Wikipedia talk:Disambiguation/Archive 35

This is an archive of past discussions. 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 30

←

Archive 33

→

RM and hits and links

I'm seeing more WP:RM requests for moves that are not contested and are solely based on links and or view counts and or Google results. I'm becoming more convinced that these should be closed as no consensus or something else that does not result in a move since relying solely on counts does not meet the spirit of the guideline. Is my thinking here wrong? Vegaswikian (talk) 22:21, 25 September 2011 (UTC)

I'd say so, but I'm slightly biased. I only came here because some editors visited some pages I watch and decided to start moving them around solely on the basis of view counts. Wiki-Ed (talk) 22:56, 25 September 2011 (UTC)

I'm inclined to complete those moves if I think they're appropriate, whether for stated reasons or for other reasons, but I'm uncomfortable with page views being given as the sole justification for moving. It doesn't seem appropriately thoughtful and considering of a broad set of factors. That's precisely what brought me here yesterday, to the above section. It's also what, together with the above discussion, inspired the edit I just made to the guideline today.

So, I'm with you in spirit, Vegaswikian, but I think the above section provides ample proof that some Wikipedians are not. -GTBacchus^(talk) 23:42, 25 September 2011 (UTC)

An admin should not close an uncontested request as no consensus just because the closer disagrees with the request. In a case like that, the admin should either register his opposition as an editor and let someone else decide, or relist for comments, or just leave it alone. Station1 (talk) 03:41, 26 September 2011 (UTC)

So you are saying that an admin should ignore policy and guidelines? That is the base problem with these. They are not supported by anything other then page views which are not the final decider of what primary use is. Vegaswikian (talk) 05:24, 26 September 2011 (UTC)

No, I'm saying that in cases where an editor has proposed a move in good faith, whether because he can't move it himself and thinks it might be controversial or genuinely wants input from others, and no one has opposed the move, there should either be discussion about the move before it's closed or an admin should just move it if uncontroversial, but an admin should not close it as no consensus if the only opposition is the admin himself. That would be acting as prosecutor and judge. That's regardless of the type of request or the reason for opposing it. As to specifics of pageviews, there's absolutely no consensus that pageviews alone cannot be a reason for a move. Many editors cite them. If John Xyz (carpenter) consistently gets 500,000 pageviews per month while John Xyz (painter) gets 500, many editors would agree that's probably all that need be said. Station1 (talk) 05:56, 26 September 2011 (UTC)

If I see a move with no opposition, and I think it's contrary to policy, then my approach is to register an oppose and let someone else make the call. -GTBacchus^(talk) 07:46, 26 September 2011 (UTC)

I agree with Station1 & GTBacchus. Admins should either contest or carry out uncontested move requests; that's honoring, not ignoring, policies and guidelines. -- JHunterJ (talk) 11:01, 26 September 2011 (UTC)

Another RM issue: if an RM request is to move a disambiguated page to the base title, at which there is currently a disambiguation page which will have to be moved too, there should surely be a notification of this on the disambiguation page. But there are a batch of such suggestions on the RM board at present, such as Sean Kelly (cyclist), where anyone watching the dab page (Sean Kelly in this case) would not be alerted to the move proposal. This seems wrong and I have raised it at Wikipedia_talk:Requested_moves#Notifications_of_Requested_Move. PamD 09:12, 26 September 2011 (UTC)
Agreed. Any affected page should be engaged on their Talk for a Move request. The closing admin should also recognize this when they attempt to carry out the move and find an unengaged page or redirect at the intended target. -- JHunterJ (talk) 11:01, 26 September 2011 (UTC)
Strongly agree with this. Too many times I only notice such a move on a disambiguation page on my watchlist after the move has been completed. Seems it should be standard operating procedure for those closing moves to not close a move involving multiple pages unless the other pages affected by the moves have been properly notified. older ≠ wiser 18:23, 26 September 2011 (UTC)
I call 'em as I see 'em. Redirect pages that are affected? Hard to care about. The more going on with the affected page, the more likely that I care about notifying it. Actual talk page traffic is enough to make it essential. -GTBacchus^(talk) 19:57, 26 September 2011 (UTC)
So if someone wanted to move Usa, Japan over Usa, you wouldn't have any problem if the only place the discussion took place or was noted was at Talk:Usa, Japan? Redirects in general are easy to care about, even if specific trivial redirects are trivial. -- JHunterJ (talk) 20:29, 26 September 2011 (UTC)
You came up with an example where I obviously care unless I'm a great fool. Very clever. You may rest assured that I'm not a great fool, and that if you find a problem that incredibly dumb in the literal reading of my words, it means that's something I wasn't thinking about when I composed that particular sentence. That's such a screamingly clear case that I see no point talking about it. Specific redirects that are obviously very important... are obviously very important. Is stating very obvious things really worth our time? Remember that our policy pages are not meant to be interpreted by robots as strict algorithms, much less auxiliary discussions such as this one. Sheesh. -GTBacchus^(talk) 21:17, 26 September 2011 (UTC)
You're welcome -- happy to help sharpen your position. Yes, stating the obvious is worth our time, since so many editors will ignore the obvious in favor of their pet topics, and point to their absence in the guidelines if challenged. -- JHunterJ (talk) 00:28, 27 September 2011 (UTC)
My position is not sharpened. It has always been to use my common sense, and that's served me well. The above was not a "statement of position". It was me having a conversation. -GTBacchus^(talk) 00:44, 27 September 2011 (UTC)
I think redirect pages are a judgement call. But the original comment in this thread wasn't about redirects. I think it is a problem when there is insufficient notification of moves that affect multiple pages. older ≠ wiser 01:32, 27 September 2011 (UTC)
Thanks, BK, for bringing this back to my original point: I don't see a problem if a page is moved to over-write a redirect which was pointing to the page anyway; I do see a problem if a page "(XYZ (whatever)" is moved to a title already occupied by an actual page "XYZ" (usually by a dab page which has then to be moved to "XYZ (disambiguation)"), without there being any notification at Talk:XYZ. It's been confirmed at Wikipedia talk:Requested moves that such moves ought to be nominated as Requests to Move Multiple Pages, but this hasn't always been done correctly. I hope indeed that closing admins will notice such cases and not make the move until proper notification has been made, but from BK's experience above this does not always happen. PamD 08:17, 28 September 2011 (UTC)

Guidance for when both primary topic criteria apply

Primary topic is now defined in terms of two criteria: "importance" as well "likelihood of being sought". There is wording in there that suggests these two criteria are rarely in conflict:

In many cases, a topic that is primary with respect to usage is also primary with respect to importance. In many other cases, only one sense of primacy is relevant. In a few cases, there is some conflict between a topic of primary usage and one of primary importance. In such a case, consensus determines which article, if either, is the primary topic.

All it says about such cases is that "consensus determines" which article if any is primary. In other words, no guidance is given. Anyone can argue anything on either side, and there is no criteria to determine which is preferred, much less any clear objective criteria to make that determination.

In any case, if a good case can be made for each side, wouldn't that indicate that neither choice is primary? That seemed pretty obvious to me, so I added the following accordingly (this diff includes an improvement from Hesperian):

In most such cases, if a good case can be made for each side, that's a strong indication that there is no primary topic, and so a disambiguation page should be at the base name.

However, Station1 (talk · contribs) reverted this change with an edit summary of, "that would heavily bias discussions toward dab pages as default)". It would? How so? It seems to me it would do that only in situations where "there is some conflict between a topic of primary usage and one of primary importance" and "a good case can be made for each side". Don't we want to bias toward using a dab page under those conditions? What's the problem? --Born2cycle (talk) 08:11, 30 September 2011 (UTC)

"...if a good case can be made for each side, wouldn't that indicate that neither choice is primary?" No, it only indicates that there is no consensus as to which is primary. Until now, when there is no consensus after discussion about a proposed page move, the default has been the status quo. Your edit would fundamentally change that so that if there were no consensus the closing admin would change an article to a dab page. This could also encourage gaming the system by someone favoring a dab page by proposing a move and just hoping for no consensus rather than a consensus for a move. It would increase the number of dab pages as opposed to articles at controversial titles. "Don't we want to bias toward using a dab page under those conditions?" No, there is no consensus that we do. It's often better to get at least some, if not most, readers directly to an article with a hatnote than to a dab page where no one wants to be. Station1 (talk) 08:37, 30 September 2011 (UTC)

To be honest, I don't think this would be a bad thing. If there is no consensus that there is a primary topic (and provided that opposing positions are supported by evidence and reasoned arguments), then what is the problem with defaulting to a dab page? I don't see that inconveniencing (or deprecating) a significant minority by placing one topic as primary is a good thing. older ≠ wiser 16:29, 30 September 2011 (UTC)

I don't follow Station1's reasoning. It's not gaming the system to use the RM process to discover situations where an article is at a base name but there is no consensus about the topic of that article being primary for that base name. That's an example of improving Wikipedia for the readers.

Now, if you want to argue that readers are sometimes better off when an article remains at a base name even though there is no consensus for it being the primary topic, please, argue that. But really the edit I made is consistent with what it has long said and continues to say: " If there is no primary topic, the term should be the title of a disambiguation page (or should redirect to a disambiguation page where more than one term is disambiguated on one page). ". --Born2cycle (talk) 18:01, 30 September 2011 (UTC)

"If there is no primary topic, the term should be the title of a disambiguation page" is fundamentally different from If editors disagree about which term is the primary topic, the term should be the title of a disambiguation page, so your edit would be a major change. If I wanted John Paul Jones to be a dab page, instead of making that proposal at Talk:John Paul Jones, as someone recently did in good faith, and having it soundly rejected, I could instead propose moving John Paul Jones (musician), hope for an Anne Hathaway situation where half point out pageviews and half point out historic importance, and wait for a closing admin to decide that there was no consensus as to primary topic so guidelines say we should make this a dab page. That's what I was referring to as gaming, but that wasn't my main point, as most proposals would be good faith but still result in a dab page by default rather than status quo by default.

The problem with defaulting to a dab page is that dab pages are not articles; they have no content and few people are looking for them. Unfortunately, when 2 articles would have identical titles, someone must be inconvenienced. In the case of John Paul Jones, large numbers are looking for the musician and large numbers are looking for the naval hero. We can inconvenience 1/3 of them by making the musician the primary topic, or 2/3 of them by making the naval hero the primary topic, or virtually 100% of them by using a dab page. I think the last option is the least desireable. Station1 (talk) 18:55, 30 September 2011 (UTC)

So you are arguing that it's sometimes preferable to treat a topic as if it is the primary topic even though it does not meet primary topic criteria. BTW, we can't "make" a topic be the primary topic - either a topic is primary or it is not - whether we treat it as the primary topic (by putting its article at the base name) is the only issue over which we have control. This is an important distinction because I think people sometimes conflate (as you just did) the process of determining whether a topic is primary with the decision to treat a topic as primary.

Let me ask you this. Do you think dab pages at base names are ever desirable? If so, when? I mean, per your argument, isn't it better to always put the most sought topic, or the most important topic, at the base name, regardless of whether it is "much more likely to be the one being sought"? Your argument, as I understand it, interprets primary topic very differently from what it says. --Born2cycle (talk) 19:08, 30 September 2011 (UTC)

I agree "treating" would have been a better choice of words than "making". The rest is I think going off on a tangent from discussing the specific edit in question, but having said that: No, dab pages at a base name are never desireable, although they are sometimes necessary, when there are multiple articles with no primary topic. But the important point here is that coming to a consensus that there is no primary topic is not the same as failing to reach consensus as to which of two articles is the primary topic during a move discussion. Station1 (talk) 19:53, 30 September 2011 (UTC)

I don't think the rest is a tangent since it seems to be a cornerstone of your objection to the edit in question. Thank you for being so clear. I think we can agree to disagree on the following point:

Whether dab pages at a base name are ever desirable. I think dab pages at a base name are desirable whenever there is no clearly-much-more-sought-after topic for that base name because all alternatives (putting any of the non-much-more-sought-after topic articles at the base name) are less desirable. They are less desirable because that would mean the majority of readers searching for that name would be taken to an article that they are not seeking when they would be better served by a dab page.

Of course I agree that reaching consensus and not reaching consensus are not the same about anything, but in the specific case of not reaching consensus on whether either of two topics is the primary topic for that name, and with good arguments supporting each one, I still say that particular lack of consensus is a strong indication that no topic is primary for that name. I also think that that statement, as well as the edit in question, is not a declaration that there is no primary topic in that situation. I can suggest a slight rewording:

In most such cases, if a good case can be made for each side, that's a strong indication that there is no primary topic, and probably means users would be best served with a disambiguation page at the base name.

Is that more acceptable to you? It would be nice to get input from others on these issues. --Born2cycle (talk) 21:30, 30 September 2011 (UTC)

I'm not bothered by a dab page at a main title, and more to the point, lots of Wikipedians are fine with it, too. Here's an argument from a purely statistical point of view for why it can be preferable: Suppose we have 4 topics called "XX". They receive 40%, 25%, 25%, and 10% of page views. If we put the 40% topic at the main title, with a hatnote to the dab page, then 60% of people simply typing "XX" into search expend two extra clicks each. If we put the dab page at the main title, then 100% of people typing "XX" into search expend one extra click each. For 100 surfers, that's 120 extra clicks if we treat the 40% topic as primary, and only 100 extra clicks if we treat none as primary. That's why the statistical rule has always been that a topic needs more views than all others combined to be statistically primary.

Taking into account that many Wikipedians care about considerations other than statistics, it seems to push the matter even more in that direction. If there is no >50% topic, and no major considerations of "importance", then put the dab page at the main title. If there is a >50% topic, but many Wikipedians arguing that some other topic(s) beat it out in terms of "importance", then all the more reason to put a dab page at the main title. -GTBacchus^(talk) 23:03, 30 September 2011 (UTC)

I was afraid this would happen. We're getting sidetracked on to a totally tangential issue of whether dab pages are good or bad. The point is that the effect of Born2Cycle's proposed edit is to say that disagreement about which of two articles is a primary topic means there is no primary topic and therefore an article should be turned into a dab page whether there is consensus to do so or no consensus to do so. To me the premise still seems illogical and the recommendation a fundamental change to current practice. Station1 (talk) 23:17, 30 September 2011 (UTC)

Edits to this page do not change current practice. God, I get tired of hearing this superstition aired. People who think that practice is based on guidelines end up very, very frustrated; I've seen this happen so many times... so many times. The question is whether Born2cycle's edit is accurate.

If you make the argument - which you did - that dab pages at main titles are undesirable, then people will point out that you're wrong in many cases, from a purely convenience-based point of view.

Now, if there is significant disagreement as to which topic is primary, then it's relatively common for the dab page to end up at the main title. A case in which that has not occurred is Corvette, although that one is complicated by the issue that the most popular potential topic is located at Chevrolet Corvette, and not at something like Corvette (Chevrolet).

Have we got good examples of cases where we've chosen to keep articles at main titles in the face of significant disagreement as to primacy? Let's look at them, and ask whether Born2cycle's edit reflects what has happened in those cases. If it doesn't, let's fix it. -GTBacchus^(talk) 23:36, 30 September 2011 (UTC)

I guess John Paul Jones is another example we can add to our list further up the page. Born2cycle, does your edit reflect what's going on with that page, do you think? Is the arrangement of John Paul Jones pages in line with consensus? Is that based on general principles, or on ad hoc considerations? -GTBacchus^(talk) 23:40, 30 September 2011 (UTC)

I don't think your repeated assertion that guideline and policy pages are purely descriptive and not prescriptive at all holds true. Why document guidelines and policy at all, if they are purely descriptive? Powers ^T 17:14, 2 October 2011 (UTC)

Because they are helpful for instructing new users in the community norms and they are prescriptive when individual users repeatedly flout community norms. Beyond that, there is value in documenting community norms as clearly as possible so that they can be examined, challenged and amended as appropriate. Unless such community standards are documented, it has more the status of tribal lore than of anything useful. older ≠ wiser 17:21, 2 October 2011 (UTC)

Right. Since they are helpful in instructing users, then it's not at all improper to talk about how guidelines provide "recommendations" and "encourage" certain behaviors. Powers ^T 19:54, 2 October 2011 (UTC)

Maybe so, but GTBacchus' statement that you seemed to be criticizing, that [e]dits to this page do not change current practice is accurate. Many wikipedians are quite content to remain oblivious to the arcane minutiae of guideline pages and only become concerned when other editors attempt to bludgeon them with some guideline. Considering that there is a tendency for some editors to wield policy/guideline pages like cudgels, I think it is prudent to be careful in recognizing the limitations and not to inadvertently give credence to the position that guidelines are akin to laws. older ≠ wiser 20:12, 2 October 2011 (UTC)

But policy and guideline pages are here to provide structure and consistency across disparate subject areas within the encyclopedia. Each quite wisely allows for exceptions, both because we cannot document every possible variation and edge case, and because we know they're not perfect and may not mesh with best practices at all times. But that doesn't make them purely descriptive, and to claim that not only can't they affect current practice, but that they shouldn't seems to defeat the whole point. Powers ^T 00:20, 3 October 2011 (UTC)

This is a very interesting philosophical conversation (no sarcasm) but it's a tangent from what we're working on here. Treating guideline and policy pages as purely descriptive is a good way to think when you're helping write them. Let's talk about how to make the page more accurate, and not about the extent to which these pages should be descriptive versus prescriptive. We can have that conversation somewhere else. How does that sound? (outdenting...)

This section is about how we can provide accurate guidance for when more than one primary topic criterion may apply. Thoughts? -GTBacchus^(talk) 00:42, 3 October 2011 (UTC)

(after ec) GTB can speak for what the intent was, but I read the statement as saying simply that there is not necessarily a direct correlation between changes made to a guideline page and the behavior of editors in the field; i.e., you can't edit a guideline to say what you think it ought to say and expect that editors will go along with that. The guidelines should describe as clearly as possible the best practices as they are presently understood. older ≠ wiser 01:09, 3 October 2011 (UTC)

If that's all GTB was saying, then getting huffy at Station1 for "airing a superstition" was rather out of line. Station1 was merely pointing out that B2C's proposed wording could bias outcomes, not that it would immediately result in a change in behavior site-wide. Powers ^T 11:17, 3 October 2011 (UTC)

Whoa! I thought the point was that the current guideline no longer accurately described what actually happens. B2C proposed some language that attempted to more closely approximate current practice. Station1's objection to changing the guideline seems retrograde, since there is such a noticeable gap between the current language of the guideline and actual outcomes. older ≠ wiser 11:49, 3 October 2011 (UTC)

I hate to speak for Station1 here, but I believe Station1's objection was to a specific choice of words that changed the "recommendation" found in the guideline from defaulting to the status quo to defaulting to a disambiguation page (in certain cases). Powers ^T 12:03, 3 October 2011 (UTC)

Protracted disagreement about whether there is a primary topic was part of the guidance for a long time until it was somewhat controversially removed a while back. However, as a practical matter, actual move discussions have in some cases continued to produce outcomes as if that guidance were still in effect. I think what we're struggling with is how to articulate guidance that reflects what appears to be a real community value in some cases. older ≠ wiser 12:50, 3 October 2011 (UTC)

I'm not sure we can, frankly. Powers ^T 11:17, 3 October 2011 (UTC)

Mind if some of us try? -GTBacchus^(talk) 13:47, 3 October 2011 (UTC)

Noting that the existence of good arguments in favor of two different topics being primary for one term is a strong indication that neither is primary (and there is no primary topic) is an attempt to do exactly that ("provide accurate guidance for when more than one primary topic criterion may apply"). --Born2cycle (talk) 18:00, 3 October 2011 (UTC)

That's appreciated, Born2cycle, at least from my end. -GTBacchus^(talk) 20:06, 3 October 2011 (UTC)

Proposal/Poll

The following discussion is an archived discussion of the PRIMARYTOPIC section edit proposal. Please do not modify it. Subsequent comments should be made in a new section on the talk page. No further edits should be made to this section.

The result of the PRIMARYTOPIC section edit proposal was to keep discussing.

I propose a poll to determine how close we are to consensus:

Do you support or oppose adding this statement to the description of WP:PRIMARYTOPIC as accurately reflecting what's happening out there:

In most such cases [where there is some conflict between a topic of primary usage and one of primary importance], if a good case can be made for each side, that's a strong indication that there is no primary topic and that a disambiguation page should be at the base name.

Support. I think this accurately reflects how decisions are often made, and, more importantly, how they must be made to accurately reflect consensus. --Born2cycle (talk) 21:46, 3 October 2011 (UTC)
Close the poll - I propose we don't jump into voting mode. Discuss, patiently. I don't think going to a vote is the way to do things. Let's just keep talking, and it will be clear what has consensus support. Let's try to avoid voting as much as possibly possible, please. Please? -GTBacchus^(talk) 01:06, 4 October 2011 (UTC)
Yeah, close the poll, per GTBacchus. Hesperian 03:06, 4 October 2011 (UTC)

The above discussion is preserved as an archive of the proposal. Please do not modify it. Subsequent comments should be made in a new section on this talk page. No further edits should be made to this section.

Merging list articles with non-article dabs

Issues with merging list articles into non-article dabs and mosdab formatting under discussion at Talk:Black Saturday#How should we merge?. -- JHunterJ (talk) 10:51, 4 October 2011 (UTC)

Discussions of possible interest to disambiguation

Some potential impact on discussion of primary topic (in a sense, making a claim that there is no primary topic for a title, even when there is only one existing article with the title). older ≠ wiser 12:23, 4 October 2011 (UTC)

Very interesting, thank you. -GTBacchus^(talk) 14:35, 4 October 2011 (UTC)

Primary Topic: Plea for clearer wording

I came here seeking clarification of the policy, but the wording of "Is there a primary topic?" is too vague. Having read through recent discussions I appreciate that this formulation may be deliberate (or the result of an unhappy consensus), but it cannot be left in its current state because of the confusion it can generate across a wide range of topics.

There appear to be two schools of thought regarding the meaning of the word "primary".

Primary usage: Article naming should be determined purely on the basis of visitor statistics. Advocates believe that for an online source accessibility is most important.
Primary importance: Article naming should incorporate common sense consideration of relevant factors including, among others, visitor statistics. Advocates believe that for an encyclopedia educational value is most important.

I would like to concentrate on the improving the current wording. I think there are three problems with it:

It favours the 'primary usage' school by according too much importance to statistics;
Where it does list other factors it does not articulate what they mean or how to measure them against one another;
It misses out key factors, some of which are already used by editors in certain fields.

I do not wish to reopen old arguments (although I suspect this is inevitable) so before going into more detail and spending time drafting alternative text does anyone object to me raising this issue again? I don't think we can leave this page as it stands. Wiki-Ed (talk) 22:52, 19 September 2011 (UTC)

I don't think most people would find 1) to be a problem; 2) is unavoidable, as the factors involved vary from case to case; 3) this is not the place for that kind of detail -- subject-area-specific criteria belong elsewhere. Powers ^T 13:31, 20 September 2011 (UTC)

The first point is the subject of some debate in the discussion archives; clearly it is a problem. The second is avoidable by rewording to enable editors to understand the concepts behind the extant statements without having to read 34 discussion pages. The third point: key factors should belong on the relevant guideline page (i.e. here). Wiki-Ed (talk) 15:09, 20 September 2011 (UTC)

I came here to start a thread about this, and I see one already begun. People are requesting page moves per "primary topic", giving page statistics as the only argument. I think this should be strongly discouraged. We should be basing decisions on something other than page-view stats, but the guideline doesn't really say anything about how to do that.
An example that comes to my mind of a page where the primary topic is not somehow automatically the one with the most page views is Anne Hathaway. I saw people arguing during (and following) a recent move discussion at Anne Hathaway (actress) that a currently popular actress is "obviously", by "common sense", primary over a person with centuries of enduring historical notability, the wife of one of the very most important writers in the history of the English language, and a source of interest and inspiration for dozens of generations of scholars and writers.
Page view stats make it very clear that anyone currently popular on the internet is far-and-away primary over someone of more academic than pop-culture interest. This is not encyclopedic, and I think we should address this directly. I'm not sure exactly how to do it. -GTBacchus^(talk) 19:41, 24 September 2011 (UTC)
- Well you can see this currently in one days nominations at WP:RM. Not picking on the editor that made these nominations or stating an opinion. However this is an good example of what GTBacchus is pointing out. Vegaswikian (talk) 20:33, 24 September 2011 (UTC)
  - People here may want to review this discusion, which led to the intentionally vague "educational value", which seems to cover the Anne Hathaway scenario. Does it not? I'm thinking consensus here is already in agreement with GTBacchus' position, so maybe we only need to clarify the wording, if that. --JaGa^talk 20:56, 24 September 2011 (UTC)
    - I was one of those arguing that the actress is the primary topic, although I really hope I didn't use words like "obviously" and "common sense" (no matter how obvious it may be). I often analyze pageviews to see what most readers of WP are searching for when they use an ambiguous phrase. I also analyze incoming wikilinks to see what articles editors are linking to, and when those two are inconclusive I sometimes go to Google for an appropriate search when possible. The reason is that while none of those are perfect, they are objective. I've seen fundamental differences of opinion as to whether article titles should direct readers to what most people are most likely looking for (objective) or to what is most important or "vital" or educational or historical (subjective). "Historic notability" is not inherently and incontestably more important than "popularity", or vice versa; it's simply a choice. This has been discussed often before and frankly I don't think there is a true consensus as to which way we should go. The primary topic guideline should reflect that, and not "strongly" encourage or discourage any considerations that multiple editors use in determining a primary topic. Station1 (talk) 21:00, 24 September 2011 (UTC)
      - Station1: I disagree with your assertion that statistics are objective. They are not; They reflect systemic and internet bias. Wikipedia will never be taken seriously as an educational resource if it panders to arbitrary Google stats. This article [1] from The Onion illustrates my point. It still makes me laugh, but there is a serious side to the problems it highlights. JaGa: I did read that discussion - which raised some interesting points - and my proposal is that we do indeed go a bit further to clarify the wording. The guidance here might not necessarily encourage/discourage particular considerations (Station1's point), but it should explain how to analyse factors and compare them against one another. Wiki-Ed (talk) 22:19, 24 September 2011 (UTC)
        I agree that statistics require analysis, and perhaps something could be written about that, although I fear that would open a can of worms. Although they may be biased or incorrect or misinterpreted, statistics are objective by definition. Systemic bias has to do with what gets put into WP by its editors, but once an article is in here its viewership can be objectively if imperfectly measured, and primary topic has to do only with topics already covered by WP. Internet bias is irrelevant and inevitable since WP is part of the internet and we serve only internet users. Pageviews (note, not "Google stats"), when properly used, often show what WP readers are seeking. As I said, it can be reasonable to ignore objective measures in favor of subjective criteria, but even assuming that it is a (realizable) goal of WP to "be taken seriously as an educational resource", as opposed to being a popular internet resource, I doubt the naming conventions are a significant factor in that regard. Station1 (talk) 23:23, 24 September 2011 (UTC)
        
        (edit conflict)I think Wikipedia is taken seriously as an educational resource by millions of people. Although I agree with the position you're taking here, Wiki-Ed, there's no need to invoke such ideas. Station1, I appreciate that you avoided the language "obvious" and "common sense"; that's helpful. My issue with the current guideline is that it gives very specific ways to determine what's more popular, but doesn't really address how we can determine what's more important or educationally significant. I think this contributes to people not even taking the latter into consideration. We could encourage a more balanced approach. I'm not sure I see how "historical" = "subjective", either. We can look at sources that address history, and we can do so objectively. -GTBacchus^(talk) 23:25, 24 September 2011 (UTC)
        I understand what you are saying about the current guideline. I think it gives specific ways to determine what's most popular because we can point to specific tools that will give us objective statistics to help determine what is more popular. On the other hand, what is more important or educationally significant is purely a matter of opinion; there are no tools or standard methods that I know to help determine that, other than discussion and consensus. Station1 (talk) 00:00, 25 September 2011 (UTC)
        I wouldn't say it's "purely" a matter of opinion. Insofar as there is any consensus, we can describe it. -GTBacchus^(talk) 03:51, 25 September 2011 (UTC)
        
        I think those of you who are ranting "internet bias" are missing the point of primary topic. Primary topic is "the subject being sought when a reader enters that term in the Search box". Primary topic is not an issue of merit or importance. It's for convenience. We cater to our audience. To do otherwise would defeat the purpose entirely. 99.999% of our readers are accessing Wikipedia from the web. We are not a paper encyclopedia. Marcus Qwertyus 05:29, 25 September 2011 (UTC)
        "Ranting" is a nice choice of word. I see reasonable people making valid points, on all sides of the question. You're not ranting. Neither am I. Let's continue to not rant, nor to accuse others of it. Erring on the side of more respect is always best.
        On what do you base your claim that "primary topic" is not an issue of merit or importance? Why is it only for convenience? Considering how few people access Anne Hathaway (actress) (no more than 6% of her viewers, at the most) via the dab page, for example, I don't see convenience being a major issue. Where and when was it decided that it's all about convenience? -GTBacchus^(talk) 07:16, 25 September 2011 (UTC)
        Convenience is the root of the whole primary topic idea. There would be no "primary topic" if Wikipedia could telepathically know which topic each reader was typing in. On the lack of traffic on disambiguation pages you mentioned, I acknowledge this. The majority of our page views come from clicks on search engines. But if we aren't going to decide primary topic based on convenience, why would we base primary topic on merit or importance if this structure yields no benefit to the reader (from conveniency)? Marcus Qwertyus 08:43, 25 September 2011 (UTC)
        If we're going to base primary topic at least partly on merit or importance it will be for reasons of merit or importance. Let's see what other people think. I think you've made your position quite clear.
        I sure never suggested that telepathy should enter into it; thanks for representing my position that way. Have I tried to disparage or ridicule your position? If so, please give me a chance to apologize. Erring on the side of more respect is always best. -GTBacchus^(talk) 12:09, 25 September 2011 (UTC)
        No one is ridiculing anyone. I just like posing my thoughts as questions. Marcus Qwertyus 14:01, 25 September 2011 (UTC)
        "There would be no "primary topic" if Wikipedia could telepathically know which topic each reader was typing in." That's not a question. It's apparently an attempt to suggest that someone, somewhere, is suggesting that Wikipedia "telepathically" know what people are looking for. Nobody suggested that, and that is ridiculous, hence my use of the term "ridicule". If that wasn't an attempt to equate any argument with something ridiculous, then I guess I don't get what was going on in that sentence. -GTBacchus^(talk) 14:25, 25 September 2011 (UTC)
        "Convenience is the root of the whole primary topic idea." If WP could know what you were looking for automatically, the concept of primary topic would be obsolete. Just trying to back up my claim that navigation is the whole reason behind Primary topic.Marcus Qwertyus 15:05, 25 September 2011 (UTC)
        "If WP could know what you were looking for automatically, the concept of primary topic would be obsolete." I don't see the relevance of that. Nobody is talking a situation where Wikipedia knows anything automatically. Given that we don't know anything automatically, do we try to cater to convenience alone, or to other considerations as well? Answer: we do what consensus supports. -GTBacchus^(talk) 16:18, 25 September 2011 (UTC)
        
        Separately, there's something very odd about the question (which is a question) "If we aren't going to decide primary topic based on convenience, why would we base primary topic on merit... if this yields no benefit to the reader (from conveniency)?" If we aren't basing it on convenience, then I don't understand questioning the convenience-value of other considerations. That's a weird kind of switch. It's like asking "If we aren't going to serve bacon, then why would we serve something that has no nutritional content (as bacon)?" That last parenthetical is very strange, considering the initial clause. -GTBacchus^(talk) 14:25, 25 September 2011 (UTC)
        What benefit does the proposed naming scheme have to the reader if it just takes you to the wrong article? Marcus Qwertyus 15:05, 25 September 2011 (UTC)
        I'm not sure what "proposed naming scheme" you're talking about. I'm proposing that the guideline, and our actions, reflect consensus. No matter what naming scheme we choose, some people are taken to the "wrong article". The question is whether minimizing that number should be our only consideration. I don't believe that there is consensus support for that position. -GTBacchus^(talk) 16:18, 25 September 2011 (UTC)

Replying to no one in particular, I'll reiterate what I've written in previous discussions. When there is ambiguity in the titles of topics, a primary topic essentially means that one topic is so much more likely to be expected to be found at a particular title that it is worth inconveniencing anyone who might happen to be looking for one of the other topics. Personally, I think that bar should be relatively high and the guideline should have a stronger default to having a disambiguation where there are conflicting indications. older ≠ wiser 23:03, 24 September 2011 (UTC)

It's not clear that inconveniencing people is an issue here. In the Anne Hathaway case, lots of people talked about this, but very few noticed that almost nobody looking for the actress ever has to expend a second click, due to internal links, Google searches, and auto-complete in our own search box. It's not about convenience, one way or another. -GTBacchus^(talk) 23:25, 24 September 2011 (UTC)

I very much disagree with that. 43,659 viewers hit the Anne Hathaway dab page in August, of whom roughly 93%, or 40,000, wanted Anne Hathaway (actress). That's not almost nobody. One can argue that the deliberate decision to slightly inconvenience tens of thousands of readers per month is worth it for more important reasons, but not that they don't exist. Station1 (talk) 01:37, 25 September 2011 (UTC)

Huh. I went back to July, to avoid the month of the page move discussion itself, which would certainly skew the statistics. I see that in that month 23,150 people viewed the disambiguation page, while 26,600 people viewed Shakespeare's wife. By what argument can we say that most of those viewing the dab page were looking for the actress? How can you tell that 93% of the people visiting the dab page wanted the actress? Are you seeing some stats I don't know about? -GTBacchus^(talk) 03:56, 25 September 2011 (UTC)

Although I didn't find any diffs for this, sometimes the disambiguation page links are deliberately pointed away from the actual articles to see how many people clicked certain links. Marcus Qwertyus 05:29, 25 September 2011 (UTC)

That is one way to learn more about reader behavior. I'm not aware that it's been tried in the case under discussion here. -GTBacchus^(talk) 07:17, 25 September 2011 (UTC)

To answer GTBacchus: 23,150 readers viewed the dab page in July. 379,222 viewed Anne Hathaway (actress). 26,630 viewed Anne Hathaway (Shakespeare). It is true that most people who access any article on WP do not come through a dab page; those people are irrelevant when considering pageviews except as evidence of the proportion seeking each article. It's also true that we do not know directly how many clicks come off a given dab page per month, but it is logical to assume that the proportions will be roughly the same as those who come via a different route. Since 93% of all the people seeking Anne Hathaway land on the actress's article, it's reasonable to assume a roughly similar percentage of 23,150 people per month click on her article after reaching the dab page. Even if we discount that to, say, 80%, it means 18,520 people in July were inconvenienced by the current set-up. Additionally, if even a bare majority of readers landing on the dab page clicked through to Anne Hathaway (Shakespeare), it would mean that 43% of her readers came through the dab page but only 3% of the actress's readers came through the dab page, which seems extremely unlikely. Station1 (talk) 05:54, 25 September 2011 (UTC)

Station1: "it is logical to assume that the proportions will be roughly the same as those who come via a different route". I don't see the logic behind that assumption at all. How do people get to the dab page? If you Google Anne Hathaway, then the top Wikipedia hit is to the actress' article. If you click on an internal link, you go straight to the article you're interested in. If you type "Anne Hathaway" into the search box, you see your auto-complete options. I suspect that a lot of visitors to the dab page are there because they're curious to whom or what else "Anne Hathaway" may refer. That's the only reason I'd be there.

You say it "seems extremely unlikely" (extremely subjective!) that very few of the actress' readers come through the dab page, but look how many internal links point to her, versus the historically notable Anne Hathaway. The vast, vast majority of readers about the actress come via Google, via internal links, and via the auto-completed text in the search box. I find it extremely easy to believe (yes, subjective!) that 3% or fewer arrive via the dab page. Even if all the dab page viewers are looking for the actress, that's only 6% of her page's views. The vast majority never see the dab page, no matter how you cut it.

As for the historically notable Anne Hathaway, I expect that a significantly larger proportion of her traffic is driven via the dab page. She's not getting as many visitors, and she's liable to get views from the curiosity of readers who wonder who besides the actress is called "Anne Hathaway". The actress probably gets views in the same way: someone looking for Shakespeare's wife wonders who else is called "Anne Hathaway". It is very easy for me to believe that the historically notable figure gets a much higher percentage of views via the dab page than the currently popular actress.

All of this aside, what do you have to say about the actual topic of this thread? I suggest we take the argument about Anne Hathaway to a more relevant page, and address here how this guideline can inform editors about primary topic decisions. What are your thoughts on that issue? -GTBacchus^(talk) 07:10, 25 September 2011 (UTC)

I think it's understood by all that we're both using Anne Hathaway just as an example, per your first bullet above (no need to discuss the specific case again elsewhere). And I think it's a pretty good example to show the competing values of different editors. As stated at the top of this section, some editors see an article title as mostly a navigation aid to get most people where they want to be with the least inconvenience, to the best of our ability; those editors would make the actress the primary topic with a hatnote to Shakespeare's wife. For a long time, that postion was the consensus, or at least that's what was reflected in the guideline. I think a majority of active RM editors probably still land in that camp to varying degrees, and therefore I'm not in favor of your proposal to strongly discourage using statistics as arguments when discussing page moves. But I also recognize that other editors think other considerations can be equally or more important. I wouldn't discourage those arguments either. The guideline should only reflect consensus, not try to impose rules. Station1 (talk) 08:53, 25 September 2011 (UTC)

I'm more opposed to rules than 99% of editors on this site. I don't think the guideline should reflect anything other than consensus, which is what this site is based on. Consensus is one of our 5 pillars; convenience is not. If there is a broad consensus of Wikipedians to base primary topic purely on convenience, then I'll assent to that. So far, I don't see it.

It's not clear to me that there's consensus support for giving very concrete guidelines for using statistics to determine primary topic, while giving almost no concrete guidance as to what is meant by: "An exception may be appropriate when recentism and educational value are taken into account, especially if one of these topics is a vital article. In such a case, consensus may determine that the article should be treated as the primary topic regardless of whether it is the article most sought by users."

If such a consensus has emerged in multiple cases, which the quoted text seems to suggest, then perhaps it would be acceptable to give a bit more detail? -GTBacchus^(talk) 12:15, 25 September 2011 (UTC)

I think we need rules for consistency. Consensus various from topic to topic. I was going to propose we say something like: 'Wikipedia is not Google and does not duplicate its function; Wikipedia does not need to arrange its structure to correspond with the way that Google's search engine directs people to Wikipedia articles. Google uses some algorithm to calculate popularity. We do not need to use this method because Google does it for us. We should not use it because:

Usage changes over time and if followed exactly would lead to chaos;
It reflects internet bias (previous discussions have already raised hypothetical examples of a new band calling itself "Jesus" or Justin Bieber changing his name to "water"); and
The first and foremost purpose of an encyclopedia is to educate (that's where the word comes from) and that is why core policies (NPOV, V, NOR) focus on this rather than accessibility issues.

The fact that Wikipedia happens to sit on the web does not mean that its educational value should be compromised in order to save a user a single mouse click. Next thing someone will be suggesting having "like" buttons to determine article quality. Wiki-Ed (talk) 09:42, 25 September 2011 (UTC)

I couldn't agree more.--Ykraps (talk) 10:19, 25 September 2011 (UTC)

So we should redirect every article in existence to Cold fusion? They are going to end up at the article they wanted regardless of what you force them to view so why force them to make the extra clicks? Wikipedia is supposed to be impartial. No knowledge is more or less important to another. Marcus Qwertyus 10:29, 25 September 2011 (UTC)

Yes, that's precisely what people are saying. Cold fusion. That's man's made of straw; there's no way he'll fight back. If no knowledge is more important than any other, what's the point of WP:VITAL? -GTBacchus^(talk) 12:09, 25 September 2011 (UTC)

WP:Vital is about articles that are empirically or hierarchical (i.e. Actor is empirical to Anne Hathaway (actress)). Plus I don't think that was ever written as a guideline or policy. The argument Wiki-Ed seems to be making is that we should be in-impartially ranking articles by perceived educational value. This project is an attempt to document all the world's knowledge not go on some crusade to educate the public. Marcus Qwertyus 14:01, 25 September 2011 (UTC)

I think the purpose of this project is precisely to educate the public. At least, I think a large proportion of Wikipedians would say it is. -GTBacchus^(talk) 14:25, 25 September 2011 (UTC)

If I type in Robin Wright and land on Robin Wright (author) I'm going to just go strait to the actress's article regardless of what you believe is educational. So why not save me a few clicks? Marcus Qwertyus 15:28, 25 September 2011 (UTC)

My opinion of what's educational has nothing to do with anything. We decide nothing based on my opinions, and everything based on consensus. I make no argument against consensus. If consensus supports the all-for-convenience viewpoint, then I will complete moves accordingly. So far, that's not the case. -GTBacchus^(talk) 16:18, 25 September 2011 (UTC)

A very interesting discussion. I don't think the guideline ever advocated "ease of navigation" as a reason for primary topics. I think that rationale has been derived from the principle of least surprise, which was explicitly the basis behind primary topic, both at this guideline and at the article title (formerly naming conventions) guideline page. But I think the principle of least surprise encompasses more than just ease of navigation or page view statistics. I've always felt page view statistics are the weakest indicator of primary topic, though unfortunately because they are the most easily obtained evidence they often are given inordinate weight in discussions. IMO, the mission of wikipedia is to create a free encyclopedia, not to create a web site that perfectly optimizes page views. Determination of whether there is a primary topic is a matter of consensus and a variety of factors should be taken into consideration. Where there are conflicting indications, as between a historically significant subject versus a contemporaneous hot topic, I think would suggest that a disambiguation page is appropriate rather than designating one as "primary". older ≠ wiser 13:28, 25 September 2011 (UTC)

That's the solution we settled on in the Anne Hathaway case. -GTBacchus^(talk) 00:07, 26 September 2011 (UTC)

It wasn't. There was no consensus in that discussion, "so we'll fall back on our usual practice of not moving an article in the face of significant opposition", according to the reasonable decision of the closing admin (i.e., you). Station1 (talk) 02:51, 26 September 2011 (UTC)

Hmm. I guess that is a more fair description. I concur, and apologize for letting my memory be colored by my opinion in the current discussion. The effect of the lack of consensus is that we didn't select a primary topic there, but you're right about how it came about. -GTBacchus^(talk) 07:49, 26 September 2011 (UTC)

Made an edit

Here. I think it's accurate, anyway. Thoughts? -GTBacchus^(talk) 20:25, 25 September 2011 (UTC)

A comment on the section about links, but not an objection to your rewording. In my mind, links can actually be a better case for why something is not the primary topic. Using them to establish primacy is more problematic. I wonder if an appropriate rewording to reflect this would have consensus? Vegaswikian (talk) 22:15, 25 September 2011 (UTC)

I think that's a really good start. I was going to start drafting something, but got bogged down by the bit you've just rewritten. However, I think we also need to look at expanding the section on the factors that editors should use to consider cases, as per my first post. If no-one else does it in the meantime I'll have a look at that tomorrow. Wiki-Ed (talk) 23:03, 25 September 2011 (UTC)

I agree there's more to say, and I look forward to reading more of your thoughts on the matter. It was your original post that inspired half of what I wrote. -GTBacchus^(talk) 23:53, 25 September 2011 (UTC)

Whither recentism? --JaGa^talk 08:27, 26 September 2011 (UTC)

Ah. I didn't care about losing the link, because it's... how do people say?... "just an essay", and people don't take it very seriously. That's only half (or a third) tongue-in-cheek. If there's a nice way to re-add it, then please feel free.

Every time someone says "X is just an essay by Y is policy", God kills a kitten. When it's said in response to someone using the word "X" without linking to - or thinking about - the essay, it's a whole litter of kittens, plus two baby seals and a hard-working pastry-chef. -GTBacchus^(talk) 14:50, 26 September 2011 (UTC)

Strong Oppose. The definition of primary topic has always been stated relative to usage in reliable sources. Creating a competing and especially vague definition based on something as nebulous as "importance" (even stated as "significantly greater enduring notability and educational value") accomplishes nothing except opening the door even more for JDLI rationalizations. ~~If this sticks, just watch the RM backlog grow even more.~~
We should be striving to refine our rules to reduce editor strife. Changes like this, which make them more vague, is just throwing fuel on the fire. Bad, bad idea. --Born2cycle (talk) 19:50, 26 September 2011 (UTC) --claim stricken. --Born2cycle (talk) 20:06, 26 September 2011 (UTC)
- You just made a prediction. Bet you $10? I'm incredibly certain that this will not have any noticeable effect on the RM backlog. Make it $100? -GTBacchus^(talk) 19:55, 26 September 2011 (UTC)
  - No, I'm not certain it will grow the backlog. It is more likely to create more "no consensus" decisions because it gives reason favoring either side for any RM proposal that involves a name that can refer to a topic with wide usage in sources as well as a different one that can be argued to be most important. The longstanding original definition provided clear guidance on what to do in such situations; the current wording provides none. --Born2cycle (talk) 20:06, 26 September 2011 (UTC)
    - You seem to think that I'm making an edit that will change how people behave. This is the common superstition about policy... People already behave this way. I'm trying to document it accurately, instead of the inaccurate documentation that was there before. People empirically have been considering the criteria I described since the site's inception. Do you disagree?
      "The longstanding original definition provided clear guidance on what to do in such situations; the current wording provides none." The clear guidance you claim we provided has always been ignored as necessary. All we're doing now is admitting that. -GTBacchus^(talk) 21:20, 26 September 2011 (UTC)
      - It's circular: behavior influences policy and policy influences behavior. No matter what rule you choose, you will be able to find some number of cases where that rule is ignored and/or some other rule is used. We already account for that with WP:IAR. But if you also use such cases to justify "loosening" the rule, then loosening that rule will cause even more cases of the exceptional behavior.
        There is a good reason we limit how much we determine criminal law based on behavior. I mean, making a point with an admittedly very extreme example, by rules-follow-behavior reasoning, murder and rape should be lawful because people kill and rape.
        The whole point of policy and guideline is to help editors decided what to do - to guide them - in areas where guidance is needed. This section provides guidance for the process of determining primary topic. Before this change the guidance was clear regarding what to do when one use meets the traditional criteria (the most likely to be sought) while another use is perceived to be more important: go with the traditional criteria. Now, that doesn't mean in some cases a consensus can't invoke IAR, and they do, but by explicitly giving that IAR condition equal status (if you will) in the policy, you've effectively removed any guidance, and the whole point of having this section, at least in the cases where "importance" can be even dubiously argued to conflict with the traditional criteria, is lost. --Born2cycle (talk) 23:13, 26 September 2011 (UTC)
        You're wrong, and that's why neither consensus nor practice support your theory. Watch what happens, and learn. Or perhaps you're going to teach us all about "rule of law"? -GTBacchus^(talk) 00:23, 27 September 2011 (UTC)
        You call that a response to what I wrote? Apparently, there is no reasoning with you. --Born2cycle (talk) 00:44, 27 September 2011 (UTC)
        I'm tired of the same-old same-old, Born2cycle. Experience supports what I've been doing for years: helping to smoothly run the site in accordance with consensus, and documenting practice in guideline pages. You've been predicting chaos if we don't become rule-bound for a long time, and it hasn't happened, and it's not going to happen. Your ideas about the purpose of policy are not supported by consensus, and they never have been. -GTBacchus^(talk) 01:14, 27 September 2011 (UTC)

Proposed guidance

I propose we split this section to reflect how we differentiate meanings of the term "primary" and also include a reminder of the key guidelines about writing neutrally. The factors I've listed only reflect the discussions I've been reading across the project; there may well be many others.

Usage
- Incoming wikilinks from Special:WhatLinksHere
- Wikipedia article traffic statistics
- Google web, news, scholar, or book searches (NOTE: adding &pws=0 to the google search string eliminates personal search bias)
Importance
- Educational value: Certain topics are academically important and should take precedence over pop-culture uses of the same term, particularly when the meaning is different. This factor should be used to counter Systemic bias. For example Madonna is a disambiguation page because the term has greater religious, historical and artistic significance than the modern entertainer whose article attracts significantly more page views than all the other uses combined.
- Historical legacy: Certain topics are perennially important or are the primary source of derivatives which share the same name. This factor should be used to counter Recentism. For example, the city of York is historically important and occupies the namespace even though there are many places/people/objects which have borrowed the name in recent centuries. This factor is also relevant to artistic works where recent multimedia derivatives may gain more page views than the original work upon which they are based (for example The Return of the King).
- Pre-eminent contribution (use Churchill example)
- Population (Ireland example? Could be a tricky one to include and would need careful wording)
Contrary factors
- Systemic bias, including (specifically) internet bias
- Geographic imbalance
- Recentism

(Or should this section link to the projects devoted to counter measures, e.g. Wikipedia:WikiProject Countering systemic bias?)

It's getting a bit late here so for now this is just a framework to illustrate where I think we should be going. Comments? Wiki-Ed (talk) 22:26, 26 September 2011 (UTC)

I've been tweaking the section above since I originally posted it. It has been suggested that it might be wise to draw attention to this fact since the discussion has rolled on without reference to what I was proposing, hence inserting a new section break below. Wiki-Ed (talk) 09:58, 30 September 2011 (UTC)

Yep, that's a problem with trying to move the goal posts in the middle of a discussion. You might bullet list the diff links to your edits that changed the proposal while the discussion was going on. And if you later feel more edits are needed, add the new proposal at the end of the current discussion, so that readers will be aware of the sequence. -- JHunterJ (talk) 11:36, 30 September 2011 (UTC)

I even posted this bit in the wrong place. Doh. The section above is the one I've been adding to. I haven't been moving the goal posts, just fleshing out the points I'd already made. Wiki-Ed (talk) 12:45, 30 September 2011 (UTC)

Discussion

The edit discussed above, and this, is all asinine, for reasons I've given above. None of this providing guidance. It's describing chaos as policy, which makes the policy chaotic. That's not guidance. It's, well, asinine. --Born2cycle (talk) 23:43, 26 September 2011 (UTC)

Asinine, very good. Get consensus support for your theory, Born2cycle. We'll keep doing what's succeeded here, and we'll keep succeeding. Your predictions of chaos have never come true, because your theory has never been right.

Are you badlydrawnjeff? -GTBacchus^(talk) 00:25, 27 September 2011 (UTC)

I'm not badlydrawnjeff. Never heard of badlydrawnjeff, until now.

By chaos I mean "unpredictable outcomes", and I'd say that already applies to a significant number, perhaps a majority of RM decisions.

Here's a challenge, start a subpage in your user account where every day you review all of the new RM requests for that day and predict their outcome: move, consensus to not move, or no move due to no consensus, and we'll see if you can bat 500. I bet you can't. That's what I mean by chaos, and policy changes like this will only make RM decisions more chaotic (less predictable). --Born2cycle (talk) 01:08, 27 September 2011 (UTC)

You have heard of him, because I've mentioned him to you before. It was when you started talking about "rule of law" on ANI before leaving the site for a while recently. I suggested that you look at his experience and learn from it. I still suggest that. I make no claim that I can predict what consensus will say, and there's no need for me to do so. My batting average at RM is that the vast majority of my closures are never challenged. Those that are, I invariably take to AN for review, and I've yet to have one overturned.

Thinking about your challenge, though, I could bat over .500. Wanna put money on it? I do. I could use the money. How much, Born? Put up the money, and I'll bite. -GTBacchus^(talk) 01:21, 27 September 2011 (UTC)

Policies on article content should not be simply extrapolated to navigation pages. Coverage of various topics can be given without bias while still directing the audience to the likeliest target with navigational aids to get everyone to their unbiased coverage. -- JHunterJ (talk) 00:42, 27 September 2011 (UTC)

I've never seen that consensus supports the "likeliest target" in all cases position. -GTBacchus^(talk) 00:47, 27 September 2011 (UTC)

Who said anything about "the "likeliest target" in all cases position."??? Like I said above (which you ignored), we already have IAR based exceptions, which requires good reason to choose an alternative title (which importance is particularly strong), and sometimes that's a consideration. But to give "importance" equal weight to "most likely to be sought"? That's totally new, and not supported in practice, by the way. --Born2cycle (talk) 01:08, 27 September 2011 (UTC)

Who said anything about "equal weight"? Nobody's revisions or suggestions have said this. We've simply affirmed that there are two kinds of "primary topic" that people talk about. When IAR is used repeatedly, then we write it down so people don't have to consider it an IAR invocation anymore. This has always been the case. -GTBacchus^(talk) 01:21, 27 September 2011 (UTC)

Rewording, since the point was dodged. Ahem, "Policies on article content should not be simply extrapolated to navigation pages. Coverage of various topics can be given without bias while still directing the audience to the appropriate target with navigational aids to get everyone to their unbiased coverage." -- JHunterJ (talk) 02:09, 27 September 2011 (UTC)

I didn't mean to "dodge" this point, but I admit that I didn't understand it. I don't see why you bring up "unbiased" coverage, because I don't see anyone suggesting a danger of bias one way or the other. I'm not sure who's suggesting that anything be "extrapolated to navigation pages". Nobody's opposed to "directing the audience to the appropriate topic", either. There's just disagreement on whether the "appropriate topic" should be defined as the one sought by the greatest numbers. -GTBacchus^(talk) 05:08, 27 September 2011 (UTC)

The "Contrary factors" in the opening list are all assumed from article content guidelines, but disambiguation pages are not articles. The article content can be made unbiased while still permitting the navigational pages to best serve the readership in getting to whichever topic they are most likely, or most appropriately, seeking. Trying to counter encyclopedia bias is one thing; trying to "correct" perceived readership bias is another. If the readers are biased toward the latest pop singer over a medieval monk of the same name (I don't have one in mind, just making that up), then we shouldn't try to ensure they first see the more "important" monk before allowing them to navigate to the article sought. The goal of disambiguation has been to deliver the likeliest article to the reader; that has been the long-standing consensus, with multiple criteria (not just traffic) since we cannot yet read their minds, and not in all cases, since no matter what criteria we put here some local consensuses will continue to ignore it in favor of a less likely but favored target. -- JHunterJ (talk) 11:51, 27 September 2011 (UTC)

IMO, this is bogus. Content guidelines should apply to disambiguation pages as well as to articles. There might be some qualifications based on their function to aid navigation, but I don't there should be a blanket exemption for disambiguation pages. I'm not sure about geographic imbalance, but recentism and systemic bias can both affect the placement of disambiguation pages and should be taken into consideration. older ≠ wiser 13:17, 27 September 2011 (UTC)

If we apply content guidelines to disambiguation pages, then there can never be a primary topic, per WP:NEUTRAL. The ambiguous title must neutrally cover all ambiguous topics. IMO, that is bogus, as is extrapolating encyclopedia anti-bias concerns ("the encyclopedia should cover both recent and non-recent topics!", which is WP:RECENT) to reader anti-bias impositions ("the encyclopedia readers should be just as likely to seek recent and non-recent topics!", which is not WP:RECENT). If the readers are wildly more interested in a recent topic or a topic more closely aligned to the English-speaking geography, why should we try to "fix" them by interposing some other, more "important" or "unbiased" topic or disambiguation page in their way first? -- JHunterJ (talk) 15:08, 27 September 2011 (UTC)

That's not even remotely close to what WP:NEUTRAL demands. This isn't about "fixing" readers, it is about creating a free encyclopedia, and the most important part of that is the participation of a wide range of editors and editorial discretion in all aspects of the encyclopedia. Content guidelines are an attempt to describe the application of such editorial activity and disambiguation pages most certainly fall within that purview. older ≠ wiser 15:15, 27 September 2011 (UTC)

I disagree: using navigational pages to direct readers to articles other than what they are (most likely) seeking is about fixing readers. Once they get to the topic they are looking up, then they have reached the encyclopedia content level. -- JHunterJ (talk) 15:31, 27 September 2011 (UTC)

This guideline is as much about what to do when there is ambiguity in article titles as it is about navigation and as such it is absolutely part and parcel of other content guidelines. Pretending that disambiguation pages are not part of the encyclopedia content is disingenuous. older ≠ wiser 15:51, 27 September 2011 (UTC)

I am not trying to be disingenuous. I am genuinely suggesting that the content be made unbiased per the existing content guidelines and the readers then be delivered the content they seek in the most efficient manner possible, even if that means that we serve up more articles on recent topics than on non-recent ones, despite having equal coverage of both. Navigation pages (redirects and disambiguation pages) are absolutely not articles (content). -- JHunterJ (talk) 16:59, 27 September 2011 (UTC)

And what guidelines describe what content be delivered to readers under a particular name? That would be this guideline (along with Article titles). This guideline does not operate in a vacuum and it is disingenuous to try to exempt it from content guidelines. There might be some qualifications based on the function of the pages, but they are not exempt. older ≠ wiser 17:04, 27 September 2011 (UTC)

The article title guidelines determine what title to use and, if needed, what disambiguator to append. Those two questions are indeed independent (not in a vacuum, but in an encyclopedia) of the decision of which of multiple ambiguous-title topics, if any, to first serve up to a reader based on their request. And let me make my previous implied request explicit: stop ascribing disingenuousness to me; it is incorrect and insulting. -- JHunterJ (talk) 17:13, 27 September 2011 (UTC)

Fair enough, my apologies for some careless invective. You are sincere, even though I think you're wrong. It is highly artificial to separate consideration of primary topic from that of article title. They are two sides of a coin. The determination that there is a primary topic often affects the title of that article (and perhaps other articles as well). older ≠ wiser 17:19, 27 September 2011 (UTC)

Thanks. The determination of a primary topic affects only the disambiguator portion of non-primary topics. And what that portion is or would be is determined separately from whether it is needed in the first place. -- JHunterJ (talk) 17:22, 27 September 2011 (UTC)

Not necessarily. Natural disambiguation is also used. Whether a title has a disambiguating phrase or not is a non-trivial issue. And the factors to consider in determining primary topic are not deterministic and require editorial judgement which should draw upon content guidelines. older ≠ wiser 17:37, 27 September 2011 (UTC)

Same effect here, whether the disambiguator is natural or parenthesized. If (a) the article would have one title if there were no ambiguity and (b) needs another one because it is not the primary topic of the ambiguous title, then (c) it needs a disambiguator (natural or parenthesized). I never said that was trivial. I do still see that selection of disambiguator (step c) as independent of the determination of the need for a disambiguator (step b). A disambiguator could be identified in advance of the determination (and then not used if not needed), and the disambiguator selected by the naming conventions of the article project would not change based on the need for a disambiguator. The factors for determining primary topic, which I agree are not deterministic and unlikely to become so, require judgement. That judgement, though, has nothing to do with the judgement of the content of the article (content which will be unchanged) and the guidelines for content, nor with the judgement for the selection of disambiguator (the presence of which may change, but not the wording) and the guidelines for the wording of the disambiguator. The need for a disambiguator and the selection of a needed disambiguator are independent (non-trivial, but independent). -- JHunterJ (talk) 18:27, 27 September 2011 (UTC)

I find this simply baffling. How is an article title not content? How is it possible for editors to work on pages that affect how readers experience an encyclopedia exempt from guidelines describing content. It makes no sense to me to create such an artificial distinction. older ≠ wiser 18:31, 27 September 2011 (UTC)

OTOH, I find it baffling to call the name part of the content. The content of a book is not its title, the body of an email is not its subject, the name of an animal is not its being, and the content of an article is not its title. The distinction between naming and content is very natural. Don't judge a book by its cover, and a rose by any other name would smell as sweet. -- JHunterJ (talk) 18:36, 27 September 2011 (UTC)

Choosing the name of a book is a highly significant part of the editorial process and can have a huge impact on the success of a book. It is true that it is a distinct component of a book (or other work), but it is nonetheless a part of what that book or work is. For an encyclopedia topic, the relationship is, I'd argue, even more integral. Determining whether there is a primary topic requires examination of the content and is not and never has been a numbers game or a matter of simplistic optimization of page traffic. older ≠ wiser 19:05, 27 September 2011 (UTC)

I'm still not calling it trivial. It is part of the work, but not part of the content. The numbers and traffic are irrelevant here. If the criteria for primaryness remain unchanged (or if they change), that decision is still independent of the content of the articles. And, as separate from the article content, attempting to force extrapolation of content guidelines to non-content aspects is not warranted or necessary. -- JHunterJ (talk) 19:12, 27 September 2011 (UTC)

There is no forcing or extrapolation required. It is in fact more work to keep a boundary of artificial separation between them. older ≠ wiser 19:17, 27 September 2011 (UTC)

Other proposed guidance

My own opinion has not changed since the last time this came up: it should be reduced to a simple traffic analysis, with guidance on how to experimentally get traffic results in the cases where test redirects are needed. Continuing to add more complications on top of the existing complication additions will lead to further additions of complications later. -- JHunterJ (talk) 00:42, 27 September 2011 (UTC)

I think that's a terrible idea, that will never receive consensus support, thank Zeus. No complications are being added, or being considered for adding. Reality is being reflected. -GTBacchus^(talk) 00:46, 27 September 2011 (UTC)

I agree that reducing primary topic to a simple traffic analysis is a terrible idea. Though I think in might be beneficial for the concept of specialized links on disambiguation pages to help gauge traffic to be articulated in the guideline. It often provides interesting results for consideration and can help decide between primary topic or disambiguation page in some extreme cases. older ≠ wiser 01:43, 27 September 2011 (UTC)

Especially if you have a case with bad links driving those views. Vegaswikian (talk) 01:54, 27 September 2011 (UTC)

There will be cases where local consensus would properly WP:IAR in favor of doing the right thing (or popular thing anyway), and we wouldn't be any worse off than with the current muddle of un-resolving guidelines. -- JHunterJ (talk) 02:09, 27 September 2011 (UTC)

When IAR is invoked repeatedly for the same reason in similar circumstances, we often try to write those circumstances down, to alert people as to what often happens. -GTBacchus^(talk) 05:10, 27 September 2011 (UTC)

I thought we already wrote in an exception for certain highly important topics that may be briefly overshadowed by another topic's ephemeral popularity. (The Hathaway case was not such a case, IMO, but that's neither here nor there.) I don't see why anything more than that is needed. Getting the reader to the topic they're interested in is the primary purpose of disambiguation pages, and if we can shortcut that for a large proportion of readers, we should. Powers ^T 13:25, 27 September 2011 (UTC)

Is that the consensus position? That's all I'm trying to reflect. -GTBacchus^(talk) 14:06, 27 September 2011 (UTC)

It is the most recent consensus, yes. -- JHunterJ (talk) 15:03, 27 September 2011 (UTC)

Evidence? Where was this consensus established? -GTBacchus^(talk) 15:21, 27 September 2011 (UTC)

The project page history. The talk page archives. Unless you're contending that the guidelines didn't reflect the consensus. In which case: evidence? -- JHunterJ (talk) 15:28, 27 September 2011 (UTC)

That's easy. The current discussion. Also, each case where Wikipedians have chosen to "IAR" and use primary importance to determine primary topic. Also, each case where Wikipedians have chosen to "IAR" and not grant primary topic status to a topic of primary usage, such as Anne Hathaway (actress). Every time Wikipedians go against a guideline, it's prima facie evidence that the guideline does not enjoy complete consensus support. When we can determine a pattern to our rejection of a guideline, it's pretty clear that the guideline is incompletely written, so we update it. -GTBacchus^(talk) 15:33, 27 September 2011 (UTC)

You misinterpret why I mean by "most recent consensus". And ignoring a rule in a specific case doesn't change the consensus in general. Otherwise, yes, I agree, when we reach a new consensus, we update the guidelines. This discussion-in-progress does not yet reflect a new consensus. -- JHunterJ (talk) 17:01, 27 September 2011 (UTC)

It reflects a lack of existing consensus, and from the trenches of Requested Moves, I can report that there's nothing new being said here. These attitudes go back years. I'm not talking about a specific case, I'm talking about an identifiable pattern of specific cases. -GTBacchus^(talk) 19:10, 27 September 2011 (UTC)

The reason I raised this is because, as an editor, I find the existing guidelines are insufficient. It is a fact, currently, that there is no consensus across a wide range of topics and traffic is not used universally as the determining factor. This guideline needs to be updated to reflect what sorts of factors editors do use to decide primacy and place them in a logical framework. Wiki-Ed (talk) 21:23, 27 September 2011 (UTC)

I would humbly suggest that the primary purpose of disambiguation pages is to disambiguate, not navigation. I agree with GTBacchus that the consensus is against deciding primary topic on page hits alone and the evidence can be seen in talk page discussions. Furthermore, as a large number of requested moves cite page hits as a reason; I believe there will be less work to do at RM, not more as one editor supposed.--Ykraps (talk) 07:31, 28 September 2011 (UTC)

You are correct in that it has never been about page hits alone -- that's my suggestion and unlikely to attain any consensus. You are incorrect about navigation -- we disambiguate ambiguous Wikipedia topics in order to navigate the reader to their sought topic. This disambiguation is navigation. That's why we only list Wikipedia topics, not other topics (dictionary definitions, editors' non-notable companies) that editors sometimes try to disambiguate without navigational purpose. It has be consensusfully about likelihood (with discussion informed by page hits, Google book/scholar/news/web hits, and incoming Wikilinks) for some time, with a relatively recent addition for various definitions of educational value. -- JHunterJ (talk) 10:58, 28 September 2011 (UTC)

What I meant was: The primary purpose of disambiguation is to prevent multiple pages occupying the same slot. Taking people to where they want to be may be a secondary purpose but it certainly isn't the primary purpose. There is no necessity to take someone to the ~~least surprising~~ most sought after subject, it is just something we choose to do. And not always with consensus, as far as I can work out.--Ykraps (talk) 16:07, 28 September 2011 (UTC)

We could disambiguate article titles without creating disambiguation pages. The disambiguation pages, the list of articles that have been disambiguated, is a navigational tool. It allows the reader to efficiently reach the page-with-ambiguous-title-plus-disambiguator easily. That is the primary purpose of disambiguation pages, and it's navigational. Selecting a primary topic determines in what order one topic and the corresponding disambiguation page will be encountered: dab first (no primary topic) or topic first (primary topic). Still navigational. -- JHunterJ (talk) 17:53, 28 September 2011 (UTC)

While you and perhaps some others might like to see this as purely navigational, there has been abundant discussion suggesting that the selection of a primary topic is a decision with ramifications beyond mere navigation. Of course navigation is a core function of disambiguation, but the disambiguation process and the guidelines encompass more than just navigation. older ≠ wiser 18:35, 28 September 2011 (UTC)

You say "of course navigation", Ykraps says "not navigation". I'm trying to close that gap. The process has become enhanced or bloated, depending on your perspective, with additional considerations, but disambiguation pages and redirects are non-article navigational tools, and adding enhancements (or changes to the priority given to some articles over others in selecting a primary topic in front of the tools) doesn't change that. -- JHunterJ (talk)

I, for one, agree with everything JHJ has said so far in this section. I will add this: it is my impression that because of the name of "primary topic", at least some people have mistakenly interpreted it to mean "most important topic", when it was never intended to mean that. Unfortunately, I can't think of a better name that more accurately means, "most likely to be sought topic". --Born2cycle (talk) 21:56, 28 September 2011 (UTC)

Again I feel obliged to draw your attention to this early version of the disambiguation guidelines[[2]], which clearly states, "If the title clearly has one central most important meaning..............one alternative is to have the full article about the primary meaning under the simple title". So to say it was never intended to mean most important topic", quite simply isn't true and I don't understand why you continue to say it.--Ykraps (talk) 22:52, 28 September 2011 (UTC)

I think this disussion is going around in circles along a set of well worn tracks. Whatever the intention of the guideline, a narrow statistical interpretation of primacy does not reflect reality (literally and in terms of how articles have been named). We can either (1) enforce a strict statistical interpretation; (2) restore the ambiguous wording and leave editors to fight it out; or (3) try to provide a framework for discussions that reflects what is already taking place. The third option is viable and would agree with core editorial policies (particularly those on neutrality). The other two lead to chaos. Wiki-Ed (talk) 23:12, 28 September 2011 (UTC)

Ykraps, "again"? Where have you brought this to my attention before? I was not aware of that early interpretation.

Anyway, as they say, that was long ago and not true. For many years it has been stated clearly only in terms of likelihood to be sought.

Personally, I don't care what it says, as long as the definition is clear. If it's the most sought after topic except when some other topic meets certain "importance" conditions, then let's say that, and be clear about what those conditions are.

Wiki-Ed - I'd like to do better than merely reflect what already takes place, which, in those cases where one topic is supported by being most sought, and another is arguably "more important" (and those are the only cases that matter to this discussion), is chaotic. That's why I think we need to be clear on what conditions need to be present to go with "most important" rather than "most sought". --Born2cycle (talk) 23:54, 28 September 2011 (UTC)

Archive 34, 22.2 (PRIMARY TOPIC word changing proposal/Discussion) 10th paragraph, in direct response to a post left by you on 10th June.[[3]]--Ykraps (talk) 07:19, 30 September 2011 (UTC)

I'm not sure it's clear that there's a set of well-defined conditions that must be present to go with "most important" over "most sought". What we do it let people discuss and arrive at a consensus in cases where there's such a conflict. If we can look at a bunch of examples, we might be able to abstract just where the line gets drawn, but until we can do that, the best we can do is to say that, in the event of conflict, consensus decides. You can call that "chaos", but I call it an honest admission that we haven't yet distilled precise criteria. Such distillation comes from experience.

If these well-defined conditions exist, then I definitely want to know what they are. -GTBacchus^(talk) 01:31, 29 September 2011 (UTC)

Well, that's the problem. If we don't determine and specify what those conditions are, the "consensus" regarding which is used is determined by the preponderance of JDLI arguments on one side or the other. It can only be JDLI arguments because, since we specify no basis (what those conditions are), people can use whatever basis they feel like using. Basically, it will come down to those who believe the more "important" topic is more important because of a supposed nebulous education value vs. those who believe readers are best served when more get to their desired topic with fewer clicks. What we're really saying, effectively, is that in each such case there should be a vote of those who happen to be participating, and the side with the most votes wins. I suggest we stop obscuring reality and just say that so that the contradiction with other fundamental principles in this wording (like WP is not a democracy) is clear. --Born2cycle (talk) 16:22, 29 September 2011 (UTC)

I don't see this as a problem. It is rather precisely the strength of Wikipedia that brings people with a wide range of backgrounds (although there is system bias in this) who discuss and determine how best to handle particular situations that are not amenable to rule-driven criteria. older ≠ wiser 16:30, 29 September 2011 (UTC)

What. He. Said. Also, just because we don't know ahead of time precisely what principles are involved in an argument doesn't make it "JDLI". Have you no faith in the ability of intelligent humans to discern good reasons from pure whimsy? Nobody is calling for a vote. We're calling for intelligent discourse. You can trust intelligent discourse to arrive at decisions that are different from chance or whimsy. The principles may appear somewhat nebulous when we've only dealt with a few examples, but they become clearer over time. Instead of equating reasoned discourse with "JDLI" and "democracy", why not cite a few examples, and then try to abstract the principles involved?

Name three examples of cases where "primary importance" and "primary usage" are in conflict. I've only got one in mind right now: Hathaway. What principles are in play there? I know them, but treat this as an exercise, Born2cycle. What's going on with that distinction? Identify at least two concrete principles. I know you can do it.

You say: "It can only be JDLI arguments because, since we specify no basis (what those conditions are), people can use whatever basis they feel like using." What nonsense. If we don't hold people's hands, and tell them precisely how to think, then they're incapable of intelligence, and we're all infinitely gullible? Really? How about, we're a little bit savvy, and we can tell bullshit from reason? Identify the principles already, instead of repeating over and over again that they don't exist. You're not even trying. Stop complaining and start thinking. -GTBacchus^(talk) 17:45, 29 September 2011 (UTC)

In addition to Anne Hathaway (though again I dispute that "primary importance" is in conflict with "primary usage" there), there's also Syracuse as a major example where the two are certainly in conflict. Powers ^T 18:03, 29 September 2011 (UTC)

Examples, awesome. :D

I think the fact that you (and others) dispute the conflict in the Hathaway case, while I (and others) see a clear conflict, is what makes it a great example. If a conflict is determined by good-faith, intelligent editors conflicting, then there's a conflict. To me (and others), it's screamingly obvious that primary importance indicates Shakespeare's wife, while primary usage indicates the currently popular actress. Oh, and before someone insults participants in this discussion with another link to dubya-pee-jay-dee-ell-eye, we've all read it, and yes, I can enumerate specific, objective criteria. I'd rather see Born2cycle list those criteria, as an exercise in thinking about these things.

What's primary usage for "Syracuse"? -GTBacchus^(talk) 18:12, 29 September 2011 (UTC)

I don't think you're really being fair to B2C here, GTBacchus. JDLI doesn't mean one doesn't have an intelligent reason for not liking something. ErikHaugen (talk | contribs) 18:20, 29 September 2011 (UTC)

If he put 1% as much work into trying to read consensus out of discussions as he does into petulantly repeating that there isn't one, I'd feel a lot more generous towards him. If there are intelligent reasons, then some of them might be able to be abstracted into general principles that, it would turn out, enjoy consensus support. What are the intelligent reasons? That is the question I'm asking Born2cycle to consider addressing. -GTBacchus^(talk) 20:37, 29 September 2011 (UTC)

I agree about the conflict at Anne Hathaway, but, absent consensus on what the guiding basis is for deciding such conflicts, there is by definition no basis to go with the currently popular actress, the wife of the famous writer, or a dab page. It's all whimsy, on both sides. We might as well toss coins to decide these. --Born2cycle (talk) 18:59, 29 September 2011 (UTC)

In both cases, Anne Hathaway is spelled with an 'e'. You've clearly researched this. What are your observations of the principles cited in the discussion there? You have studied that discussion, right? Where is consensus regarding the guiding principles going to come from, except from observations of what arguments people buy, and why? Have you done any work on this question? Observe consensus, Born2cycle, and then tell us what you observed. -GTBacchus^(talk) 20:37, 29 September 2011 (UTC)

I think the answer has to be that, if there is no consensus on a primary topic, then the default should be a dab page at the primary title. --R'n'B (call me Russ) 19:10, 29 September 2011 (UTC)

I would slightly tweak that: if there is no consensus for a change of primary topic arrangement (to or from no primary topic, or from one primary topic to another), then the default is the current state. -- JHunterJ (talk) 19:50, 29 September 2011 (UTC)

More to the point, absent clear guidance, what happens is people rationalize whatever their whimsical preference is, and there is no objective standard by which to judge one rationalization to be "better" than another. Only if we have clear guidance spelled out in policy can anyone say one reasoning is better than another.

Oh, and I can't think of any clear objective criteria that could be used in an effective and useful fashion to distinguish when to go with "historically important" over "much more likely to be sought". I also don't think it matters which one is used, and, to the extent it does, serving readers in terms of getting them to their desired destination sooner is preferred. More importantly, since we do have a reasonably objective means to determine "much more likely to be sought", I favor doing only that. --Born2cycle (talk) 19:08, 29 September 2011 (UTC)

Born2cycle, if you "can't think of any clear objective criteria,..." then you're not trying. I can think of many. Have you studied the discussions in which this has come up, and looked at the arguments that people use? I'm assuming you have, and that you've tried to abstract the principles being cited. Otherwise, what are you talking about?

You have not begun to try to figure out what "primary importance" means to Wikipedians in context, have you? -GTBacchus^(talk) 20:37, 29 September 2011 (UTC)

Yes, I have thought about it. A lot. Again, I cannot think of any clear objective criteria that would clearly tell us, for example, whether Corvette should be the historical ship type or a redirect to the car, and which Anne Hathaway, if anyway, is of "primary importance". On both sides there are reasonable opinions, each of which uses difference criteria and weighs the same criteria differently. Which is best? Which is "correct"? There is no way to determine this, because there is no basis on which to make such a determination, especially since your edit. --Born2cycle (talk) 21:27, 29 September 2011 (UTC)

Why do you think "Anne Hathaway" is spelled without an 'e'? Are you fucking with us? Is this funny to you?

What you've just said makes no sense: "especially since your edit". My edit means nothing! Whether or not there is a basis to make determinations of primary importance is not a question about a policy page, it's a question about the world! What you just said is equivalent to suggesting that my edit makes it harder to determine what color the sky is. Anyone who consults a guideline in order to determine whether a topic is primary is engaging in idiocy. You consult a consensus-building discussion to determine that.

This page has no effect on our ability to determine primary topic, unless we're slaves to the words written here. We're not slaves. When will you drop this ridiculous fantasy that these pages are so powerful? They're just documentations of what happens. You want to make it inaccurate again, revert me! I've practiced 0RR for years; I won't edit war with you! Make your edits, do your word magic. See how it affects decisions in the field; it won't. Neither do my edits.

You've proven that you haven't thought about this, because you refuse to address specific issues in specific examples. You don't even know how to spell "Anne Hathaway". I don't believe that you've read the discussion, because you apparently can't cite a single principle that was brought up, not even to refute it.

Name and discuss a single criterion, or I can't believe that you even know what they might possibly be. As for which criteria should get the most weight, we won't know that until we've observed what consensus does with it in action. We don't decide these things on these stupid talk pages; we're not lawmakers! These things are decided in the real discussions that you refuse to observe. You seem to think that failing to observe consensus-building discussions is how we should document what consensus is. That's so incredibly misguided. Just go make some observations, and then tell us what you saw. Not what you failed to see, but what you saw. Until then, you're wasting our time. You won't even edit the guideline; you just want to complain on the talk page. Do something, already, Born2cycle. Anything. -GTBacchus^(talk) 01:04, 30 September 2011 (UTC)

Asking for any old criterion is very different from establishing clear objective criteria for determining when to go with most likely to be sought vs. most important.

Trying to provide clear objective criteria, rather than ambiguous guidance, does not make us lawmakers. --Born2cycle (talk) 04:00, 30 September 2011 (UTC)

I don't think there is a primary topic for Syracuse, personally, though I haven't looked at the numbers recently. As for Anne Hathaway, it's not that I dispute that others think there's a conflict; all I meant is that the modern actress is at least as important culturally as Shakespeare's wife (at least the actress has actually created cultural works of lasting value, for instance). Powers ^T 19:20, 29 September 2011 (UTC)

Well, that's a purely subjective opinion. How do we measure cultural importance? There are ways. Of the ways I've listed, most favor the historical figure who has been a subject of scholarship and speculation for centuries, enduring through scores of generations of very, very popular actresses who are now forgotten. When you say "lasting value", you do mean on the scale of centuries, right? To me, it's a complete no-brainer, but I'm not the measure of these things. Neither are you. Consensus is.

Regarding Syracuse, if none of the topics is primary with respect to usage, then there's no conflict between "primary usage" and "primary importance". There may be a conflict between "primary importance" and "no primary topic", and that's worth analyzing as well. -GTBacchus^(talk) 20:37, 29 September 2011 (UTC)

Sorry; I misunderstood. (Some people say "primary usage" when they mean "primary topic".) Syracuse, New York appears to be the primary meaning when measured by popularity or usage. As for Anne Hathaway, I'm still trying to figure out what she did of note besides marry Shakespeare. Powers ^T 20:57, 29 September 2011 (UTC)

What did Mercury, the planet, do of note, besides find itself in an orbit in our solar system? What did Mercury, the Roman god do? He didn't even exist. Neither did Eve. Neither did Helen of Troy. What could possibly make them important?

You seem to be arguing that personal accomplishments must be the sole determiner of importance for a person. What the historic Anne Hathaway did was inspire centuries of interest, scholarship, and speculation. She continues to do so. Whether she did that by getting married or by discovering fish is immaterial. The point is that many scholars and writers have considered her of great interest for a very long time. That is getting to the heart of one principle that we can use to determine importance: Enduring coverage in reliable sources over a long period of history.

For a long time, our criterion for "primary usage" has been that the article is more likely to be accessed than all the others combined that share the title. Is that true of Syracuse, New York? It might be. Is it? Never mind, if I want to know anything around here, I have to look it up myself. Thanks for your help. -GTBacchus^(talk) 01:12, 30 September 2011 (UTC)

Syracure, New York should be the primary topic for "Syracuse", since that best serves the readership. Using "cultural importance" as an excuse to send most readers to the historical figure that they don't care about and aren't going to read anyway is putting the editors ahead of the readers and reduces the usability of the encyclopedia. That's why primary topic navigational should reflect usage. -- JHunterJ (talk) 23:30, 29 September 2011 (UTC)

Does Syracuse, New York fit the long-standing criterion for primary usage: i.e., is it accessed more often than the other topics combined? Or are you arguing that we need a looser criterion, and that we should give the primary topic to a simple plurality of "page hits"? That would be a change; good luck getting consensus support for it! -GTBacchus^(talk) 01:04, 30 September 2011 (UTC)

I think it is correct that there is no primary topic for "Syracuse". We don't send readers to the historical figure (whatever that might mean in this case). We do send readers to a disambiguation page because the term is ambiguous and there is no clear-cut predominant sense of the term. older ≠ wiser 23:59, 29 September 2011 (UTC)

This is not the the place to be arguing about whether "Syracuse" has a primary topic. This is the place to be discussing whether the definition of "primary topic" on this page makes clear whether "Syracuse", and countless other titles, has a primary topic. I suggest the current definition provides not much more guidance than would the statement, "toss a coin: heads it's "primary"; tails it's not". Useless. --Born2cycle (talk) 00:22, 30 September 2011 (UTC)

I will prove you wrong. You can take that to the bank.

You, Born2cycle, seem to persist in your ridiculous superstition that these pages are rules. They simply document what would happen anyway, whether or not this page existed. We can delete this page, and the principles will still reign in the field. You refuse to document what happens, because you won't even comment on your observations. You've convinced me that you've made no observations. If you had, you would be able to name at least one principle that's been cited in a "primary importance" argument. You've failed to do so, despite being asked over, and over, and over again. Are you going to make me hold your hand and spell it all out? Seriously? -GTBacchus^(talk) 01:04, 30 September 2011 (UTC)

Every now and then I pop in here and find this discussion going on for the bazillionth time, and so I say for the bazillionth time, that the primary topic for a term should be ascertained by assessing how reliable sources use it. This offers a beautifully elegant symmetry with the article titles policy, which recommends entitling articles according to how reliable sources name topics, and it captures neatly our concerns about recency and encyclopedic "importance". Hesperian 01:08, 30 September 2011 (UTC)

That's beautiful, and useful sounding. Can you say something about how it applies to, say, Syracuse? -GTBacchus^(talk) 01:13, 30 September 2011 (UTC)

The nice thing about the way this approach handles recency is that it is possible for a recent phenomenon to wrest primacy from an older topic, but only if it does so in real world scholarly dialogue. Titanic the movie can wrestle primacy from Titanic the ship, but only if the movie proves of such monumental significance that reliable sources are using the term Titanic much more often to refer to the movie than the ship. This doesn't happen very often, and rightly so...

... but "Syracuse" does seem to be a case where it has. Syracuse, Sicily has 2,700 years of history behind it, but Syracuse, New York is so much more the subject of modern scholarly dialogue, that, if "Syracuse" has a primary topic at all, it would be the latter. Hesperian 01:43, 30 September 2011 (UTC)

As the subject of scholarly dialogue, Anne Hathaway the wife of Shakespeare absolutely smashes the actress. Shakespeare's wife is the primary topic there. Hesperian 01:54, 30 September 2011 (UTC)

Earlier you referred to making the determination based on usage in reliable sources, which includes newspapers, magazines and the like, per WP:RS. For determining primary topic for "Anne Hathaway", you seem to be limiting to scholarly ones. I don't see how doing that exemplifies the beautifully elegant symmetry to which you referred above.

Now, if you compared usage in all reliable sources, not just that which is limited to "scholarly dialog", I think you'd find the actress is not smashed at all, and there is no primary topic. So I agree with you in principal, I just think you did not following the very principle you advocate in this example (but you did with Titanic - for which "scholarly dialog" is also still dominated by the actual ship). --Born2cycle (talk) 04:13, 30 September 2011 (UTC)

Somewhere in the middle. For information on the life, career and achievements of a living actress, magazines like Film Comment and American Cinematographer are reliable sources. Women's Weekly is not; nor is whatever TV magazine comes with your Sunday newspaper. On that basis, I reassert that Shakespeare's wife smashes Hathaway the actress. Hesperian 04:27, 30 September 2011 (UTC)

That's no doubt due to the bias inherent in WP:ANCIENTISM. After all, Shakespeare's wife does have a 426 year head start. Perhaps if we divided number of articles about each by years since birth it would even the score a little. Station1 (talk) 07:10, 30 September 2011 (UTC)

Perhaps the weight of 426 years of scholarship is worth something? Perhaps it's worth something to a relatively broad consensus of Wikipedians? -GTBacchus^(talk) 23:42, 30 September 2011 (UTC)

Field observations

The fact that you can't extrapolate a trend from one or two data points is not proof that a trend cannot be extrapolated. Refusing to consult more data points, and at least make a good-faith effort at documenting a trend, is incredibly counter-productive. Those who are here to document what's actually happening on the Wiki, instead of make wishes about what might happen in the Never-Never Land version of Wikipedia, please help me collect a list of principles that are cited in the discussions we're talking about.

Three examples to start with: Anne Hathaway, Syracuse, Corvette. Is real discussions, that have really occurred, what principles have people cited that they find useful in determining whether a topic is of primary importance. When we start listing these, we at least have a chance of observing some patterns. Meanwhile, those who prefer ignoring actual practice can continue to repeat that there can be no patterns.

The churchmen who refused to look in Galileo's telescope argued that they didn't need to look; they knew Jupiter could have no moons. I'm here to be an astronomer, not a dogmatist. I'll start listing discussion links. -GTBacchus^(talk) 01:13, 30 September 2011 (UTC)

Corvette

Discussions:

Main title:

Corvette - a type of ship - July pageviews: 35,492

Anne Hathaway

Discussions:

Talk:Anne Hathaway#Requested move [added by Station1]
Talk:Anne Hathaway (actress)#Requested move
Talk:Anne Hathaway (actress)#Requested move redux

Main title:

Anne Hathaway - dab page - July pageviews: 23,150

Syracuse

Discussions:

Main title:

Syracuse - dab page - July pageviews: 4,007

Principles to determine primary topic

I'm going to try to make a neutrally worded list of distinct principles that I think real Wikipedians cite in primary-topic discussions. This is informed by the above, of course, including Born2cycle's list. I would encourage others to add more, but please try not to duplicate terms, and please try to keep them neutrally worded, and in the spirit that they have been offered. -GTBacchus^(talk) 19:51, 30 September 2011 (UTC)

The following are criteria that we have observed Wikipedians applying in primary topic discussions. Inclusion in this list is not meant to imply anything about relative importance, or the advisability of including each criterion in the guideline. This list is simply a tool for discussion.

Page views - If an article associated with a term gets more hits than all the others combined, then it has a claim to primacy.

Incoming links - If an article associated with a term has move incoming links than all the others combined, then it has a claim to primacy.

Coverage in sources - If a topic is discussed in reliable sources significantly more (can we make this precise?) than other topics associated with the term, then it has a claim to primacy.

Historical usage - If a topic has been associated with a term for a significantly longer period of time than other topics associated with that term (can we make this precise?), then it has a claim to primacy.

~~Ancestorship~~ Legacy - If a topic is "ancestral" to other topics associated with a term, in the sense that they're named after it, then it has a claim to primacy.

Dictionary definition - If a topic is the dictionary definition of a term, then it has a claim to primacy.

Default due to No Title Conflict - If a topic is the only one claiming a term as its article title, i.e., if all other topics to which that title might apply have articles with other titles (or have no articles), then it has a claim to primacy.

add more...

Of those, I think that the "Dictionary definition" criterion carries the least sway in these discussions. I also think there are perfectly natural ways to make the "historical usage" and "coverage in sources" criteria more precise, but let's take this one step at a time. Other comments? -GTBacchus^(talk) 19:51, 30 September 2011 (UTC)

I added "Ancestorship", which is a terrible name, and not the weightiest consideration, but it's one that does come up in these discussions, and seems to mean something to many editors. -GTBacchus^(talk) 23:45, 30 September 2011 (UTC)

For now, my only comment is... I like where this is going... --Born2cycle (talk) 23:57, 30 September 2011 (UTC)
- That's funny, you've been slowing us from getting to this point for a long time now. -GTBacchus^(talk) 00:22, 1 October 2011 (UTC)
  - I've never liked the "Ancestorship" argument; if an article can't win primary status based on other concepts ("Page views", "Historical usage", etc), it definitely should not be primary. Consider Boston and Winston Churchill. So why should we list "Ancestorship" here, and imply it has equal weight to other factors? --JaGa^talk 16:35, 2 October 2011 (UTC)
    - I don't think anyone's assigned weight as of yet; GT has merely listed factors that have been cited in discussions. There was actually a rash of move requests several months ago advocating moving localities in England to their base names in preference to American or Australian cities that the advocate had "never heard of". Granted, ancestorship alone is rarely accepted as valid by a consensus of participants, but sometimes it can be used as an objective substitute for "historical usage". Powers ^T 17:08, 2 October 2011 (UTC)
      - While I agree that ancestry alone is rarely sufficient to establish a topic as the primary topic, it may in other context be a factor in determining whether or not a primary topic exists for a particular term. older ≠ wiser 17:19, 2 October 2011 (UTC)
        Also, perhaps "derivation" might be a better descriptor than "ancestorship". older ≠ wiser 17:28, 2 October 2011 (UTC)
        Thank you, Powers. JaGa, are you familiar with the concept of a "brainstorm", or even an "observation"? Assigning weights is so much further down the line that I suspect it will never happen. I've explicitly said that the "Dictionary definition" criterion is shit, so your suggestion that I'm giving them equal importance seems to be a creative way to spell, "I haven't read what you've said". I think that "ancestorship" (or "derivation"; thanks Bkonrad) is a stupid criterion, at least in most cases, but I'm listing crap that I've seen people say. That's how we start to document consensus, and I wish fewer people were so interested in derailing the process. -GTBacchus^(talk) 18:18, 2 October 2011 (UTC)
        It would help if you showed a little more good faith in your fellow editors, GT, and stopped accusing everyone of deliberately obstructing your noble efforts here. Powers ^T 19:55, 2 October 2011 (UTC)
        I'm not "noble", Powers, and I've never claimed to be. I don't like being told that I'm promoting shit that I'm not promoting, or being accused of implying things that I'm not implying. I especially don't like it when a cursory reading reveals that I'm not doing the crap I'm being accused of. I list factors, and I explicitly say that some of them are junk, but that I'm trying to observe consensus as I see it. Then someone says that we're wrong to make preliminary lists, and that we're implying that each factor we deign to consider has equal weight. That's unadulterated, unjustified horseshit.
        Have you noticed how much criticism I've had thrown at me, people claiming that I'm trying to intentionally muddy the guidelines, trying to push my ideas of what's educational on others, and other complete bollocks? I'm doing my best to document the muddy and mixed up criteria that I've seen applied in the field, and to make our guideline pages accurate. It's not "noble"; it's because I get tired of people who fetishize rules claiming that we have to slavishly follow inaccurate crap that's written on policy pages. Thank you, Powers, for understanding that.
        Now what precisely is useful about accusing us of giving everything on that list equal weight? How is that more helpful than, let's say, asking how we might reflect weight that these factors are truly given, in the field? I'm very curious. We do not assign weights here, and we never will. We. Document. Consensus. In order to document it, we have to start somewhere, and that might mean writing down principles that we'll end up dropping because they're not well supported. Got a problem with that? -GTBacchus^(talk) 20:51, 2 October 2011 (UTC)
        I'll tell you what I have a problem with. You start your "brainstorming", you come here to completely rewrite the primary topic guideline (although you're no disambiguator that I've seen), and as soon as you get some comments that don't agree with every single thing you throw out, you start cursing and accusing people of not parsing every word of your TLDR essays. I was wondering why more actual disambiguators have not been participating in this discussion and now I think I know why. I also see you've prepared for disagreement by declaring anyone whose participation doesn't suit you is "so interested in derailing the process". Of course it couldn't be that they have legitimate concerns or gasp may actually understand the topic better than you. But don't you worry; I'll not participate in this any longer. Does that suit your style of consensus building? --JaGa^talk 16:32, 3 October 2011 (UTC)
        JaGa, no, it doesn't suit my style if you leave. You're right that I was wrong to yell at you like that, and I've posted an apology to your talk page. I hope that you'll continue to participate here, and I'll repeat here that I've made a commitment to check my attitude, which has been getting out of hand. Thanks for calling me out.
        I'm not here to rewrite any guideline. I've made one edit to it, and I'm unlikely to make more. Furthermore, anyone who doesn't like my edit is welcome to undo it. I practice 0-RR; there will be no edit-war.
        I'm trying to help facilitate a conversation about how Wikipedians determine primary topic. Whatever conclusions we arrive at regarding consensus, if they're added to the guideline in any form, will be added by someone besides myself. I mean, if I'm going to claim that something I wrote down represents consensus, but then I've got to add it to the guideline myself, then I'm not really fooling anyone, am I? I made one edit, with the summary that I think it's accurate, and now I'm just discussing. (...and shouting a bit. Toning that down...)
        The thing JaGa, is that you haven't disagreed with anything I've thrown out. I think the legacy criterion is not particularly well-supported in discussions, and we would be silly to claim that it carries equal weight. I have no, nor ever had any intention of making such a claim, and I'm ready to discuss at some length observations of how it's actually used in the field. How convincing do Wikipedians tend to find it? My suspicion is that it's given less weight than most of the other principles on the table, in most cases. If that turns out to be the case, then a sufficiently accurate and detailed guideline will reflect that fact.
        I'm making no definitive claim about how much weight Wikipedians tend to give these guidelines beyond, "I've got some guesses, but let's consult discussions that have really occurred to be sure." Therefore, I'm not sure what it is I'm saying that betrays a lack of understanding, but I hope you'll let me know. I promise not to bite. I'd rather understand than fight.
        Oh, you're right that I'm no "disambiguator", if that's a... what is that? I mostly work in requested moves, where we see a fair share of primary topic discussions. That's the source of my experience on the subject, and I'm willing to read any other relevant discussions that anyone can share. -GTBacchus^(talk) 04:25, 4 October 2011 (UTC)
        
        Wow. That's rather harsh. This guideline is not OWNed by "disambiguators". In fact, I think a large part of the problems with the current guideline may be the result of it having been edited with myopic insularity and without sufficient input from those who are not steeped in disambiguation minutiae. The disambiguation guideline affects more than just disambiguation pages. It has a direct role in article title and requested move discussions. older ≠ wiser 16:43, 3 October 2011 (UTC)
        Harsh? Moi? Well, when a user responds to an innocuous comment with a profanity-laden "OMIGOD DONT U KNOW WHAT BRAINSTORMING IS CANT U READ WHY DO U HATE CONSENSUS", that tends to elicit a harsh response. I thought we were working together, but GTBacchus let me know my views are neither welcome nor particularly valuable, and frankly I have better things to do than fight for approval from an attitude like that. PRIMARYTOPIC will survive without my input. So, yeah, harsh. --JaGa^talk 21:58, 3 October 2011 (UTC)
        That response is so childish, I'm not even sure why I'm bothering to reply. First, you are dramatically overreacting. I don't see that GTB ever made any indication that your views are neither welcome nor particularly valuable. GTB's only slightly strident remarks here were in response to LtPowers. If you want to be treated with respect, I suggest that you not over-react so easily and try to give others the benefit of a not so hostile reading. older ≠ wiser 22:36, 3 October 2011 (UTC)
        JaGa is right. I was out of line. Please expect a better reply here soon; I have to run right now. -GTBacchus^(talk) 01:04, 4 October 2011 (UTC)

I tried to articulate some of these points further up, but unfortunately my edits were obscured by ongoing discussions. Historical/ ancestry: I had paired "history" with "legacy" (your "ancestorship") as I'd see these as pretty much the same thing with application to a wide range of topics ranging from people (e.g. Hathaway) to cities (e.g. York). However, the usual counter argumment to this seems to be to highlight cultural/ social/ economic/ political impact (or words to that effect). This partially explains why Boston or Churchill or Lincoln occupy the primary space rather than historical antecedents. I think this has to be covered too. Dictionary definitions: I would not dismiss this - a "ford" was a river crossing 500 years ago and it will be a river crossing in 500 years time. Currently it redirects to a company that makes motor vehicles. There should be a way to explain this better, but it's still a valid argument even if it's not the strongest. Coverage in sources: I take your point about documenting consensus, but I feel that this should carry some weight since verifiability is so important to this project. However, it should carry a caveat about geographical bias. This cuts both ways, but if we rely on Google books we have to be clear that it does not reflect a global perspective. Wiki-Ed (talk) 13:13, 3 October 2011 (UTC)

A comment on incoming links -- if this criterion is used, it must be applied very cautiously, especially if there has been a history of page moves and/or redirects. A link to the disambiguation page may have been created at a time when that title did not contain a disambiguation page. To take a purely fictional example, say that Roger had been created in 2004 as an article about the use in radio jargon, and Roger (name) was created in 2005. Then, in 2010, the first article was moved to Roger (radio jargon) and a disambiguation page placed at the main title. Someone who comes along in 2011 might look at the incoming links to Roger and see they are nearly all about radio jargon, not about the name, not realizing that is because that was the way the articles about these two topics had been titled for the previous six years. In other words, the prevalence of incoming links reflects past Wikipedia naming decisions more than it does either frequency or importance of usage. --R'n'B (call me Russ) 20:36, 3 October 2011 (UTC)
Furthermore, links to disambiguation pages are generally incorrect, and will often be fixed. As a result, the state of incoming links to a disambiguation page today may not reflect what it was in the past, because editors spotted some of the past links and corrected them. In particular cases, where some links are very easy to fix and others are difficult, all the easy cases may be gone and this criterion would incorrectly indicate that the "easy" case was not often accessed. --R'n'B (call me Russ) 20:36, 3 October 2011 (UTC)
- I agree with R'n'B on this point. A better metric is the intended destination of new links added to disambig pages that regularly have new incoming links. For example, I regularly fix newly added links to EP, Painter, and LLC. Around 99% of new links that are made to these pages are intended to go to Extended play, Painting, and Limited Liability Company, respectively. I also regularly fix links coing in to Boxer, Bass, for which probably 90% are intended to go to Boxing, Bass (guitar), followed by Boxer (dog) and Bass (fish) or Bass (voice type). By contrast, new links made to Amazon are closer to an even split between Amazon.com and the Amazon River or Amazon Basin. If we were going solely by what people seem to expect when they make links, we would have clear primary topic winners for the first three mentioned, a likely target for the next two, and a clear case for disambiguation of the last one. bd2412 T 16:20, 21 October 2011 (UTC)

Can a topic called "XY" be primary topic for "X"?

This issue is not quite the same as determining primary topic when topics share an exact name. Anne Hathaway is Anne Hathaway is Anne Hathaway. On the other hand, "Chevy Corvette" isn't precisely "Corvette", although a lot of people call it that. I think there has long been agreement that it's possible for topic "XY" to be primary for "X", if most people just call it "X". Comments? -GTBacchus^(talk) 19:51, 30 September 2011 (UTC)

I suggest the wording of the question in the heading would be more accurately stated as, Can the topic of the article currently at "XY" be the primary topic for "X"?

I think the answer is certainly, and perhaps the most obvious examples are famous people with articles at their full names who never-the-less are the primary topic for the surname too... Reagan, Nixon, Obama... Of course this practice is not limited to US presidents or even to people; we see it in place names too, like Monterey, Sacramento, Baton Rouge (these are all redirects to articles whose titles include ", State")... It can be in company names as well, such as Ford and GE (the latter is not strictly an example of literal XY/X, but it is a short name redirecting to a longer name, which relies on the same underlying principle). And these are just examples off the top of my head - WP is replete with more. And WP:PRIMARYTOPIC has addressed this point explicitly for a long time; it states: " The fact that an article has a different title is not a factor in determining whether a topic is primary.".

Does this prevent people from arguing otherwise when it suits the title they prefer? Often not.

So, here we have a principle that is about as consistently supported in practice as is any in WP, is explicitly stated in policy, and yet people will still ignore it for no reason other than following it results in an article or dab page placement that they don't like. --Born2cycle (talk) 23:51, 30 September 2011 (UTC)

You know, we've all read it. I don't keep linking the same thing to you, because if I'm going to be insulting, I'll just do it in so many words. Conversations are not made of links; is your goal to piss people off? I'm reading the content of what you read now, but Christ, man. Ugh. Links to project space pages that we've all read get so OLD! Ugh. -GTBacchus^(talk) 00:21, 1 October 2011 (UTC)

Yeah, pretty much in agreement here, Born2cycle (except regarding how to best trade barbs). There's something, though...

Case I: We've got a significant and popular topic called "X" and another significant and popular topic called "XY", which is often known as simply "X". Case II: We've got two significant and popular topics, both called simply "X".

In case II, we have to either decide that one is primary, or else let the dab page sit at the main title. (I don't think that defaulting to one topic at the main title as status quo is a stable solution.) In case I, however, the fact that one of the topics can be called "XY" buys us an escape from having to make the call. We can call the articles "X" and "XY", and this might be more satisfying.

I dunno, whaddya think? -GTBacchus^(talk) 00:32, 1 October 2011 (UTC)

In some cases, that solution flies; in others it doesn't. It depends in large part on the specifics of each case. Powers ^T 01:19, 1 October 2011 (UTC)

Yes. I agree. Trying to extract some kind of pattern from the specifics of each case is the only reason we have policy and guideline pages. In this case, I'm asking what pattern there might be. Any ideas? -GTBacchus^(talk) 20:53, 2 October 2011 (UTC)

Discussion involving using patronymics as disambiguators

There is a move request discussion with potentially wider implications at Talk:Vladimir Mikhaylovich Smirnov over whether to use Russian patronymics ("middle names") as disambiguators or whether to use parentheticals. Please weigh in if you care. — AjaxSmack 03:25, 10 October 2011 (UTC)

WP:QUALIFIER isn't explicit (though the William Henry (actor) example illustrates this): perhaps it needs to say something like "Where an additional given name, patronymic or initial is not commonly used, it is preferable to distinguish by a bracketed qualifier rather than by using this element" - could even give this ski-er and fencer as examples if the discussion goes in the direction I hope it will! PamD 08:07, 10 October 2011 (UTC)

What is an "incomplete disambiguation"?

I do a lot of disambig cleanup, and I am now seeing more and more dab pages tagged as "incomplete disambiguations". But what exactly is this? I've read the description 10 times, and still don't understand it, nor do I understand just what is needed to cleanup an "incomplete" disambig. And if I, a disambig regular, can't figure it out, heaven help the rest of our users. Is there anyone who can explain this in a more clear way? --El on ka 19:05, 20 October 2011 (UTC)

As an example, if there are 3 equally popular songs — "Xyz" (Beatles song), "Xyz" (Gershwin song) and "Xyz" (folk song) — then Xyz (song) is an incomplete disambiguation from Xyz. In other words Xyz (song) disambiguates from Xyz but it's not enough, so Xyz (song) should redirect to the dab page for Xyz (which would be either Xyz or Xyz (disambiguation), depending on whether there's a primary topic).

Personally, I think there are a few cases where it's better to keep Xyz (song) as a dab page itself, for instance when Xyz is long and messy and the songs are highly sought in comparison to other uses (a real example might be Avatar (film)). The downside is the need to keep two dab pages updated instead of one. Station1 (talk) 19:58, 20 October 2011 (UTC)

It might make more sense for the incompletely disambiguated dab not be a dab page but to redirect to the appropriate section of the real dab. So to use your example, Avatar (film) would redirect not to Avatar (disambiguation) as it currently does, but to Avatar (disambiguation)#Film. That avoids the duplication problem that requires multiple dab pages to be updated. In fact, I just change it do that [4]. --Born2cycle (talk) 20:08, 20 October 2011 (UTC)

Yes, that's often a good idea, but in this case it bypasses the one film that almost everyone wants, which is at the top of the page, not in the film section. Station1 (talk) 20:22, 20 October 2011 (UTC)

Thanks for trying to explain, but I'm still lost. For example, in Category:Disambiguation pages in need of cleanup from July 2011, there are several pages listed for cleanup such as Cleveland (NFL), Oakland (AFL), Kansas City (AFL), etc. Just what exactly is needed to get these "cleaned"? Redirecting them to the city names does not make sense to me. --El on ka 20:14, 20 October 2011 (UTC)

I think these are exactly the type of page I was talking about that should remain a separate dab page. I agree it makes no sense to redirect these to the dab pages for Cleveland, Oakland or Kansas City. In these cases I would simply remove the tag. Station1 (talk) 20:22, 20 October 2011 (UTC)

My understanding is that the content of Cleveland (NFL) should be moved to a new NFL section of Cleveland (disambiguation), and Cleveland (NFL) should be turned into a redirect to Cleveland (disambiguation)#NFL.

The idea is to serve someone looking for a Cleveland NFL team but can't remember the name of it. The Cleveland what? As it is currently he enters "Cleveland" and goes to the article about the city, but sees a hatnote for the dab page. However, clicking on that does him no good because the NFL uses of "Cleveland" are not currently listed there. --Born2cycle (talk) 20:32, 20 October 2011 (UTC)

If they are ambiguous (which I am skeptical of), there should be an "NFL" or "Sports" section for Cleveland (disambiguation) et al. For these, it might be better moving them to List of NFL teams based in Cleveland et al. -- JHunterJ (talk) 20:49, 20 October 2011 (UTC)

(ec)That's a good point. The other way to take care of that is to add Cleveland (NFL) to the Cleveland dab page. Double disambiguation does make sense in some cases. Usually, adding numerous sports teams, which are really partial title matches, to a city's dab page, clutters it in my opinion. Station1 (talk) 20:52, 20 October 2011 (UTC)

If they're PTMs, the list articles are definitely the way to go (and the lists linked from the dabs' See also sections). If they're ambiguous, they should be included in the disambiguation page, not as clutter but as ambiguous entries needing disambiguation. -- JHunterJ (talk) 20:55, 20 October 2011 (UTC)

I do not agree with linking Cleveland (AFL) to the Cleveland (disambiguation) page. In fact, those are the exact kinds of links that I would normally remove from a page such as Cleveland (disambiguation), because it would not be likely that anyone would have linked Cleveland when they really meant to link to a sports team from Cleveland. --El on ka 02:34, 21 October 2011 (UTC)

That's a good argument for turning them into list articles -- they are not ambiguous in that view (which I agree with). But the alternative is that they are ambiguous and get listed on the dab page -- JHunterJ (talk) 02:45, 21 October 2011 (UTC)

I agree generally that if we're talking about a collection of links to articles on football teams that have been located in Cleveland, we need a list and not a disambig. That situation is running dangerously close to WP:DABCONCEPT territory. bd2412 T 02:52, 21 October 2011 (UTC)

I agree that most if not all of these teams are not ambiguous with the term "Cleveland" (or at least the ambiguity is extremely slight as with other partial title matches). Because they are not ambiguous, I don't think the pages Cleveland (AFL) and Cleveland (NFL) are named appropriately. I'd suggest that both should redirect to List of Cleveland sports teams. older ≠ wiser 13:07, 21 October 2011 (UTC)

That's one of the better ideas I've heard so far. I'm also thinking that we might need a "disambig cleanup instructions" page of some sort. Maybe Wikipedia:Disambiguation/Common problems? Then the "incomplete disambiguation" templates would have something better to link to, with specific suggestions on how these things (and other common cleanup problems) can be fixed. --El on ka 14:27, 21 October 2011 (UTC)

Okay, now I'm confused. I'm puzzled by JHunterJ's skepticism about whether "they are ambiguous", and Bkonrad's comment: "most if not all of these teams are not ambiguous with the term "Cleveland" (or at least the ambiguity is extremely slight...)". Since it is very common to refer to a team name by city name only, as in, "They're playing Cleveland tomorrow", I suggest "Cleveland" is unquestionably ambiguous for each of the Cleveland teams. More generally, the gray area between ambiguity and partial title matching is so broad I'm not sure there is any point in trying to draw a distinction between the two. I mean, not all ambiguous uses are partial title matches, but many if not most partial title matches are ambiguous references to particular topics. Am I missing something?

Also, I'm wondering about Elonka's suggestion that something not be linked to the Cleveland dab page "because it would not be likely that anyone would have linked Cleveland when they really meant to link to a sports team from Cleveland.". This statement seems to assume that the reason we link ambiguous terms to dab pages is only to reasonably handle linking to these ambiguous terms. But what about searches? Aren't these links to dab pages supposed to also aid readers find the article they are really seeking? Again, if someone is looking for a Cleveland team but can't remember its name, or might actually be trying to find out what the name is, he is likely to just search with "Cleveland". The idea that the Cleveland teams should not be listed on the Cleveland dab page because Cleveland is merely a PTM makes no sense to me. --Born2cycle (talk) 15:40, 21 October 2011 (UTC)

It might be common to say, "They are playing Cleveland tomorrow" in a verbal conversation, yes. But in a Wikipedia article, the sentence would more likely be something like, "The Detroit Ostriches will be playing the Cleveland Sugar Gliders" (or whatever). It would be highly unlikely for an editor to link "Detroit will be playing Cleveland", without some other context. For another example, like with "M", a stock symbol. It would be unlikely for someone to link M when they really meant McDonald's, or N when they really meant Nitrogen. The guidelines for disambig pages, are that the only entries there should be those that might reasonably be linked with that exact term. --El on ka 17:31, 21 October 2011 (UTC)

I can certainly imagine that in an article about a particular game, perhaps an article about a particular Super Bowl game, either team could be referred to by city name only, as in, "Cleveland dominated the second half by ..." In fact, I just found this sentence in the lead of Super_Bowl_XXXII: "Despite suffering a migraine headache that caused him to miss most of the second quarter, Denver running back Terrell Davis (a San Diego native) was named Super Bowl MVP." Here, of course "Denver" refers to the Denver Broncos. Denver is not linked to the page about the team here, but it could be. And yet, if I click on Denver and then the hatnote link to Denver (disambiguation), I find nothing there about the team, though there is a link (and properly so in my view) to the University of Denver athletic program, Denver Pioneers.

Regardless, this point again seems to be based on the assumption that the reason we link ambiguous terms to dab pages is only to reasonably handle linking to these ambiguous terms. I ask again, what about searching? What about readers searching for information about a team by entering only "Cleveland" in the Search box? Shouldn't we set up our redirects and dab pages to help them get to their desired article as efficiently as reasonably possible? --Born2cycle (talk) 18:08, 21 October 2011 (UTC)

Sadly, that Denver (disambiguation) page is not a good example, as it's gone a bit off the rails. I have tagged it for cleanup. See also Wikipedia:Manual of Style (disambiguation pages)#Examples of individual entries that should not be created. --El on ka 20:19, 21 October 2011 (UTC)

Thanks. Note this part of it:

The [guidance against including partial title matches on dab pages] does not apply if the subject is commonly referred to simply by Title. For instance, Oxford (disambiguation) should link University of Oxford and Catalina might include Santa Catalina Island, California. If there is disagreement about whether this exception applies, it is often best to assume that it does.

Professional sport teams are commonly referred to simply by the name of the city they are from. Disagree? "It is often best to assume that it does." :-)

Note also that it does not say: "The [guidance against including partial title matches on dab pages] does not apply if the subject is commonly linked to simply by Title." That is, the criteria to use is whether it is commonly referred to by Title alone, not the much more narrow criteria (which you seem to be using) of whether editors are likely to link to that topic using [[Title]]. --Born2cycle (talk) 20:33, 21 October 2011 (UTC)

I still think it boils down to : Include entries that are ambiguous in the disambiguation page. If the page is long, use sections to aid navigation of it. If the topic are not possibly titled with the ambiguous term, do not put them on the disambiguation page but rather in a list article. Those are the ways to help readers find the articles they are seeking. -- JHunterJ (talk) 20:36, 21 October 2011 (UTC)

I agree, and I think that almost any city with enough sports franchises to justify disambiguating them would be better served by a "List of sports teams in Foo City" article, which could sort all of the sports out by the type of sport and founding and ending dates for each team, and could include teams not necessarily notable enough for an article, such as college or high school teams that might be referred to by the name of the city when playing against teams from another city. bd2412 T 20:40, 21 October 2011 (UTC)

This is helpful on how to deal with the sports teams, thanks. There are still many other incomplete dabs though to go through. For example, how should this one be handled? Route 61 (public transport). That one, and many others like it, are in Category:Incomplete disambiguation from May 2011. I don't see any obvious redirect target or related list page to link it to, so, should we leave it alone? Or any other ideas? --El on ka 22:21, 22 October 2011 (UTC)

I'd merge them back to the disambiguation at Route 61. The title "Route 61 (public transport)" is an artificial contrivance. If the listed routes are known as "Route 61" then they should be listed on the correct disambiguation page. And personally, I'd then delete Route 61 (public transport), but there would no real harm in leaving it as a redirect to the disambiguation page. It there were a huge number of such transport routes, it might make sense to break out a List of public transport routes numbered 61, similar to what happens with some roadways, but I think that would be rather exceptional. older ≠ wiser 22:51, 22 October 2011 (UTC)

Aha, I replied a little too quickly. I didn't realize that Route 61 is a redirect to Highway 61. And there already exists List of highways numbered 61. Oy. older ≠ wiser 23:01, 22 October 2011 (UTC)

Any other suggestions? If not, I'm inclined to just remove the tags from these articles. --El on ka 05:04, 25 October 2011 (UTC)

In the case of Route 61 (public transport), I would move it to Route 61, making this a different and separate dab from Highway 61. They may be very similar but in terms of the article titles they are different. I would then just add a link to Highway 61 and List of highways numbered 61 in a see also section. France3470 (talk) 16:58, 29 October 2011 (UTC)

Demonym dabs

I've recently been discussing some new dabs with their creator. They are for demonyms such as Sri Lankan and Aragonese, replacing redirects to Sri Lanka, etc. I can see the logic of the change, because Indian, Chinese etc have dabs. However, I'm concerned that terms like Sri Lankan may have a primary topic or at least be broad concepts.

I noticed the new pages because a report[5] flagged them as dabs with many incoming links. Although it shouldn't be our main consideration, one advantage of the redirects is to route the demonyms' existing and future incoming links to an appropriate article.

Should we be consistent by treating Sri Lankan like Indian, or do practical differences suggest a different treatment? I'll also notify WikiProject Geography, asking them to place any comments here. Certes (talk) 13:07, 31 October 2011 (UTC)

I don't think consistency needs to be a goal here. If one title has a demonym meaning as primary, it's primary; another title might have a demonym meaning that's not primary, and so it's not primary (either another topic is primary or no topic is primary). So Chinese might be a dab while Sri Lankan might not be, just like Elvis has a primary topic but Madonna doesn't, or Alaska has a primary topic but Georgia doesn't, or Reagan does but Lincoln doesn't. -- JHunterJ (talk) 13:34, 31 October 2011 (UTC)

i understand that sometimes a dab is unneeded and undesirable, and i thought about it

what i don't see is a fairly simple criterion, such that two or three people will likely reach the same conclusion — Preceding unsigned comment added by 70.189.170.229 (talk) 19:03, 31 October 2011 (UTC)

I think you've got it. Consensus is against a straightforward number-crunching criterion for primaryness. If two or three editors do not reach the same conclusion, WP:CONSENSUS would come into play. -- JHunterJ (talk) 19:19, 31 October 2011 (UTC)

Well, first try to determine facts. Having 241 incoming links is great - that means 241 various contexts that you can examine and determine whether the term is largely referred to in an ambiguous manner or not. Obviously a semi-random sample could do, too. Note also a related example at Talk:Moroccan. --Joy [shallot] (talk) 19:37, 31 October 2011 (UTC)

At the risk of tl;dr I'll sample links 1, 21, 41, ... 241 from today's list[6] and suggest some improvements.

Adjectivals and demonyms for countries and nations (in table, row Sri Lanka, column Adjectivals) – a difficult choice but probably unlink like Brazilian
Dimitri Mascarenhas has two links:

is an English cricketer of Sri Lankan Tamil heritage – Sri Lankan Tamil people conveniently also removes the link to dab Tamil

a replacement for Sri Lankan fast bowler Lasith Malinga – Sri Lankan cricket team or perhaps Cricket in Sri Lanka

Free Media Movement Journalists from Sri Lanka – Media of Sri Lanka
Henry Steel Olcott His view of Buddha influenced Sri Lankan leaders, such as Anagarika Dharmapala – Buddhism in Sri Lanka
Jeyanthan Brigade battles fought against the Sri Lankan Forces – Sri Lanka Armed Forces
Lalith Kotelawala is a Sri Lankan businessman – Sri Lanka
Mervyn Silva is a controversial Sri Lankan politician} – Sri Lanka or perhaps Politics of Sri Lanka
Palitha Perera nationality in infobox – Sri Lanka
Ran Banda Seneviratne was a Sri Lankan lawyer, author, lyricist, television and radio presenter – Sri Lanka
Sarath Fonseka nationality in infobox – Sri Lanka
Stigmata (band) is a Sri Lankan heavy metal band – Sri Lanka or perhaps Music of Sri Lanka
Thusitha Laknath is an actor in Sri Lankan drama, teledrama, and film – Sri Lanka or perhaps Cinema of Sri Lanka
Zero degree the custodial treatment of Sri Lanka refugees (piped link to Sri Lankan) – Sri Lanka or possibly Human rights in Sri Lanka if the treatment occurred there

That should help with the cleanup, though all it shows is that one editor wrote a generic link to Sri Lankan when a more specialised article existed (or was created later). I am not suggesting that every one of those targets deserves a place on a dab page, or even that there should be a dab page. Certes (talk) 23:53, 31 October 2011 (UTC)

I agree. If there was a single language called "Sri Lankan" or a single eponymous ethnicity, the case for disambiguation would be clearer. This is a borderline case that you should probably {{rfc}}. --Joy [shallot] (talk) 11:37, 1 November 2011 (UTC)

I'll add the RFC template here, with a new section header and summary for it to pick up. Please put further discussion after that summary rather than here. I'll change the header, as I don't think the terms are necessarily all demonyms. Certes (talk) 14:18, 1 November 2011 (UTC)

Country and region adjectives

Which country and region adjectives should redirect to a primary topic, and which should have disambiguation pages like Indian? Please see the discussion of Sri Lankan at #Demonym dabs above. Certes (talk) 14:18, 1 November 2011 (UTC)

I'm not sure why this is an RFC here. I thought you were going to collect comments on specific adjective pages' Talk pages. But here: Titles (including that that are ambiguous with demonyms) that have a primary topic should redirect to that primary topic, and titles (including those that are ambiguous with demonyms) that don't should have disambiguation pages like Indian. To figure out whether "Sri Lankan" has a primary topic, you should find consensus at Talk:Sri Lankan. -- JHunterJ (talk) 14:59, 1 November 2011 (UTC)

Ah, perhaps I misread Joy's suggestion. I was looking for some general guidelines, so that we can boldly do the right thing where that's obvious, then discuss the others on the individual talk pages. Unless anyone has other ideas, what you just wrote seems like a good starting point. We can start with pages having recent major changes and consider whether to widen the scope later. Certes (talk) 15:41, 1 November 2011 (UTC)

That's what I mean. The general guidelines are the existing ones that apply to all titles. The general guidelines don't need separate tweaking for demonyms. -- JHunterJ (talk) 16:37, 1 November 2011 (UTC)

I wanted you to RFC it over *there* :) on a case-by-case basis. Talk:Sri Lankan, and you can also link the discussion from the relevant wikiproject talk pages. --Joy [shallot] (talk) 18:42, 1 November 2011 (UTC)

This issue affects several countries, though it is probably best to deal with one at a time. Rather than raise an RfC for each, let's use the existing RfC for this page to draw editors' attention to the following page-specific discussions:

Sri Lankan

I'll expand the above list in situ as appropriate. Certes (talk) 22:38, 1 November 2011 (UTC)

Please use the individual pages' talk pages for the comments. The guidelines don't need to be discussed. I've removed the RFC tag from this section. -- JHunterJ (talk) 01:46, 2 November 2011 (UTC)

initialism disambiguations

(a) Do initialism disambiguation pages (i.e. GLP) require the articles listed have the initialism in their article? (b) Does the entry have to be cited as an initialism, or does the obviousness of the initialism supercede? — Fourthords | =Λ= | 19:21, 1 November 2011 (UTC)

(a) Yes, in the article, subject to WP:N, WP:V, and WP:RS as usual. (b) I'm not sure I understand the question; but looking at GLP, the entries there (just the wikilink to the article using the expansion of the acronym) are fine, for the entries that are identified as GLP or G.L.P. or G. L. P. in the linked articles. -- JHunterJ (talk) 19:26, 1 November 2011 (UTC)

Thanks for answer number one. As to number two, let me explain. My example DAB page of GLP lists the German Longhaired Pointer as an entry. Now the Pointer article has the initialism ("GLP") listed in the article after the bolded article title, but there is no citation that "GLP" is a verifiable or allowable initialism for this article. The GLP initialism is uncited. Now, does it need to be cited even though the title of the article makes the initialism obvious? If it's not cited, should the entry be removed from the DAB page since the initialism isn't verifiable? — Fourthords | =Λ= | 19:40, 1 November 2011 (UTC)

As long as there is consensus for it to remain on German Longhaired Pointer, that's sufficient for inclusion on the dab page. If an editor objects to the inclusion on the dab page, they should remove it (boldly or with consensus, based on WP principles) from the topic article as well (and restore it if their bold edit is reverted). IMO. -- JHunterJ (talk) 19:43, 1 November 2011 (UTC)

Purpose

The page Purpose was recently changed from a redirect into a (so-called) disambiguation page, which contains nothing but dictionary definitions and links to synonyms. Normally I would be bold and revert, but this change was the result of a "consensus" developed after a discussion on the talk page. Unfortunately, none of the discussants bothered to advise this project that they were proposing to create a disambiguation page that deviates so widely from the accepted guidelines. Others may wish to participate in re-opening the discussion there. --R'n'B (call me Russ) 20:11, 1 November 2011 (UTC)

Manifold (automotive)

(More incomplete dab stuff) Manifold (automotive), seems to be generating lots of mixed opinions. It has been tagged for a speedy, this was contested and the tag removed. Today I prodded, but after rereading WP:INCDAB decided it was best to redirect it to the relevant section of the dab page. This was then reversed. I thought it might best for us to discuss it here. Thoughts? France3470 (talk) 17:33, 29 October 2011 (UTC)

Or at the discussion I started at Talk:Manifold (automotive)#Incomplete split, unless there's a bigger issue at stake. Dicklyon (talk) 18:09, 29 October 2011 (UTC)

The page originally covered both inlet and exhaust manifolds but it was convoluted and confusing, so we decided to split it to inlet manifold and exhaust manifold. Since we couldn't leave a page with a single redirect, we left a disambiguation page that points to both of the new pages. There are only a very few articles that link to this page (GM Quad-4 engine, Singer Roadster, Mitsubishi 3B2 engine, Triumph Super 9) and a lot of talk pages. Most of the former links to this page have been change to the more specific inlet/exhaust pages. To aid disambiguation, I have also added links from this page to Manifold (disambiguation) and manifold (in the mathematical sense). Manifold (disambiguation) also has links to inlet manifold and exhaust manifold (I have just done minor tweaks to this page). It would be tempting to change all Manifold (automotive) links to Manifold (disambiguation) except the topic of these links is clearly automotive - to change it to the less general page would not be helpful to the reader. In short, it's not the most fascinating page on Wikipedia but it does its job and does not cause any confusion. Stepho talk 19:01, 29 October 2011 (UTC)

The inbound links from article space to manifold (automotive) were all appropriately redirected to inlet or exhaust manifold. Talk pages and user pages were left as before, for historical reasons. The handful of articles that weren't redirected were those were it wasn't clear which manifold was meant, or one (on a pre-war car) that was referrring to a cooling water manifold, neither inlet nor exhaust.

It makes it all worthwhile to then be told this was a half-assed job 8-( Andy Dingley (talk) 20:10, 29 October 2011 (UTC)

(ec) It should instead redirect to (a section of) Manifold (disambiguation) -- there's no reason for the dab page to link to both the incomplete disambiguation and the entries of the incomplete disambiguation. Or manifold (automotive) could be made into a (short or stub, if needed) general topic article that links to the others for full details -- and if it's a valid link target, then it's an article. But there's no real benefit to having it as an incomplete disambiguation. -- JHunterJ (talk) 20:17, 29 October 2011 (UTC)

I noticed Manifold (automotive) near the top of the Disambiguation pages with links report [7] on 24 September. I diverted most incoming links – a lot went away with one edit[8] to {{CarEngine nav}}. It looks as if I missed a few, but I'm happy to tidy them up once we've reached a consensus and it's safe to destroy the evidence.

I agree with User:France3470 than WP:INCDAB applies and we should just redirect to Manifold (disambiguation). One we do that, or decide not to, then the job is done. Manifold (automotive) can sit there quietly doing its job, unless editors frequently create new links to it. If and when that happens, we should probably give it another prod (or an RfD) to avoid confusion. Certes (talk) 12:50, 31 October 2011 (UTC)

If there are no additional comments/objections I'm going to reinstate the redirect. I don't know whether there is any benefit to creating a concept article for this subject but if another editor with more knowledge of the topic took it upon themselves to write one I wouldn't have any issues with this (or does Manifold (engineering) already cover this?). However, I think for now a redirect is the best option. The dab links will still be targeted to manifold (automotive), so those who are knowledgeable on the subject could have another look to see if these could be disambiguated. I had a look myself but it was completely out of my depth; I think Andy is quite right that these might be problem cases. France3470 (talk) 17:42, 3 November 2011 (UTC)

I have looked at the remaining links again. Triumph Super 9 is a water manifold; the rest are unclear. Short of spraying everything with {{dn}} or {{clarify}} to attract a subject expert, I think we've done all we can with them. Certes (talk) 20:39, 3 November 2011 (UTC)

Okay. Page is now redirected to Manifold (disambiguation)#Science and engineering. France3470 (talk) 21:43, 3 November 2011 (UTC)

Disambiguation categorization

I'd like to get some feedback regarding an inconsistency in approaches and direction around how disambiguation pages are categorized.

The instructions in Template:Disambiguation/doc with regarding to category parameters are straightforward — "Only add a parameter if the disambiguation page has several items of that type.". Further direction is available at WP:MOSDAB — "Parameters can be added to place the page additionally into other more specific disambiguation categories. For example, if a page includes multiple places and multiple people with the same surname (and possibly other items), use {{disambig|geo|surname}}." This language ("for example") seems to be suggestive rather than prescriptive, but acts as a useful guideline.

On the other hand, the descriptive text on some of the category pages — Category:Airport disambiguation and Category:Broadcast call sign disambiguation pages, to name two — and on some of the documentation pages for alternate disambiguation templates describes the categories as being for disambiguation pages with one or more entries of those types.

If the purpose of the separate disambiguation categories is to categorize pages of a type, I'm unsure why a page would need multiple entries to be considered a member of a given category, and so I think the instructions on the category pages are the right approach. However, it's possible that I'm missing history in how this structure was originally developed, so I'd really like to see how others feel. Whatever the consensus, I think it's worth making sure that the template documentation, style pages, and category pages all provide consistent advice for our editors.

Full disclosure — I originally created Category:Broadcast call sign disambiguation pages almost three years ago, and have been among those most involved in populating it. So, I have some skin in the game, so to speak. Mlaffs (talk) 21:04, 3 November 2011 (UTC)

"Partial primary disambiguations"

Sorry for the odd heading, but I didn't quite know how to phrase it. Anyway, to explain, I recently had an interesting discussion on my talk page (see here) on the topic of whether an article can be the primary topic of a parenthetical disambiguation. The article that sparked it was Ray Wilson (English footballer), which I recently moved from Ray Wilson (footballer) because of the existence of a few other footballers called Ray Wilson. From hanging around RM for a few months, it was my understanding that an article cannot (or rather should not) be at simply "Person X (footballer)" when, say, "Person X (Australian footballer)" exists, i.e. the article is either the primary topic or needs a completely unique disambiguation. But I couldn't find this explicitly mentioned anywhere on this guideline, so I'd appreciate some input and if we get a consensus, it would be great if we could make a clarification to the guideline. Jenks24 (talk) 12:19, 3 November 2011 (UTC)

I tend to agree with you on the semantics. But the choice of qualifier for ambiguous titles that need a qualifier is up to the subject-matter project. So for instance, the film project might require that Avatar (2009 film) be distinguished from other less-well-known films named Avatar, while the music project might allow Thriller (album) despite the existence of Thriller (Lambchop album). From a disambiguation perspective, either works -- the articles end up with unique titles, and hatnotes or a disambiguation page will get you to the others. -- JHunterJ (talk) 12:28, 3 November 2011 (UTC)

But surely what's best for the reader is the same for all subjects - i.e. it's not related to wikiprojects (which obviously consist of editors not readers). Dpmuk (talk) 12:40, 3 November 2011 (UTC)

(edit conflict)I started the discussion on Jenks24 talk page and so feel I should comment here. The whole point of our article naming rules is to get users quickly to the page they want without causing confusion / surprise. In my view this is best served by allowing primary topics for a disambiguated title (in this case having the English footballer at plain "(footballer)" - assuming it is determined he's primary). As I see it the argument for not allowing a primary topic essentially comes down to the fact that no one is likely to search for "Person X (foorballer)" so it makes no difference how we disambiguate as people will arrive at the "Person X (disambiguation)" page and having unique identifiers will probably surprise people arriving at the disambiguation page least. My counter argument is that someone may search for "Person X (footballer)", however unlikely, and so primary topic should apply here as well. I think that the extra convenience for our readers that allowing primary topics for disambiguated pages outweighs their surprise as ultimately surprise won't stop them getting where they want to go. I do however see both sides and my support for allowing primary topics on disambiguated pages can best be described as weak. I'm also more than happy to be persuaded either way as there may well be points I've missed. Dpmuk (talk) 12:38, 3 November 2011 (UTC)

"But surely" often isn't sure. What's best for the reader is identifying the primary topic (if any) and adding qualifiers to the rest. The purpose of qualifiers is to allow the articles to exist at all, since there are good technical reasons why two articles can't have the same title. Once unique titles are identified, the reader is best served: they can get to their sought article by searching (primary topic) or navigating from a search (hatnotes and disambiguation pages leading to non-primary topics). The qualifier in the title is a technical requirement, and unrelated to the content of the article or the navigational path there. The closest it can get is when a Wikipedia-savvy reader skips the built-in navigational tools and edits the URL or guesses at the qualifier in the search box -- if they enter "Avatar (film)" they'll find themselves at a dab page, but if they enter "Thriller (album)", they'll find themselves at an article. I don't see either as obviously better than the other, so it seems like a fine inconsistency to leave in the hands of the content projects. Not to mention that I would expect serious and unneeded drama if the disambiguation project tried to dictate qualifiers to the content projects. -- JHunterJ (talk) 13:28, 3 November 2011 (UTC)

The point I was trying to make, and I may have included some too strong statements in making it, is that we're here for the reader and I'm not sure that a bunch of different decisions by wikiproject members is going to result in what's best for the reader. Dpmuk (talk) 13:44, 3 November 2011 (UTC)

The point I was responding with is that that conclusion isn't foregone. A bunch of different decisions by Wikiproject members might indeed result in what's best for the readers, or at least something that's no worse than what Emerson called a "foolish consistency". We are here for the reader, and so are the editors at the content Wikiprojects. -- JHunterJ (talk) 13:58, 3 November 2011 (UTC)

And that's a valid point - it's my view that it's best decided across the board and not by wikiprojects but I made that point too forceably and implied something was a fact when it may not be. That was a mistake. We obviously disagree about how to get to what's best for the reader but we both have the same end goal. As I say on this topic I don't have strong feelings what the right answer is I just want what's best for readers and some consistency so that we don't have different outcomes on very similar pages as that's definitely worse for readers. Dpmuk (talk) 14:21, 3 November 2011 (UTC)

I expect that very similar pages would have consistent qualifiers for non-primary topic, since they would be covered by a single content project. So there should be (and would be) consistency within a content segment. -- JHunterJ (talk) 14:57, 3 November 2011 (UTC)

There's a similar discussion at Talk:Pat Murphy (writer) - a new stub is for Pat Murphy (sports journalist) but its creator wants the "... (writer)" to be renamed to something like "... (science fiction writer). I reckon I've fixed the problem by improving the dab page. PamD 14:07, 3 November 2011 (UTC)
The concept behind the policy at WP:PRECISION and guideline at WP:PRIMARYTOPIC is the same whether or not an article title has a parenthetical disambiguator; namely, that titles should be as easy to search for and link to as possible for the greatest number of people. Wikiprojects should not overrule policy. While only a small percentage of readers search for the phrase "Avatar (film)" or "Avatar film", that still amounts to a few thousand every month. Since over 95% of them are looking for the 2009 film as opposed to two earlier films, we make it more difficult than necessary for those 3 or 4 thousand people by forcing them to a dab page. In the case of Independence Day (film), the opposite situation applies; most people typing that phrase get straight to the article they want, where the small minority seeking the 1983 film click directly on a hatnote. There are detailed discussions at Talk:Independence Day (film). The numbers for Ray Wilson will be much smaller but if Ray Wilson the English footballer is the primary topic among footballers, there's no good reason to make even a few people go through a dab page unnecessarily, especially since the 2 or 3 other footballers can be listed in hatnote. Hatnotes are often a less intrusive means of disambiguation than lengthening an article title. Station1 (talk) 03:51, 5 November 2011 (UTC)

No piped links on disambiguation pages

There seems to be an epidemic of these, but there is good reason for it now, and the page should probably reflect that. Italics are used in displaying the names of articles that are not italicized in their entirety but include italics, which wouldn't work if the links themselves contained italics.Vchimpanzee · talk · contributions · 18:30, 4 November 2011 (UTC)

This is covered in detail in a section of the Manual of Style. But I suppose it wouldn't hurt to add a link to WP:PIPING after the sentence "Do not pipe disambiguation links. Showing the entire linked article title avoids confusion, which is the reason for the link in the first place." under usage guidelines. France3470 (talk) 18:35, 4 November 2011 (UTC)

WP:PIPING has this sentence in bold: Subject to certain exceptions as listed below, piping or redirects should not be used in disambiguation pages. Under Exceptions, the sub-section Where piping may be appropriate addresses this. Short repeating things in the guideline, what else can be done to make this clearer? older ≠ wiser 18:51, 4 November 2011 (UTC)

I pointed this out since I had seen a guideline with no exceptions, and may have inappropriately changed some entries on other pages. However, I believe the need is limited to those situations where the entire title is not in italics for some reason, and putting italics within the link would require a redirect, which would be pointless.Vchimpanzee · talk · contributions · 19:22, 7 November 2011 (UTC)

I've added another link to the list of exceptions, which already includes the use of italics for titles. WhatamIdoing (talk) 15:26, 8 November 2011 (UTC)

Dabs with honorific titles

I've come across two cases recently, Rabbi Yochanan and Reverend John Hancock which has me raising the question 'are these proper dabs?' In one respect the items listed at these pages are partial title matches (WP:PTM) in that none of the entries contain the word "Rabbi" or "Reverend". But in a catch-22 situation the articles simply cannot, because titles should not contain honorifics, titles, professional qualifications etc. In my mind these types of pages should not exist because they are potentially problematic and could set a precedent for all sorts of dabs which would be nothing more than lists of people with Foo (title-qualification-profession). Any thoughts? France3470 (talk) 00:48, 19 November 2011 (UTC)

I am aware that Reverend John Hancock is not the best example, as it does contain an item which is not a name but it still work as a jumping off point about whether these items are partial matches. France3470 (talk) 00:54, 19 November 2011 (UTC)

Rabbi Yochanan gets more than 100 hits each month, and at least some of the entries are for topics that could be known as "Rabbi Yochanan". The titles themselves do not need to contain the word "Rabbi" -- these are not WP:PTM examples. If Rabbi Yochanan could have been a redirect to more than one article (like Rabbi Akivah is a redirect to one article), then there's ambiguity to be resolved, even if the naming conventions say that the honorific title should only be a redirect to the non-honorific title. IMO. -- JHunterJ (talk) 15:08, 19 November 2011 (UTC)

I agree that these pages can be valid in order to resolve ambiguity but I also think that the issue is more complicated. For instance, should the entries on these pages use a redirect, to show the title with the honorific name, as opposed to the actual article title. WP:DABREDIR suggests that such redirects should be used or else the term should appear in combination with the article title separated by an "or". Perhaps this would be bordering on unnecessary, I don't know. (This is more an issue of how to style such as page than my initial concern over validity.) I suppose I'm not wholly convinced that hands-down all of these types of pages are automatically valid and still feel these should be very much case by case situations. As you pointed out, Rabbi Yochanan is clearly a reasonable and well used search term, so there should be no reason not to have it. But take for instance, Duke William (disambiguation) as an example. Would this be a valid dab page to create? There are a large number of Williams who are also Dukes and who might reasonably be referred to as Duke William. However, these subject would most likely be commonly refer to as something else. In another situation, say we created Major Fisher which contains Foo R Fisher, Foo Fisher, Foo Fisher (major). Would there be a benefit to such a page? What would be the advantage of it being a separate page as opposed to it being as redirect to a section title "Major" in the dab page Fisher (as we treat incomplete dab pages)? France3470 (talk) 18:44, 19 November 2011 (UTC)

I do not see this as a special case. (Q) Is the title ambiguous? (N) No: no dab needed. (Y) Yes: Is there a primary topic? (Y.N) No: Dab page or redirect to dab at the base name (Y.Y) Yes: Can the ambiguity be handled through hatnotes on the primary topic? (Y.Y.Y) Yes: hatnotes it is. (Y.Y.N) No: dab page or redirect to dab page at (disambiguation) title. If the dab page is necessarily a subsection of another dab page, the redirect can be used for editor efficiency. However, if anyone wants to create or maintain the separate dab page, there is no harm to the reader and possibly a benefit in loading and reading a shorter dab page to find the sought article. Honorifics do not make for incomplete disambiguations or partial title matches, as far as I can tell. -- JHunterJ (talk) 20:46, 19 November 2011 (UTC)

For the record, I have previously created 54 "Justice Foo" disambiguation pages, from Justice Adams to Justice Wright, to accommodate the fact that lawyers in a given jurisdiction seem to be predisposed to write as if they believe that theirs is the only one in which a high court exists. To lay out all the Justices Smith in the surname page for Smith would be redundant (as they are already listed on that page by their full name). bd2412 T 22:16, 19 November 2011 (UTC)

I think this resolves any concerns I had. Must have been thinking about this too much. I've gone ahead and removed the incomplete dab tag from Reverend John Hancock on the basis of this discussion. Thanks, France3470 (talk) 01:25, 20 November 2011 (UTC)

Links to disambiguation pages in 'See also'

Practice is not to pipe links to dab pages in 'See also' sections to make it clear to the reader that it is a dab page to which they are being taken. I recently came across an instance of [[Foo (disambiguation)|Foo]], which I am sure is sub-optimal, but I can't find this situation discussed specifically in the guideline. I think that we should add a clarifying sentence to section 'Links to disambiguation pages'. Thoughts welcome, please. Bridgeplayer (talk) 03:37, 20 November 2011 (UTC)

I agree with you on the practice. I fix them when I see them. The guidelines mention not piping entries but don't explicitly call out the See also section. I've got no problem if the language can be made clearer. -- JHunterJ (talk) 03:56, 20 November 2011 (UTC)

Speedy deletion of "Foo (disambiguation)" redirects

Greetings! User:Bwilkins has speedily deleted several "Foo (disambiguation)" redirects (which I have since restored). I tried to inform him that these redirects were mandated by WP:INTDABLINK, but he has asserted that I "either don't understand the purpose of a disambig page, or really don't understand that policy!" See User talk:Bwilkins#Intentional disambiguation redirects (context). Can somebody else please set him straight? Thanks! bd2412 T 17:55, 20 November 2011 (UTC)

This is pretty relevant to the above discussion. Although we often cite WP:INTDABLINK as the reason for creating Foo (diambiguation) pages the guideline doesn't really explain this perhaps as clearer as it could. I have just re-read the Links to disambiguation pages about 5 times and cannot find anywhere that actually states to pipe through (disambiguation), (hopefully I'm not being blind) although it is mentioned at WP:MOSDAB#"See also" section (and oddly says to refer to WP:INTDABLINK. I think we could potentially add a statement onto the end of "When to link to a disambiguation page...Links from one disambiguation page to another for further disambiguation: British has a link to Britain (disambiguation) for further disambiguation." Which says something about how these types of links need to be piped through (disambiguation). France3470 (talk) 18:17, 20 November 2011 (UTC)

It's in the section How to link to a disambiguation page.--ShelfSkewed Talk 18:22, 20 November 2011 (UTC)

Not really. I've always interpreted this as being more general. Merely saying that if one wants to link to a disambiguation page that should do so using (disambiguation) to show that this is intentional. And that if such a redirect does not exist it can be created. It doesn't explicitly advocate the practice but rather advises that "there is nothing wrong" with it. This also does not address links on disambiguation pages nor state that on dabs, links to other dab pages should also contain (disambiguation) it just implicit. France3470 (talk) 18:38, 20 November 2011 (UTC)

It seems to me that the guideline is direct and declarative: "link to the title that includes the text '(disambiguation)', even if that's a redirect" and "If the redirect does not yet exist, create it". The final sentence ("there is nothing wrong") is not a signal that these policies are optional but is, rather, an attempt to anticipate the objections of editors who think that redirects are somehow "broken" or otherwise undesirable.--ShelfSkewed Talk 18:49, 20 November 2011 (UTC)

In all of the deletions done by User:Bwilkins, the "Foo (disambiguation)" redirects were being used to route intentional links from other disambiguation pages. I think we need to seek an amendment to the language governing speedy deletion of redirects to at least say that a "Foo (disambiguation)" redirect pointing to an existing disambiguation page is not a candidate for speedy deletion. Cheers! bd2412 T 19:03, 20 November 2011 (UTC)

I think that for those who are actively involved in disambiguation this section is perfectly clear. All of us here know these redirects are a good thing. But I doubt that it as clear for those who don't spend so much time working with these types of pages. I too have seen so many of these redirects being speedy deleted recently (despite the fact that they were created citing WP:INTDABLINK) which indicates to me that there might be some misunderstanding at play. France3470 (talk) 19:15, 20 November 2011 (UTC)

I have initiated a request for a corrective amendment to the speedy deletion criteria at Wikipedia talk:Criteria for speedy deletion#"Foo (disambiguation)" redirects created in accordance with WP:INTDABLINK. Hopefully such an instruction will at least give pause to overeager deleters. Cheers! bd2412 T 19:34, 20 November 2011 (UTC)

Those redirects never should have been deleted under R3 in the first place. Even if an admin doesn't understand the reason for the redirect or thinks it is not helpful, it is clearly not an obvious typo or misnomer, so it should have been taken to AFD instead of speedy deletion. --R'n'B (call me Russ) 21:46, 20 November 2011 (UTC)

That is exactly why I am proposing that R3 be clarified to specify this. bd2412 T 21:48, 20 November 2011 (UTC)

Double redirects

Is there a way to ensure that a deliberate double redirect via a dab link doesn't get "tidied up" into a single redirect. Look at Saint John, a dab page. The incoming redirect from St. John should presumably go to Saint John (disambiguation) and thence to the dab page. It doesn't, and from the history hasn't done in the past. But if it did, a bot would "helpfully" take out the intermediate step, as far as I know! Can one tag a redirect in some way to stop this happening? Or does the double-redirect bot know about dab redirects already and avoid this? PamD 19:31, 20 November 2011 (UTC)

Actually, I don't think that is correct. St. John should still point to Saint John, because a link to St. John is likely an error needing to be disambiguated. In the rare event that a link to St. John is intentional, then either it should be piped as [[Saint John (disambiguation)|St. John]], or (if it is in a hatnote or "See also" section") it should be changed to [[Saint John (disambiguation)]]. bd2412 T 19:37, 20 November 2011 (UTC)

You're right. Bad example. But I'm sure I've come across some case, possibly not even in connection with disambiguation, where it was useful to redirect "FooA" to something ("FooB") which might eventually have become an article but was at present a redirect (to "FooC"), and where a bot is likely to helpfully redirect "FooA" to "FooC" so that when the "FooB" article is written it will not be linked from "FooA". Ah well, too tired to think straight tonight. PamD 19:43, 20 November 2011 (UTC)

Double redirects don't work. Period, end of story. A reader who clicked on "FooA" in your example would not be taken to "FooC" but to an ugly redirect "FooB" page that says they need to click again to get to "FooC". This is a negative user experience, and we don't do it. Ever. --R'n'B (call me Russ) 21:44, 20 November 2011 (UTC)

SIA vs. disambiguation of encyclopedic topics

Disagreement at Prikubansky could use additional eyes. I've tagged it for cleanup. -- JHunterJ (talk) 22:20, 21 November 2011 (UTC)

Naming the disambiguation page

I suggest that we add this:

According to the principle of least astonishment, the spelling of a word in general English is preferred when it is used with various spellings in different article titles, especially when one of these articles uses the most common spelling itself or is a synonym of the disambiguation page.

It would seem to me that the following already provides sufficient support for my reasoning on Talk:Flim-flam for the move to the most common spelling Flimflam:

The simplest form of the term is preferred to those containing punctuation, diacritics and articles; for example SA is preferred to S.A., and Shadow (disambiguation) is preferred to The Shadow (disambiguation).

But perhaps we need to add something like i suggested to avoid making many editors use up lots of time on similar discussions.--Espoo (talk) 21:26, 25 November 2011 (UTC)

I use the spelling that is common to the greatest number of the ambiguous topics. Which I could also say would be the least astonishing. The spelling of the word in general English would be a rule for a dictionary entry, not for a Wikipedia disambiguation page. If the "most ambiguous" name has periods or hyphens or the article "the", then use it. Or split the disambiguation page, which is often done in the case of the article "the" if the list is long enough to have editors discussing which title should be used. -- JHunterJ (talk) 00:31, 26 November 2011 (UTC)

Please participate in Talk:Flim-flam so that this discussion is not too theoretical because that example may convince you why your method doesn't always work and would cause unnecessary astonishment in that and similar cases. --Espoo (talk) 06:11, 26 November 2011 (UTC)

Your suggestion above is about changing or expanding the guidelines here; discussion about that change should be done here. The discussion about the titling of the specific disambiguation page can be had at the specific disambiguation page's Talk. -- JHunterJ (talk) 13:07, 26 November 2011 (UTC)

I just participated there. I saw nothing that indicated that my method doesn't work or that it would cause any unnecessary astonishment. -- JHunterJ (talk) 13:16, 26 November 2011 (UTC)

First name disambigations - deletion of

Please see my comments at Template_talk:Given_name#Too_much_disambiguation.3F. --_{Piotr Konieczny aka Prokonsul Piotrus| talk to me} 23:10, 28 November 2011 (UTC)

"Hatnotes" vs "disambiguation links"

The text of the guide mostly refers to "hatnotes", but the two relevant section headings (and the text of one subsection) refer to "disambiguation links". Since "hatnote" is a more specific term, I suggest we change "disambiguation link(s)" to "hatnote(s)" wherever possible. After all, one could argue that a link in a disambiguation page is also a "disambiguation link". Whenever I come to the guide looking for info on hatnotes, I scan the Contents table and am briefly perplexed that I cannot find hatnotes. Nurg (talk) 01:11, 3 December 2011 (UTC)

Some hatnotes are disambiguation links and some disambiguation links are hatnotes, but not all hatnotes are disambiguation links and not all disambiguation links are hatnotes. I think both terms might be used where they are correct. That said, I have no problem with a bold edit to consolidate on the way you think they should read, and then possible subsequent reverts or expansions where other editors think the other term is better or also correct. -- JHunterJ (talk) 15:56, 3 December 2011 (UTC)

Re:Dabs with honorific titles, and Sir Ellis

Tonight, I came across Sir Ellis which is currently prodded for deletion. This seems very relevant to the discussion I initiated a few weeks ago, see Wikipedia talk:Disambiguation#Dabs with honorific titles about the validity of dabs which differentiate names using honorifics. I'm about to head to bed but I'm wondering if we need to re-investigate this issue; from where I stand there seem to be some differing opinions. France3470 (talk) 02:50, 9 December 2011 (UTC)

Disambiguation of sub-articles

I do not see the disambiguation of sub-articles formally addressed here. For example, why would it be Georgia (U.S. state) wine instead of "Georgia wine (U.S. state)" or List of The Twilight Zone (1985 TV series) episodes instead of "List of The Twilight Zone episodes (1985 TV series)". Or Category:Mercury (planet) in fiction instead of "Category:Mercury in fiction (planet)". In other words, the disambiguation term in brackets is placed immediately after the ambiguous term in the title, and not necessarily at the very end. The reason I ask is that I noticed an edit dispute on my watchlist, and as an uninvolved admin I would like to set the record straight with those parties involved. And if that means actually adding this rule, I'd be in favor of that. Thanks. Zzyzx11 (talk) 06:16, 5 December 2011 (UTC)

I would also be in favor of specifying that a qualifier, when needed, be placed at the end of a title instead of inserted in the middle. OTOH, determining the way to make titles unique has so far been left up to the content projects. -- JHunterJ (talk) 12:58, 5 December 2011 (UTC)

"Georgia wine (U.S. state)"? I've been to Georgia, but never to a state called "Georgia wine". What is "fiction (planet)"? :-) Adding a rule about this would seem to be defy WP:COMMONSENSE and cross the WP:CREEP line, unless this is coming up all the time. These articles are all "X Y", "X in Y" or "X of Y Z" form, where X is "A (B)"; The "(B)" isn't an independent operator in the equation. — SMcCandlish Talk⇒ ʕ(^Õل^ō)ˀ Contribs. 01:14, 17 December 2011 (UTC)

Are sub-articles inappropriate for a disambig page?

I'm involved with a topic/article (RMS Titanic) where the main article is huge and the topic is huge (lots of sub-articles and related articles) with the sub articles mostly (or all) linked in-line or as "see also" in the individual sections. I think that it would be useful to communicate which sub-articles exist, I've been at the article for a half year and am still learning which exist; it's very hard to see this from in-line links and notes in the section headers. I first thought of listing them in "see also", but I think that the guidelines for that clearly exclude articles that are already linked in the article. So I though of using the disambig page to list them and then listing the dis-ambig page as a "see also" in the main article, there being no specific exclusion of this here. I was reverted at both, the person saying that dis-ambig pages are only for listing where there is true ambiguity and that the "see also" section should be used for this. I was thinking that folks here would have more of persepctive on this than I do on this. Is it a common or OK practice to list sub-articles on the dis-ambig page? If so, possibly a mention of that could be added. And if not, do any folks have experience on the best way to do this? North8000 (talk) 16:06, 14 December 2011 (UTC)

I think the usual practice is to create an index article, with the title, in this case, of Index of RMS Titanic-related articles (see, for other examples, All pages with titles beginning with Index of).--ShelfSkewed Talk 16:16, 14 December 2011 (UTC)

Thanks! I'll have to read up on those.... I didn't even know that they existed. Sincerely, North8000 (talk) 20:21, 14 December 2011 (UTC)

Alfonso Gomez (disambiguation)

Should there be an Alfonso Gomez (disambiguation) for Alfonso Gómez, Alfonso Gómez-Lobo, Alfonso Gómez Méndez, Alfonso Gomez-Rejon and Alfonso Gomez vs. Saul Alvarez?--TonyTheTiger (T/C/BIO/WP:CHICAGO/WP:FOUR) 00:06, 17 December 2011 (UTC)

No, unless Gómez-Lobo, etc., are sometimes known as Alfonso Gomez/Gómez without the additional surnames, in which case it would be an {{hndis}} page. And the fight article wouldn't be on that page at all, since it's linked to from the boxer's article. — SMcCandlish Talk⇒ ʕ(^Õل^ō)ˀ Contribs. 01:19, 17 December 2011 (UTC)

Primary topic and age or derivation arguments

A common argument (seen most recently at Talk:Engelbert Humperdinck but also at Talk:Anne Hathaway and others in the past) is that a particular topic is primary because the other use(s) came later, and sometimes additionally because the later topic was named in honor of the earlier one. This argument has no basis in policy or practice, yet it continues to be brought forth. Is this something we should look at making clearer in this guideline? Powers ^T 16:07, 20 December 2011 (UTC)

Winston Churchill is a good counter to that argument. But I would love to (re)add the criteria that do not factor into primary topic consideration, except inasmuch as they impact the existing criteria (usage and notability/educational value): age, derivation, size, population, recentness, etc., in and of themselves do not figure into it. -- JHunterJ (talk) 16:30, 20 December 2011 (UTC)

I don't agree. Ideally, the guideline describes the actual practices used in determining primary topic. It would be inappropriate to edit the guidelines to suit the preferences of individual groups of editors. The fact is that derivation in various forms is a consideration in many discussions. We could perhaps attempt to summarize or describe the limitations of using such arguments. That is, as far as I'm concerned, derivation alone should never warrant placing a topic as primary that fails primary topic by other criteria. Where it is occasionally a valid consideration is to support arguments that a disambiguation page should be at the base name, even if there is a topic other than the "original use" that might otherwise meet criteria for primary topic. older ≠ wiser 17:06, 20 December 2011 (UTC)

I don't understand your objection. I'm not suggesting changing anything "to suit [my] preferences"; I'm suggesting that what we have already is insufficiently clear, as an argument that has no basis in policy or practice is still being promulgated by some editors. Powers ^T 00:20, 21 December 2011 (UTC)

If it is successfully being used in discussions, then perhaps the guidelines do need to be updated. There is nothing in policy that prohibits it from being used. That you don't like it being promulgated by some editors is not a reason to change the guidelines to discourage its use. older ≠ wiser 01:01, 21 December 2011 (UTC)

It's an argument that is rarely successful, as far as I can tell, and the few exceptions where it has succeeded, like at Corvette, can be written off as either WP:IAR or something to be rectified in the future. I don't think it has been successful often enough to warrant reflection in the policy, and I agree that adding more clarification on this point is a good idea. --Born2cycle (talk) 02:15, 21 December 2011 (UTC)

I also don't understand that argument. If whatever is being used in discussions is to be fair game, the entirety of the guideline should be removed and replaced with "whatever editors feel like bandying about this time, on any given page". Lousy guideline, IMO, and I'd rather provide guidance such as hey, look at the actual usage and educational value, not the editors' idea of what readers should be looking for (who cares that everyone is looking for article A, A was named for the never-sought article B, so article B is primary! That'll learn 'em!). Plus, Winston Churchill. -- JHunterJ (talk) 02:57, 21 December 2011 (UTC)

That's not how it works and you know it. Guidelines describe practices that have community support. If a particular rationale is used repeatedly, that would appear to be an indication of community support. Revising guidelines to deprecate the use of such a rationale because you don't like it, is wrong. older ≠ wiser 03:35, 21 December 2011 (UTC)

Actually I'm supporting how I think it works: communicate to other editors that simple age or derivation is not how current consensus for determining primary topics goes, and using them in ignorance of the guidelines is wrong -- if you're using them for WP:IAR, fine (with a reason to ignore the rules), or if it actually is coming up beyond isolated cases among editors who are not familiar with these guidelines, then we can reflect that new consensus. Until then, editors using those arguments in discussions despite the guidelines should be discouraged, and explicitly so in the guidelines. Analogously, if a bunch of AfD discussions start using WP:IDONTLIKEIT (and they do), that's a reason to clarify the guidelines to indicate that that's wrong (and they have been, long ago), not an indication that WP:IDONTLIKEIT is acceptable. -- JHunterJ (talk) 04:51, 21 December 2011 (UTC)

Or, in other words, using the criteria listed instead of age and derivation does have community support, and the guidelines should clarify that since apparently they are not always understood that way. -- JHunterJ (talk) 04:52, 21 December 2011 (UTC)

And conveniently pretend that there is not any disagreement. Sweet. older ≠ wiser 13:39, 21 December 2011 (UTC)

Tell me, Bkonrad, then why do we have guidelines at all, if we can't make sure they accurately reflect current consensus? Powers ^T 18:49, 21 December 2011 (UTC)

And where is consensus if there is disagreement? older ≠ wiser 19:31, 21 December 2011 (UTC)

Isolated disagreement is not indicative of a lack of consensus. Powers ^T 01:04, 24 December 2011 (UTC)

Persistent or isolated. Guess it depends on your perspective. I'd say the amount of discussion on the matter indicates there is not a very strong consensus, if any, to exclude considerations as as you propose. older ≠ wiser 02:35, 24 December 2011 (UTC)

Isolated persistence does not equate to a lack of consensus. Vegaswikian (talk) 02:58, 24 December 2011 (UTC)

"Isolated persistence" sounds like an oxymoron. If by it you mean that the persistence is by isolated individuals, I think a review of past discussions show a range of people expressing similar concerns. older ≠ wiser 03:35, 24 December 2011 (UTC)

I do not pretend there is no disagreement. I recognize that the disagreement is not a new consensus, not a new lack of former consensus, in the Wikipedia usage of "consensus". -- JHunterJ (talk) 16:19, 24 December 2011 (UTC)

OK, so my point is that it is inappropriate to revise the guidelines to favor one's own preference in matters where there is disagreement. older ≠ wiser 03:02, 25 December 2011 (UTC)

Clarification of what the guidelines already say is not the same as revising the guidelines to say something different. Age and derivation have not been criteria, and are not criteria, and making that exclusion explicit doesn't change that. -- JHunterJ (talk) 17:58, 25 December 2011 (UTC)

Some editors conflate age or derivation with arguments about importance. While I'd likely often agree with you about specific cases in that age or derivation alone is seldom an good indicator that a topic is a primary topic. I think there are a significant number of experienced editors who do see age or derivation as being stronger indicators of PT than we might. And besides, I tend to agree with other that while age or derivation alone is not sufficient to establish a primary topic, it may be a factor to consider in determining whether some other topic is primary or if a disambiguation page should be at the base name. older ≠ wiser 23:59, 25 December 2011 (UTC)

Disambiguating when the disambiguator is ambiguous

Nice section title, isn't it? Here's the situation: I came across a dab page where several similarly-named school articles included the city to disambiguate; i.e., School (City). So far, so good. A couple of them, though, used the city and state to disambiguate, even though disambiguating with just the city would make each of the school articles uniquely titled. Further complicating things is that some of the ambiguous cities were in the US, where we automatically include the state in articles about the cities, and some of them were in countries that have no such automatic disambiguation policy. When I moved the US ones to School (City) for consistency with the rest, someone came to my talkpage and claimed there were guidelines that said when US cities are used to disambiguate, we use "City, State".

So my questions are: 1) In general, when a disambiguator includes a topic that is ambiguous on its own, but otherwise would sufficiently disambiguate, can we use just the ambiguous part of it? 2) If we can, do other naming conventions/project guidelines override that? Interested in your thoughts. P.S. The page at issue is Salesian College. Dohn joe (talk) 22:14, 20 December 2011 (UTC)

Our convention for disambiguating places in the U.S. is to use the city and state for clarity. I'm not sure if it's documented anywhere. Powers ^T 00:18, 21 December 2011 (UTC)

I know that's the convention for articles about the places themselves. But what about when they're just being used to disambiguate something else? The example here was Salesian College (Farnborough) vs. Salesian High School (New Rochelle, New York). The first school is located in Farnborough, Hampshire (note the disambiguation), but in the school article, only "Farnborough" is used, whereas in the second, "city, state" is used even though "New Rochelle" alone was sufficient to disambiguate. So this isn't about the city articles themselves - does that clarify the issue? Dohn joe (talk) 00:33, 21 December 2011 (UTC)

I'm sorry that I wasn't clear. Salesian College and Salesian High School are both "places", and that is the sense which I meant when I used the word. I did not mean to restrict the meaning of the word "place" to just "communities". Powers ^T 18:51, 21 December 2011 (UTC)

The convention to disambiguate cities with state never had consensus support since the half dozen or so people agreed to implement it, but we never had consensus to stop following it either. So we definitely never had consensus support for including the state as a disambiguator. Including the state when it's not needed is certainly contrary to WP:PRECISION. --Born2cycle (talk) 01:05, 21 December 2011 (UTC)

Just quickly, the answer to Q1 is definitely yes. A good example of this is the thousands of articles that use "(footballer)" as disambiguation. Jenks24 (talk) 01:18, 21 December 2011 (UTC)

That was perhaps too quickly, since Footballer redirects to Football player, an article, not a dab page, so "footballer" is not ambiguous. I can't think of an example where the disambiguator is the title of a dab page (or is a redirect to a dab page).

So I think my answer to Q1 is no, since the disambiguator must be a primary topic itself, or at least it should not be an ambiguous title which is, or redirects to, a dab page. But note that that means that using just a U.S. cityname as a disambiguator is perfectly acceptable as long as cityname is either the title of an article (e.g., San Francisco), or redirects to one (e.g., Carmel-by-the-Sea). But, yeah, if the city is truly ambiguous, as in, say Rosemont, it probably should be disambiguated itself to be used as a disambiguator. --Born2cycle (talk) 01:42, 21 December 2011 (UTC) Revised (see below). --Born2cycle (talk) 02:00, 21 December 2011 (UTC)

In my defence, a) football player is really a WP:CONCEPTDAB and b) it only recently became an article (October 30) :) Jenks24 (talk) 09:07, 21 December 2011 (UTC)

So, for the examples at hand, Salesian College (Farnborough) should be moved to Salesian College (Farnborough, Hampshire), because Farnborough is a dab page? Dohn joe (talk) 01:49, 21 December 2011 (UTC)

Oy. Well, per what I just said off the top of my head, yes, but now that I think about this some more, no. What's important is uniqueness in the given context - and the given context here is topics named "Salesian College". So, if there is only one "Salesian College" to which Farnborough can apply, then, yes, that's all that's required. You need to disambiguate the disambiguator further only if more than one Salesian College is in a city named Farnborough.

So, I'm striking what I said above and am restating here.

So my answer to Q1 is yes, the disambiguator can be an ambiguous name itself, since the disambiguator needs to be unique only in the context of the name being disambiguated, not in all of Wikipedia. But if the disambiguator applies to more than one of the uses of the name being disambiguated, then it does need to be disambiguated further. --Born2cycle (talk) 02:00, 21 December 2011 (UTC)

Okay, I found an example... Whiz Kids (Ford). Note that Whiz Kids is a dab page, listing over half a dozen uses. One of the uses is associated with the Ford Motor Company, and is disambiguated with Ford, even though Ford is a dab page itself. That's because in the context of "uses of 'Whiz Kids'", "Ford" alone distinguishes one use from the rest, and that's all that is required of a disambiguator. --Born2cycle (talk) 02:07, 21 December 2011 (UTC)

Another example: Spring Hill, New South Wales (Orange). Clearly "orange" is ambiguous (color, fruit), and yet here, int this context, it's a perfectly adequate disambiguator for Spring Hill, New South Wales. But now that I found that one, I see that the only other use of "Spring Hill, New South Wales", Spring Hill, New South Wales (Wollongong), is now a redlink, so Spring Hill, New South Wales (Orange) should probably be moved to Spring Hill, New South Wales, but it's still a useful example, I think. --Born2cycle (talk) 02:27, 21 December 2011 (UTC)

Why School (City) format? Why not to name Salesian High School (New York) when there are no other schools of the same name in the state? --Kusunose 03:08, 22 December 2011 (UTC)

Depends on the natural disambiguator. I would support state as preference without any other reason being advanced, or even [[Template:Null|Salesian High School (US). But I wouldn't go to [[Template:Null|Salesian High School (North America). Rich Farmbrough, 15:34, 28 December 2011 (UTC).

Engelbert Humperdinck

Fans of disambiguation may want to participate in the heated debate at Talk:Engelbert Humperdinck#Requested move. Is disambiguation really necessary when there are only two topics, neither of which is primary? Thanks, Bazonka (talk) 09:34, 23 December 2011 (UTC)

Yes, if there's ambiguity with no primary topic, a dab goes at the base name, even if there are only two topics for the ambiguous title. -- JHunterJ (talk) 13:29, 23 December 2011 (UTC)

I totally agree, but not everyone else does. The opinions of others at Engelbert's talkpage would be welcome. Regards, Bazonka (talk) 15:15, 23 December 2011 (UTC)

Yeah, the debate a few topics below is entirely about this question, and while not vitriolic, it's certainly heated. As for E. H., it's a no-brainer: Probably 99% of people over the age of 30 in the English-speaking world have heard of the singer, and probably less than 0.01% of them have heard of his namesake, a minor arts-historical figure who really seems to be remembered in large part because of his funny name. Two clearly notable operas and some more minor ones, all in German, but his magnum opus is simply a musical folktale retelling. Just being earlier doesn't make him the primary topic, even in music. An obvious counterexample would be Eddy Merckx vs. Eddy Merckx (billiards player). They're both world champions in their respective notable sports, both also have received Belgium's highest civilian award as national sports heroes, and they equally satisfy WP:ATHLETE, making them seem equal and suggesting a disambiguation fight (one that I actually brought, a long time ago, before I understood WP:DAB better). But the cyclist, the elder and possibly the namesake of the latter, is way, way, way more recognized, with literally thousands of articles written about him, over a long career, while the cueist is still active and not globally known (yet). In 50 years we might look back and say the younger guy is actually more notable, depending on what happens with the rest of his life. We have that kind of hindsight already in the case of the Humperdincks, and the answer is pretty clear, in favor of the younger E. H. (on English Wikipedia; on de:wiki, the other way around, almost certainly). — SMcCandlish Talk⇒ ʕ(^Õل^ō)ˀ Contribs. 00:37, 3 January 2012 (UTC)

Disambiguation of two topics

A recent discussion at Talk:Engelbert Humperdinck#Requested move looked into a situation where two topics exist with the same name, but neither seems to be the Primary Topic. Unfortunately, the discussion was closed inconclusively. However, a few interesting points were raised:

1. Should WP:TWODABS apply in this situation? Correctly applying it in its current form means that a dab page should be created, but some people felt that this was counter-productive:

If we have two articles, Foo and Foo (context), each with a hatnote pointing to the other, then people searching for the first article who type "Foo" will find it straight away; those looking for the second article will have to make an extra click, on the hatnote of the first article.
If we have Foo as a disambiguation page, and articles Foo (context 1) and Foo (context 2), then all people typing "Foo" will have to make a second click to find their article. So, it is argued that disambiguation makes things more cumbersome for those searching for the first article, and no different for people searching for the second. However, if neither article is the Primary Topic, then this approach means that people looking for the second article just have to load a dab page, not an inappropriate article.

2. The definition of WP:PRIMARYTOPIC seems to be inadequate/unclear.

What exactly is meant by "educational value", when should it apply, and by how much?
Stability should be a factor in determining primacy (e.g. a flash-in-the-pan movie may generate a lot of web action following its release, but not be of long-term interest). Stability is covered by the Importance clause ("enduring notability"), but not by the Usage clause. Should this be addressed?

3. Should the bar for determining the Primary Topic be lowered when there are only two topics?

Personally, I feel that no change needs to be made to WP:TWODABS, although perhaps the guidance could be more clearly worded to show that dab pages must be used in this situation. I have no strong opinion regarding point 2. And I do not feel that the Primary Topic bar should be altered. Bazonka (talk) 09:32, 27 December 2011 (UTC)

If there is an article at the base name for a title (or the base name redirects to an article), that is the primary topic for that title. Therefore, if there is no primary topic for an ambiguous title, the title must lead to a disambiguation page. Yes, a disambiguation page means that every reader will need to click twice, but if neither article is primary, that is the proper behavior.

You are absolutely right about the lack of useful clarity for "educational value". The addition of that criteria has not helped bring consensus to any primary topic discussion that I'm aware of, and has instead only helped increase the rancor at primary topic discussions, as editors who favor primaryness for a topic that readers are not seeking and will be surprised to find claim educational value with no consensus for it either. IMO, educational value should be reined in so that it applies only to things that no reasonable English-speaking reader would not be surprised to find at the base name (e.g., Apple leading to a fruit instead of a technology company), and not applied to topics that are dear to the hearts of some segments of the English-speaking population but not generally known (e.g., Engelbert Humperdinck). Bemoan the sorry state of education or "kids today" as needed, but Wikipedia is not a schoolmarm.

Usage can change, so primaryness can change. One editor's "flash in the pan" (used for popular topics that he has no use for) is another editor's subject for modern cinematic dissertation. If we can put some guidelines in for it (needing references beyond simple movie reviews, awards, journal citations, or reliable sources a year after release), that may help.

IMO, the "bar" can't be changed. Either there is a primary topic for a title or there's not. -- JHunterJ (talk) 14:18, 27 December 2011 (UTC)

Agree. I think there is a subtle flaw in the premise of this section that confuses the issue, namely that in the Engelbert Humperdinck case "neither seems to be the Primary Topic". Disagreement about which of two topics is primary is not the same as consensus that there is no primary topic. There are undoubtedly many thousands of two-entry dab pages that are completely uncontroversial because of consensus that there is no primary topic (example: Abdiel Colberg). In the Humperdinck case some people think the composer is primary, some the singer is probably primary, some think one or the other is primary without specifying which, and only some agree that "neither seems to be primary topic". WP:TWODABS says "If there are only two topics to which a given title might refer, and one is the primary topic, then a disambiguation page is not needed" [emphasis added]. So TWODABS can be correctly applied in its current form to the Humperdinck case if one believes, as some do, that there is a primary topic. It doesn't apply only if one believes that there is no primary topic. Station1 (talk) 20:31, 27 December 2011 (UTC)

I'm personally uncomfortable with Primary Topic, because it's permissive with the kind of thing WikiProject Disambiguation is working against: links going to places the author didn't intend. As a case in point, I had a rummage around in the links pointing to Engelbert Humperdinck and found over 10% were intended to be for the singer rather than the composer. At the moment, there's no guidance saying that nothing ought to link directly to an ambiguous title, even if it's the primary topic. That's a problem because the various tools don't regard linking to a primary topic as something worthy of disambiguating. But if one goes through the entire list of "What links here" for a primary topic and corrects all those that want to refer to the other terms, how does one maintain that "What links here" list? The only thing for it is to correct every article that links to the primary topic so that they either point to

the disambiguated redirect to the primary topic (i.e. explicitly linking to the primary topic via a redirect such as Engelbert Humperdinck (composer)); or
the intended alternative topic, such as Engelbert Humperdinck (singer)

leaving no links to the bare primary topic. Which makes me wonder, if that's the kind of effort you have to go through, and the tools don't support it, why allow Primary Topics?— Preceding unsigned comment added by Josh Parris (talk • contribs) 22:23, 27 December 2011‎

Because some ambiguous things have a topic that is primary: sought more than all others combined, and sought much more than any other. There would be little point in sending people to Science (disambiguation) rather than to Science when they enter "Science" in the search box, and similarly for William Shakespeare or Libya. -- JHunterJ (talk) 22:28, 27 December 2011 (UTC)

I agree that it's a useful concept. But far too often when there are just two topics, people insist on making one primary, like on the Engelbert H thing; sometimes, like there, there are even arguments about which one, which would appear to be strong evidence that neither is primary. Yet we tolerate it. These are sort of situations where maybe not a "ban" but a strong recommendation against picking a primary topic would be a great improvement. Dicklyon (talk) 22:33, 27 December 2011 (UTC)

I respond in sympathy with Station1's contribution, and with Josh Parris and Dicklyon. And I thank Bazonka for raising an important issue.

Station1, I'm no fan of the doctrine of "primary topic" as it is currently constructed. I think it does a great deal of harm, along with other ill-considered provisions at WP:TITLE (see discussion here, for example, and in the sections preceding that). But if we accept its utility for the moment, I agree with you: "Disagreement about which of two topics is primary is not the same as consensus that there is no primary topic," but also: "TWODABS can be correctly applied in its current form to the Humperdinck case if one believes, as some do, that there is a primary topic. It doesn't apply only if one believes that there is no primary topic." Disagreement over which is "the primary topic" (precariously predicated on the assumption that there is one) is strong evidence there is not a primary topic.

The Humperdinck case is one of several recently that illustrate how complex it all becomes, both in elucidation of principles and in applying them. The course of the discussion there makes an interesting study. Major reforms are needed at WP:TITLE, with fresh input from the wider community. It just isn't working, and the key provisions do not reflect consensus with anything like the claimed robustness.

Noetica^Tea? 22:43, 27 December 2011 (UTC)

Responding to no one in particular, it's long been my opinion that the guideline should have a stronger defaulting to a disambiguation page. Some have argued that because a disambiguation page always "the wrong page" and forces everyone to go through an extra click, in cases where there are exactly two approximately equal ambiguously titled topics it is preferable to arbitrarily select one to be primary so that approximately half of readers arrive at their desired target and the other half are still only one click away via a hatnote. I don't like this rationale for a couple of reasons. First, disambiguation in my opinion is not primarily about optimizing the number clicks that a reader has to make. Second, as noted above, having an arbitrarily selected "non-primary" topic at the base title makes it much more difficult to maintain the incoming links to the ambiguous topic. I think the bar for primary topic (regardless of whether there are two or more ambiguous topics) should be relatively high, although not such that any disagreement would disqualify the primary topic. For example, to select a topic I know Noetica and Dicklyon differ with me, I think French Quarter is correctly about the world-famous tourist destination in New Orleans. That there are other "French quarters" doesn't change the dynamics that the vast majority are looking for New Orleans topic. older ≠ wiser 23:22, 27 December 2011 (UTC)

Yes, we have to get away from the idea that disambiguation pages are bad or that there is always a primary topic. Vegaswikian (talk) 23:28, 27 December 2011 (UTC)

Disambiguation pages are bad. They represent our failure to get the reader to the content they want to read. Indeed, having primary topics does make it harder to maintain incoming links. But that's our problem, not our readers', and it's shameful to use editorial convenience as an excuse for sending readers to a disambiguation page. In my opinion, the only reason we should have a two-item disambiguation page is if it is plainly impossible to say that one or the other is the primary topic. That's the solution that serves our readers the best, even if it does make more work for us as editors. Powers ^T 01:22, 28 December 2011 (UTC)

In cases where there is an actual primary target, yes. But I disagree that there is any significant benefit to arbitrarily placing a non-primary topic at a base name under a mis-reading of WP:TWODABS. There is in fact nothing "wrong" with a disambiguation page -- it merely and only indicates that the title is ambiguous. Period. older ≠ wiser 01:32, 28 December 2011 (UTC)

As I said, if it's clear that neither topic is primary, fine. But if there are only two options, there's no reason to be overly strict about what constitutes a primary topic, because we should avoid sending readers to a disambiguation page if at all possible. Powers ^T 01:47, 28 December 2011 (UTC)

Why should we avoid this? I do not see that there is an encyclopedic benefit to avoiding disambiguation where the term is ambiguous. Wikiepdia is not a web site and the primary purpose of disambiguation pages is not to optimize web traffic. The web happens to be the primary (but not the only) medium for the encyclopedia, but apart from technical limitations that cannot be avoided I don't think media-specific issues should determine how topics are titled. older ≠ wiser 03:04, 28 December 2011 (UTC)

Why must the base assumption be that there always is a primary topic? Would the readers be better served by accepting the fact that, generally, there may not be a primary topic? Are we doing right by readers by assuming that we know best what they are looking for? Vegaswikian (talk) 03:16, 28 December 2011 (UTC)

One reason there is often a debate over primary topic is that it seems to be based on "what the man in the street would first think of", which is often boiled down editor opinion and googlehits. The problem with that is it is terribly prone to individualism, recentism, sportism and celebrity-ism. There may be a river in Germany called the "Foo" which has existed for centuries, but because "Fred Foo" happened to be a Canadian national hockey manager in the 90s, he gets to be the primary topic. Yet in 20 years time he will be forgotten and tourists will still be flocking in their thousands to the Foo Valley. If we tightened up the definition of "primary topic" it may help to reduce the amount of time wasted in debate. For example, it ought to count for something if the topic is the name-giver to all other subsequent uses, and perhaps more weight could be given to e.g. permanent features rather than people, bands, games and the like that, in many cases, have derived their name from the original. Just a thought. --Bermicourt (talk) 14:35, 28 December 2011 (UTC)

I agree that the definition of "primary topic" needs to be tightened (particularly the reference to educational value), but I disagree that the derivation of the name should be a factor. This is likely to cause more confusion as it cannot be applied consistently - in some cases it will apply (e.g. Birmingham (UK) vs. Birmingham, Alabama), and in other cases it won't (e.g. Boston, Lincolnshire vs. Boston). Bazonka (talk) 16:26, 28 December 2011 (UTC)

@LtPowers: "Disambiguation pages are bad. They represent our failure to get the reader to the content they want to read." Nonsense: if the name truly is ambiguous, it is not a "failure" to point out the ambiguity. We didn't make it ambiguous; the English language did. If you walk into any library in the English-speaking world and ask the research librarian for help finding information about Mercury, without any other context, the librarian will ask you which "mercury" you mean. That is not a failure on the part of the librarian, it is correctly recognizing an ambiguity that is inherent in the question. In fact, a librarian who didn't ask you which Mercury you meant would be failing to do their job. --R'n'B (call me Russ) 15:36, 28 December 2011 (UTC)

If our goal is to get a reader to the content they want to read as quickly as possible, any delay represents a failure to reach that goal. In some cases, such as the one you note, it may be unavoidable. To extend your analogy, if you ask the research librarian for help finding information about Englebert Humperdinck, should the librarian say "Which one?" or should the librarian say "Do you mean the composer?" And then consider that the answer might be different since we are not a research librarian. Powers ^T 15:13, 29 December 2011 (UTC)

I wonder if a direction to go might be the "subtitle" idea (as mentioned at WT:AT#Subtitles in smaller type, though it's been proposed before and quite well received). If the parentheticals were made less prominent, they'd seem to be less of a downside, and people would be happier about including them even if the article is the primary topic (except perhaps in extreme cases like Libya). So we could in some way detach the question of utility as regards primary topics (we want to get people to their sought article as quickly as possible) from the question of how the articles are titled. I mean, we could have Engelbert Humperdinck (composer) and Engelbert Humperdinck (singer), treat all links to plain EH as requiring disambiguation, but also make one of them the primary topic so that people don't get sidetracked to a dab page. Then we can fairly easily switch primary topics if relative traffic volume changes. (This solution doesn't require the smaller subtitles, but I think they would make it more palatable.)--Kotniski (talk) 15:06, 28 December 2011 (UTC)

How would you decide which is the primary topic? Unless it's obvious, this may well be a contentious issue, and so it may often be better to have no primary topic. Bazonka (talk) 16:26, 28 December 2011 (UTC)

It will be less contentious using this solution than it is now, since primary-topic-ness wouldn't affect the article titles, only which article a reader happened to get to first. We would decide on a primary topic in much the same way as we do now, but since it would be reduced to a practical, navigation-assisting choice, and one that could be easily changed at any time, we could probably rely more happily just on page hits and little else. (And in cases where there really is an obvious primary topic that all sane editors will recognize as such, we could still dispense with the disambiguator.)--Kotniski (talk) 12:05, 29 December 2011 (UTC)