different criteria for 2 dabs and 2+ dabs, we could change the existing criteria from saying "much more likely than any other" to just "more likely than any other": "more likely than any other, and more likely than all the others combined ". After all, if one topic is more likely than any other and more likely than all the others combined, to be the one being sought, isn't it the primary topic? Since it already must be more likely than all the others combined to be the one being sought, why does it also have to be much more likely than any other? Why isn't merely being more likely than any other, as well more likely than all the others combined, sufficient to be primary? If it was that, which I suggest accurately reflects how primary topic is usually interpreted regardless of what is said here anyway, then in the case of only two uses, this criteria would logically compress to "more likely than the other", which also accurately reflects reality. --Born2cycle (talk) 20:35, 27 December 2010 (UTC)

Because we have left 51/49 splits without a primary topic, in theory. That's how I continue to interpret it. -- JHunterJ (talk) 14:45, 31 December 2010 (UTC)

Do you really think we have the means to distinguish a 51/49 from a 49/51 split? If they're really that close, then we really can't claim that one is "more likely than the other". That is, the difference has to be significant enough to be discernible to be able to argue that one is "more likely than the other". So I still don't see the need for the "much".

Any objection to removing the "much" from "much more likely than any other" at WP:PRIMARYTOPIC? --Born2cycle (talk) 16:57, 31 December 2010 (UTC)

Yes; please leave our naming guidelines alone. This would worsen Wikipedia - although it would help B2C's long term agenda.

This post is incoherent; we make 90-10 splits primary topics but make 51-49 splits disambiguation pages in large part because it is hard to tell 51-49 from 49-51; so the end works against the beginnng.

The other part is because primary topic inconveniences a minority, who have to click twice, to help a majority get directly where they want to be. This inconvenience is more acceptable when the minority is 10% or 5% then when they are 49%. Septentrionalis PMAnderson 17:29, 31 December 2010 (UTC)

That isn't entirely true - it's only those whose topics won't fit into the hatnote that are inconvenienced in this way (the extra click), and if we're talking not about multiple other topics which become significant when added up, but one particular one that's significant on its own, then it can be mentioned in the hatnote. --Kotniski (talk) 17:42, 31 December 2010 (UTC)

(ec) If it's more likely than all the others combined, then it's certainly more likely than any other, so if we remove "much", we should remove that whole subclause. But I think the subclause does have some purpose, if only to avoid the confusion, ill-feeling and wrong links that tend to result in practice if we try to force one of two apparently equal topics in front of the other for no discernible reason. Though if there is a discernible reason then I'd generally be in favour of going for the primary topic solution. (So I'd perhaps favour replacing "much" by some other less intense adverb, but not dropping it altogether.)--Kotniski (talk) 17:39, 31 December 2010 (UTC)

B2C, right, the fuzziness of 51/49 vs. 49/51 is exactly my objection to removing the word "much" from the guidelines. If you are using a definition of "more likely" (but not "much more likely") that somehow excludes 51/49, that will need further explanation. And rather that adding that further explanation to the guidelines, I'd rather avoid introducing the problem. Keeping "much" avoids that introduction. -- JHunterJ (talk) 22:27, 31 December 2010 (UTC)