Misplaced Pages talk:Disambiguation

This is an old revision of this page, as edited by BD2412 (talk | contribs) at 21:50, 22 April 2013 (→Proposal that designation of a primary topic should be inherited by subtopics.: new section). The present address (URL) is a permanent link to this revision, which may differ significantly from the current revision.

Revision as of 21:50, 22 April 2013 by BD2412 (talk | contribs) (→Proposal that designation of a primary topic should be inherited by subtopics.: new section)(diff) ← Previous revision | Latest revision (diff) | Newer revision → (diff)

Skip to table of contents

This is the talk page for discussing improvements to the Disambiguation page.

Put new text under old text. Click here to start a new topic.
New to Misplaced Pages? Welcome! Learn to edit; get help.

Shortcuts

Archives: Index, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57Auto-archiving period: 21 days

Threads older than 3 weeks may be archived by MiszaBot II.

Disambiguation

This page is within the scope of WikiProject Disambiguation, an attempt to structure and organize all disambiguation pages on Misplaced Pages. If you wish to help, you can edit the page attached to this talk page, or visit the project page, where you can join the project or contribute to the discussion.DisambiguationWikipedia:WikiProject DisambiguationTemplate:WikiProject DisambiguationDisambiguation

Establishing a WP:PRIMARY within the brackets

We keep seeing at WP:RM move proposals seeking to shorten (reduce the precision) of parenthetical disambiguation, like this randomly generated example:

Let's assume for the sake of argument that the American is actually much more notable than the Brit (which he probably is). Is the current text of WP:DISAMBIGUATION clear about whether we do/don't establish who is the more notable in brackets. My understanding, given that no-one will ever type "Mark Smith (racing driver)" is that there is no benefit to such moves, and it reduces utility of both Google results and the R/H search box autocomplete. Can someone comment on this please? In ictu oculi (talk) 06:08, 25 March 2013 (UTC)

I don't think that titling decisions should entertain discussions of relative notability. We have trouble enough defining sufficient WP:N-notability without wanting to pretend that we can quantitate degrees of notability. Deferring this poor measure to ghits is an abrogation of our aim to create reference knowledge accessible to all people. Ghits are biased and non-representative of our readership.

Titles should be precise enough that no reader is ever astonished on arriving at the page. Where there is ambiguity, each should be disambiguated to equal precision, not all but the most notable.

My understanding is that titles are not written to facilitate the search function. The search function is quite capable of looking beyond the title, and searching the article text. My view of titles is that the title, as it appears elsewhere, embedded in another article or written externally as a link to a particular page on Misplaced Pages, should tell the reader what the article is about.

I therefore thnk that WP:PRIMARY should be reduced in impact, and that article titles should err on the side of being too precise. I don't think that WP:PRIMARY should have any applicability within parentheses in a title. --SmokeyJoe (talk) 07:05, 25 March 2013 (UTC)

I agree. Once parentheical disambiguation is necessary, the goal should be to disambiguate adequately between all uses of the term, though there may be difficult cases where different criteria are used for disambiguation. I am also not too happy about London (electoral district). --Boson (talk) 08:35, 25 March 2013 (UTC)

Not the first time has come up on this talk page (here's a similar discussion that springs to mind), but I can't recall a concrete consensus being reached. Personally, I agree with you that if an article is not the primary topic it should have a unique disambiguation and I'd support adding something of this effect to the guideline. Jenks24 (talk) 10:10, 25 March 2013 (UTC)
Something to this effect is present in some of the naming guidelines. IMO, that's where it should be; disambiguation will work once the titles are different, and the choice of which differentiators to use is best left to the content projects and their naming conventions. -- JHunterJ (talk) 12:58, 25 March 2013 (UTC)

I personally think that the reduction in disambiguation is poor, since it will result in even more contentious renaming battles than already occur, since as of right now, people already make a big fuss over which article should live at the undisambiguated title. If we reduce disambiguation for disambiguated titles as well, these battles will be ever more frequent, and result in more move wars. If it isn't the primary topic, we should treat all other uses as needing sufficient disambiguation to distinguish each from each other. The choice of disambiguatory term is also rather arbitrary for Misplaced Pages. In the example provided, it uses "racing driver", and "British" or "American", but someone else may use "United States" and "United Kingdom", or "US" and "UK", and "racing driver" may be replaced by "driver" or "pilot" or may use the current series/discipline they're in "F1 pilot" or "rally driver", etc. -- 65.92.180.137 (talk) 04:41, 29 March 2013 (UTC)

Well it seems there's not absolute agreement here. I understand JHunterJ's comment to say that the question of whether to establish a primary within brackets should be left to projects. I have noticed in fact that (song) seems to be a particular category where project participants seem determined to argue over who gets the prime "(song)" spot, leaving (Walter Mitty song) (John Smith song) (Joe Bloggs song) as 2,3,4 runner ups. Sort of an X-factor contest for the "(song)" spot. But is there really value to the project as a whole in having some projects do this? A lot of these projects only have 2-3 active editors and when 1 leaves, the view changes and renames happen in the other direction. In ictu oculi (talk) 07:35, 1 April 2013 (UTC)
If the project as a whole thinks that all of the content-area projects should standardize on this, I think it's a discussion for WP:VPP or WT:NC. Formulating overriding rules within the dab project for their naming conventions is IMO going to cause more arguments, not fewer. And I have no objection to that potential broader consensus. -- JHunterJ (talk) 11:08, 1 April 2013 (UTC)

URLs on category pages

There is a discussion at Misplaced Pages talk:Categorization#Reference resources about whether WP:External links should be permitted in the descriptions on category pages. Since I propose that cat-page descriptions be handled similar to dab descriptions (e.g., useful description, but no reliable sources and no external links), perhaps some people here will be interested. If you have an opinion, please comment there. WhatamIdoing (talk) 21:50, 28 March 2013 (UTC)

Get rid of WP:PRIMARYTOPIC

Wondering around RMs, I'm only becoming more convinced that WP:PRIMARYTOPIC does far more bad than good. It is very poorly explained, with people straight off making opposite readings. Probably this is because it is not well defined. It is a huge invitation to assert a majority bias. It draws people to study ghits, which are poorly aligned with the goals of the project. It creates confusion and controversy in what should be the simplest of cases. Better, it would be, to just disambiguate when there potential for ambiguity, on the part of any reasonable group of readers. I think it should be got rid of as unhelpful. --SmokeyJoe (talk) 13:27, 8 April 2013 (UTC)

I'm uncertain that I understand what you're proposing. As a couple of examples, do you advocate that Dog (disambiguation) be relocated to Dog and that Barack Obama become a disambiguation page listing Barack Obama II and Barack Obama, Sr.? —David Levy 13:38, 8 April 2013 (UTC)

I have to say, if this were a formal proposal, it would have my full support. Saying that WP:PRIMARY topic is "unhelpful" is putting it mildly: the amount of time Wikipedians waste on discussing which topic is more "primary" than the other is simply mind-boggling, and the results of such discussions often remain contentious anyway. Yet all that's often needed in such discussions is a little common sense. I have difficulty imagining someone coming up with serious reasons to turn something like "Barack Obama" into a dab page, but in most cases it's not so clear-cut. At any rate, even if something like "Dog" becomes a disambiguation page, the harm would be rather minimal, and if doing so saves up editors' time otherwise wasted on counting ghits, popularity votes, and abstract arguments about "educational value" or "enduring notability", I'd say the net positive is well worth it.—Ëzhiki (Igels Hérissonovich Ïzhakoff-Amursky) • (yo?); April 8, 2013; 14:05 (UTC)

I don't quite understand your position. We can't stop recognizing the existence of primary topics without making Barack Obama a disambiguation page. It certainly is true that many cases are less clear-cut than that one is, and it's reasonable to opine that we should adjust the threshold to draw such a distinction only in the most clear-cut cases, but abandoning the entire concept behind WP:PRIMARYTOPIC would mean never drawing the distinction; a disambiguation page would occupy the base title for any term requiring disambiguation. —David Levy 14:29, 8 April 2013 (UTC)

David, to clarify, I'm not for stopping to recognize the existence of primary topics altogether; I'm for removing of a bunch of artificial criteria which are vaguely defined, interpreted by different editors differently, and ultimately lead to wasting enormous amounts of time which could be put to better use elsewhere. Note that under any arrangement there is nothing preventing editors from continuing to discuss potential moves, but getting rid of the artificial nonsense PRIMARYTOPIC currently endorses would simplify the RM process significantly. If there is substantial disagreement as to which topic is primary, keep the dab page at the primary location. If there is a clear consensus that some topic is indeed primary, then so be it. It can't be simpler than that (and doesn't have to be) and it doesn't lead to disambiguation pages occupying all possible "primary" nodes either. Assessing whether there is a consensus or not is a lot simpler than assessing a bunch of esoteric criteria often created and enforced by people who do little productive work here beyond endlessly arguing policy and coming up with new creative way to regulate everything. If such people can be shut off, everyone else would benefit.—Ëzhiki (Igels Hérissonovich Ïzhakoff-Amursky) • (yo?); April 8, 2013; 14:56 (UTC)

David, We can stop recognizing the existence of primary topics without making Barack Obama a disambiguation page. There is no policy requiring making Barack Obama a disambiguation page except for the existence of the guideline section with the shortcut PRIMARYTOPIC.

I want to abandon the superficial, not-proven-useful section in the guideline, not abandon an entire underlying concept. But if the concept is to remain explicit, it needs to be better than what’s there now. Removing WP: PRIMARYTOPIC does not imply that we are writing into policy that everything that was once written under WP: PRIMARYTOPIC was false. --SmokeyJoe (talk) 01:16, 9 April 2013 (UTC)

David, We can stop recognizing the existence of primary topics without making Barack Obama a disambiguation page.

No, we can't. In such a scenario, there would be no reason to treat Barack Obama, Sr. differently than we treat Barack Obama II.

There is no policy requiring making Barack Obama a disambiguation page except for the existence of the guideline section with the shortcut PRIMARYTOPIC.

Let's set aside our polices and guidelines for a moment. Without the concept of "primary topics" (whether written down or not), there would be no justification for assigning the base title to a particular Barack Obama's article. The very act of doing so is an acknowledgement that said article's topic is primary.

I want to abandon the superficial, not-proven-useful section in the guideline, not abandon an entire underlying concept.

Oh. Then why did you just advocate that we "stop recognizing the existence of primary topics"?

But if the concept is to remain explicit, it needs to be better than what’s there now.

I noted above that the guideline can be modified. —David Levy 03:52, 9 April 2013 (UTC)

Hi David. On Obama: I mean that we don't need to be this explicit in recognising the primary topic to continue with current practice. Other than that, I'd like to leave Obama as not a typical page.

Then why did you just advocate that we "stop recognizing the existence of primary topics"? Trying again then... I want to stop recognising the something under the term "Primary topic". I would prefer to speak in terms a topic that overshadows all others in terms of notability/recognisability, especially when the other topics derive from the first. A problem with "primary topic" is that it sounds like it is something that is very well defined, as in there are not shades of primariness.

Modifying the text: Yes, it could be modified. Roughly speaking, do you think that is needs (1) wholesale rewriting; (2) work; (3) tweaking only, it's basically OK. --SmokeyJoe (talk) 04:34, 9 April 2013 (UTC)

Trying again then... I want to stop recognising the something under the term "Primary topic".

Thanks for clarifying. I was referring to the underlying concept, not the term used to describe it (which certainly can be changed if we come up with something better).

Modifying the text: Yes, it could be modified. Roughly speaking, do you think that is needs (1) wholesale rewriting; (2) work; (3) tweaking only, it's basically OK.

My impression falls in the range of 3, but I'm open to the possibility that more extensive change is called for (depending on what problems exist and what solutions are offered). —David Levy 09:34, 9 April 2013 (UTC)

(ec) OTOH, I've only been more convinced of the need for WP:PRIMARYTOPIC, after seeing some of the poor criteria editors sometimes use in RMs. And as David Levy notes, doing away with it leads to worse absurdities that you're finding with it, so it does more good than bad. -- JHunterJ (talk) 14:30, 8 April 2013 (UTC)

JHunterJ, can you expand on how cutting this section (with the option of starting again with different words) will (definitely) lead to worse absurdities? I don't think you can, because guidelines don't define practice, they reflect it. Also note that I am not asserting absurdities. On the whole, we make reasonable decisions despite the confusion injected by WP:PRIMARYTOPIC. My point is that for editors looking for guidance, they are being poorly served. I maintain that it does little good, that correct decisions make that are consistent with PRIMARYTOPIC were going to be made that way anyway (based on sources, etc). --SmokeyJoe (talk) 01:22, 9 April 2013 (UTC)

No need. This section will simply fill with sound and fury, signify nothing, and then be archived. If your point is that editors are being poorly served, there may be improvements that stand a snowball's chance of finding consensus, but leading with "Get rid of WP:PRIMARYTOPIC" ensures this won't be the discussion that finds them. -- JHunterJ (talk) 01:33, 9 April 2013 (UTC)

JHunterJ, I have been looking at this for a long time. I really think that removing the section will be a dramatic improvement without bad consequences. At worst, people who like to bluelink the allcaps oneword may have to think their rationale out more clearly. You say "it does more good than bad". You normally say sensible things. I would really like to understand why you say this. If you think I am just pot stirring or game playing or otherwise making noise without point, please respond on my talk page. --SmokeyJoe (talk) 01:51, 9 April 2013 (UTC)

Our guidelines are intended to document how we do things. If the one in question is poorly written, it should be fixed, not eliminated.

The logic that editors will continue to behave sensibly without such documentation can be applied to the entire Manual of Style or even every Misplaced Pages policy and guideline. Editors who regard them as needless bureaucracy (and wish to delete them en masse) actually have done so. You're presenting essentially the same argument on a smaller scale. —David Levy 03:52, 9 April 2013 (UTC)

I think the MOS is over-written. It probably happened because there are so many pages of them. On Misplaced Pages Policies and guidelines, I think they are generally well balanced on depth of detail. Am I on a limb here?

I think the one in question is built on such a poor word choice, it would be easier to start again. I guess that you would like to see a replacement first. --SmokeyJoe (talk) 04:41, 9 April 2013 (UTC)

In my view, nothing else makes sense.

Our guidelines are descriptive, not prescriptive. We don't do the things described at WP:PRIMARYTOPIC because they're written there; they're written there because we decided to do them. If we decide to do something else instead, we can adjust/replace WP:PRIMARYTOPIC accordingly. Until such time, the documentation should remain. Removing it wouldn't magically change how we do things or solve any problems related thereto. It would amount to nothing more than obfuscation. —David Levy 09:34, 9 April 2013 (UTC)

I would be more inclined to go in the opposite direction and say that for pages like Battery and Mercury and Springfield, we should just pick a "most likely" meaning, put that topic at that title, and move the disambig to the "Foo (disambiguation)" title. bd2412 T 17:16, 8 April 2013 (UTC)

Could you clarify your point, BD2412? Mercury, for instance, seems an excellent example of how impossible it would be to choose which page could stand without a disambiguator (that is, which one would be the primary topic). I don't even know how you'd "prove" whether the planet or the element was the most likely intended destination. Cynwolfe (talk) 17:32, 8 April 2013 (UTC)

There I would contend that a chemical element is more fundamental a topic than a planet, which is in turn a more fundamental topic than a god or a record company. This disposes of popularity questions, and focuses on which topic is primary from a standpoint of importance to an encyclopedia. However, I would still have incoming links to the page piped through Mercury (element), just so we could check the unpiped links to be sure they intended that meaning of Mercury. bd2412 T 20:34, 8 April 2013 (UTC)

Both the element and the planet names derive from Mercury (mythology). Ususally, the inclusion of the word “fundamental” indicates that the statement is false due to excessive simplification --SmokeyJoe (talk) 00:12, 9 April 2013 (UTC)

That's exactly the wrong approach, in my view. Our goal is to provide access to the encyclopedia articles sought, not to favor topics that we consider more important. The latter doesn't serve readers, and it's the last thing that will prevent arguments among editors.

Also, requiring the use of redirects in running prose would be a never-ending headache. —David Levy 03:52, 9 April 2013 (UTC)

"Mercury" is a textbook example of a term with no primary meaning. —David Levy 03:52, 9 April 2013 (UTC)

We can't get rid of WP:PRIMARYTOPIC, for as others have noted that would mean making Barack Obama a dab page, but we can surely improve it.

The problem with WP:PRIMARYTOPIC probably starts with the ambiguity of the meaning of the word "primary". For example, just above, BD2412 interprets the chemical Mercury to be "primary" over the planet, because the chemical is "more fundamental". Now, "fundamental" is a synonym for "primary" and an appropriate substitute in some contexts, but I've never before encountered it being used in primary topic discussions, and I thought I'd seen everything. "Most fundamental" is NOT what "primary" means in the context of "Primary Topic"!

What "primary" means in "PRIMARY TOPIC" is "most commonly used". So, what if we renamed Primary Topic to something like, "MOST COMMON USE TOPIC"?

Regardless of what we call it, to discuss it, it's critical to understand why we have it. The reason is, and has always been, to serve our readers best. The quintessential example, perhaps, is Paris. Even though there are many uses of "Paris", most people searching for "Paris" would be looking for the city in France, so we put that topic at Paris. Thus we have the traditional definition of "primary topic" of a given term:

it is highly likely—much more likely than any other topic, and more likely than all the other topics combined—to be the topic sought when a reader searches for that term.

The relatively recent addition of the second criterion to the definition, long-term significance, has made things worse, in my opinion. Not only is this criterion far more ambiguous/subjective than the traditional "likely" one is, it often suggests a different answer, thus creating, rather than resolving, conflicts. What kind of guideline is that?

I suggest that primary topic arguing has gotten much worse since the addition of long-term significance, and we could go a long way towards improving the situation by removing it.

As to arguments about "popularity" and counting ghits, it wasn't all that bad before "long-term significance" was added. Without it, unless popularity including counting ghits makes it clear that one of the uses of a given term is "much more likely than any other topic, and more likely than all the other topics" to be the one being sought, there is no primary topic, period. Hence Mercury is, and should be, a dab page. --B2C 21:19, 8 April 2013 (UTC)

Also concerning: "educational value." How and, more importantly, why are we supposed to define it? "volume of coverage in reliable sources" would be a reasonable alternative which is quantifiable for the purposes of debate and should be agreeable to everybody. For the record I don't agree with throwing out primary topic. Marcus Qwertyus (talk) 21:42, 8 April 2013 (UTC)

And if you can simply define some type of criteria, how do you objectively determine it? Vegaswikian (talk) 21:58, 8 April 2013 (UTC)

Some good responses.

The best outcomes I see with WP:PRIMARYTOPIC, it is redundant to WP:COMMONNAME, with no other topic challenging for the same COMMONNAME.

I do not have a problem with Dog. Dog could be at Dog (Canis lupus familiaris), but "Dog" is the clear COMMONNAME with no other topic has a serious, even time or place, claim to be at Dog. If the article were at Dog (Canis lupus familiaris), Dog should redirect straight to it, without any need for a PRIMARYTOPIC argument.

Barack Obama is an extreme case. A hot, sensitive, promotionally written article that I would not touch. In general, I think biographies should be more precisely titled, whether by parenthentical birth-death and/or main occupation. If this were a good idea, I'd begin with long dead people, then presidents of small countries, and would not propose a rename of the most significant person to the most significant group of editors at this time. I don't think it would even lead to a helpful debate.

Mercury is an excellent example (as I guess there are few emotive interests) of a title where you can imagine separate people in different times and places arguing that their use (bias) is "PRIMARY".

The sort of article RM that have motivated me are human commercial products, such as songs and films. Including "Big".

At the very least, I think the shortcut "WP:PRIMARYTOPIC" should be abandoned as historically confusing, and if replaced (NB we do not NEED ALLCAPS ONEWORDS for every concept), it should be something much more tightly defined, so that we can agree on what it means, and ordinary editors can interpret unambiguously. However, I still feel that every appropriate use of PRIMARYTOPIC (the relative few of them) is redundant to a better explanation for the best title. --SmokeyJoe (talk) 22:52, 8 April 2013 (UTC)

I do not have a problem with Dog. Dog could be at Dog (Canis lupus familiaris), but "Dog" is the clear COMMONNAME with no other topic has a serious, even time or place, claim to be at Dog.

"Dog" is the common name of multiple subjects. One of them is much more well known than the others are, which makes it the primary topic.

If the article were at Dog (Canis lupus familiaris), Dog should redirect straight to it, without any need for a PRIMARYTOPIC argument.

We don't redirect "Foo" to "Foo (disambiguation term)". If you believe that we should, I strongly disagree.

In general, I think biographies should be more precisely titled, whether by parenthentical birth-death and/or main occupation.

Why? —David Levy 03:52, 9 April 2013 (UTC)

""Dog" is the common name of multiple subjects". Really? For example. Are we working with different interpretations of WP:COMMONNAME, specifically that a COMMONNAME has prevalence in reliable English-language sources? Vernacular is excluded. To my reading, every other usage on the disambiguation page involves another word or some contextualisation before use.

We don't redirect "Foo" to "Foo (disambiguation term)"? I don't see a problem in principle, but I haven't deeply considered the question. I think we are agreed to leave Dog at Dog anyway.

"In general, I think biographies should be more precisely titled, whether by parenthetical birth-death and/or main occupation.

Why?" A: Because biographies are a major growth area for the project, and recognisable names are often not unique identifiers. We therefore have to disambiguate many of them somehow. Middle names, sometimes, but middle names, if they exist, may not help recognisability. Parenthetical birth and death dates are likely to help recognisability because the searcher probably has some idea of the historic period they belong to. The searcher probably also has some idea of the occupation. I think this should be done, it is often done, and I think it should be done more. See William Henry for example. In the listing, it is natural to disambiguate and sort by birth and death years as well as occupation. I think this would work well in titling. At the moment we sometimes disambiguate biographies oddly, such as William Henry (brother of Patrick Henry). My preference would be to recommend a systematic method, allowing the article writer freedom to accept or reject the recommendations. The exceptions to usual disambiguation are: (1) living people, and (2) supernotable people (where WP:PRIMARYTOPIC would apply). (1) doesn't bother me, I'm happy to see more precise disambiguation happen later. (2) feels wrong to me. We have a arbitrary-seeming line of supernotability, and assume that no one needs precision for the supernotable. This has some downsides, including providing poor examples of titling for non-regular editors. --SmokeyJoe (talk) 05:13, 9 April 2013 (UTC)

"'Dog' is the common name of multiple subjects". Really? For example.

"In general, I think biographies should be more precisely titled, whether by parenthetical birth-death and/or main occupation. Why?" A: Because biographies are a major growth area for the project, and recognisable names are often not unique identifiers. We therefore have to disambiguate many of them somehow.

Perhaps I misunderstood. I interpreted your statement to mean that we should routinely append such information to biographical articles' titles (even when disambiguation isn't needed). —David Levy 09:34, 9 April 2013 (UTC)

I do think that we should routinely disambiguate articles even if not strictly needed, for ease of recognition and consistency, as a recommendation. I think that the precision used should be consistent across biographries, and should not be inversely proportional to importance. For example, William Henry (brother of Patrick Henry) should be moved to William Henry (1734–1785, politician) (subject to an ongoing format discussion at Wikipedia_talk:Naming_conventions_(people)#RFC-birth_date_format_conformity_when_used_to_disambiguate), even if he is the only notable William brother of a Patrick Henry. I recommend moving George Washington to George Washington (1732–1799, US President) and George Washington (Washington pioneer) to George Washington (1817–1905, Washington pioneer). Consistency is nice, but I have in mind good service for readers such as the children in Borneo who have difficult access and no cultural association with these people. I think that titles should feed an indexing system, and that 50 characters is an appropriate title length. --SmokeyJoe (talk) 11:59, 9 April 2013 (UTC)

You certainly are entitled to your opinion. You're aware, I assume, that the above is far removed from anything for which consensus has been established. —David Levy 13:05, 9 April 2013 (UTC)

Moderately far, yes. But I do like the idea. However, it is tangential to the main goal here of clarifying PrimaryTopic. Cutting it would clarify WP:DAB and dissuade non RM regulars from clumsily quoting it, but it seems you are going to require me to work out what exactly the underlying concept is. Ok. My opinion is directly relevant in declaring an opposition to an effect of WP:PRIMARYTOPIC, which is reduced precision for more popular subjects. --SmokeyJoe (talk) 13:35, 9 April 2013 (UTC)

Fair enough. I don't agree that our titles should be utilized in the manner that you've described, but I understand the underlying concern. —David Levy 13:45, 9 April 2013 (UTC)

I would like to see rejected arguments based on likely searches. Arguments of likelihood are arguments from majority bias.

Regarding the quintessential Paris, looking though the disambiguation, I see that no other topic has a reasonable COMMONNAME argument to be at just "Paris". The second best would be Paris (mythology), and for this subject I note that no one refers to this Paris without the subject being first contextualised. So Paris can be justified from policy without reference to PRIMARYTOPIC. Although on the other hand, it would hurt no one if Paris were at Paris, France, and it would help people who do no immediately know Paris, foreign children, for example. I remember someone believing Paris to be the capital of Europe. --SmokeyJoe (talk) 23:01, 8 April 2013 (UTC)

New York is another example for what is the primary topic. There are at least 3 articles that could make the claim, but how do you prove that the city or the state is the primary topic? Vegaswikian (talk) 23:16, 8 April 2013 (UTC)

If I had my druthers, we'd rename Georgia (country) to the native Sakartvelo and be done with that disambig. However, looking to names for which the "primary topic" could be more subject to dispute, what about Apple and Grape? Orange and Lime are disambigs. What about Europe and Asia and Florida and Kansas and Chicago? bd2412 T 23:38, 8 April 2013 (UTC)

"New York State" versus "New York City". That one could have gone either way. The original colony was, what is now, the city, certainly not the current state according to defined boundaries. I'm sure the people of the city refer to as "New York", and the upstate people refer to their state as "New York". Normally, in my experience, it is pre-contextualised as to whether the conversation is about cities or states. I don't think there is anything in the current WP:PRIMARYTOPIC that helps decide the question, and so yet again WP:PRIMARYTOPIC is serving no useful purpose. I think the guiding principle for New York should be the choice made by the original author, Trimalchio (talk | contribs) at 01:09, 1 October 2001, subject to a possible later clear (not rough) consensus to change. NB. This is actual practice.

On Sakartvelo, WP:PRIMARYTOPIC topic is useless due to not advising on whose perspective to consider it from. This one looks like a worthy RM discussion. I see Talk:Georgia_(country)#New_move_proposal and a completely useless reference to "primary topic", as if the user thinks it means something.

As I said above, we are not suffering disputes over Apples and Grapes, Oranges and Limes, and so PRIMARYTOPIC is not existing to help them. The problems are worst with recent human products. Europe, Asia, Florida, Kansas and Chicago? What about them? None were named with reference to PRIMARYTOPIC. --SmokeyJoe (talk) 00:04, 9 April 2013 (UTC)

SmokeyJoe, "Arguments of likelihood are arguments from majority bias." Indeed. Since in this case (choosing article titles) the goal is serving the majority best, arguments from majority bias are appropriate.

"Although on the other hand, it would hurt no one if Paris were at Paris, France". Very few titles could hurt anyone. That's rarely a consideration in deciding titles. The problem is having multiple choices all of which could be reasonably supported... how do we decide which one to use? Yes, yes, I know, "by consensus", but I mean, how does each person decide which one to use? Based on what criteria, exactly? If you and I apply the same criteria to the same situation and come up with different titles, we're not looking at the criteria in the same way - the criteria is vague, and we have no objective means to resolve such differences, except to improve clarity in the criteria.

This is why I advocate more and better criteria. It would not be okay if Paris, France met our criteria just as well as Paris, for that would mean anyone could reasonably call for a move periodically, without end. Thankfully, that's not the case: because of common name, primary topic and concision, we clearly prefer Paris to Paris, France, and that's a good thing because it means this case, and countless others, are settled (at least as settled as anything can be on WP).

If we loosen the criteria, like by getting rid of primary topic, we would introduce more ambiguity and thus more conflict into the title decision process. That would be a bad thing. In contrast, by removing the long-term consideration criterion from primary topic, we would be removing ambiguity and thus reducing conflict... a good thing. --B2C 00:09, 9 April 2013 (UTC)

B2C. Is the most direct service of the majority the goal of choosing article titles? I think not. The majority, especially as measured by ghits, are already well served. We should aim instead to serve all humanity. To serve broadly means making special efforts to serve those disfavoured by the prevailing bias. We should think of readers who do not know how to search. We should think of readers who have only a print version of selected pages.

When speaking of Paris, you seem to me to be too heavy on the need for criteria. This criteria-based decision making feeds into expert-pleb style governance, disempowering the ordinary editor. Disempowering the ordinary editor is to stifle the future growth of the encyclopedia. Instead, we should trust the editors who build to be doing things sensibly. On the question of Paris, the page naming decision by Zundark at 11:45, 6 November 2001 should stand until there is a clear consensus that something should be changed.

“If we loosen the criteria, like by getting rid of primary topic”. That’s where you are mistaken. WP:PRIMARYTOPIC is not serving by providing defined criteria. Define it better and it may work better. However, I think it is in a rut.

I disagree with removing the concept underlying long-term consideration criterion. An encyclopedia is best considered as a historical document, things are best written from a historical perspective, and titles should reflect the perspective of the content. “Long term consideration” is a variation on WP:NOTNEWS and Misplaced Pages:Recentism.

“Anyone could reasonably call for a move periodically”. Certainly not. A consensus is required for a move. The longer the history of proposed moved, the clear the consensus required. “Periodic” calls for moves quickly becomes received as disruptive editing. --SmokeyJoe (talk) 01:53, 9 April 2013 (UTC)

How has such a simple idea become so complicated? If we want to get rid of Primary Topic why can't we? Mercury the element becomes Mercury (element), Mercury the god becomes Mercury (god), Mercury the planet becomes Mercury (planet), and Barrack Obama becomes Barrack Obama (44th president of the United States). Anyone who still doesn't know what they're looking for, even with a helpful drop down menu on the search bar, can go to a dab page.--Ykraps (talk) 00:08, 9 April 2013 (UTC)

The simple answer is that some editors believe that disambiguation is bad and predisambiguation is worst. Then you have those who will argue that the first article has a god given right to the main name space. Vegaswikian (talk) 00:12, 9 April 2013 (UTC)

If we get rid of Primary topic, we get rid of those arguments.--Ykraps (talk) 00:20, 9 April 2013 (UTC)

If that is due to improved acceptance of disambiguation not being bad, then maybe getting rid of the primary topic guideline is also not bad. But would editors accept this? Is it possible to get a consensus in support? Vegaswikian (talk) 00:26, 9 April 2013 (UTC)

I think most editors find PT an unnecessary distraction. Everybody seems content to go to the index if they want to look something up in a book. Nobody starts crying because it hasn't magically fallen open at the page they wanted.--Ykraps (talk) 00:35, 9 April 2013 (UTC)

Ykraps: Do you seriously advocate that we move Barrack Obama to Barrack Obama (44th President of the United States)? If so, why? —David Levy 03:52, 9 April 2013 (UTC)

So we can get rid of PT. Why shouldn't we move him?--Ykraps (talk) 06:18, 9 April 2013 (UTC)

We don't have to move Barrack to get rid of PT. --SmokeyJoe (talk) 07:12, 9 April 2013 (UTC)

Why? Where would entering Barack Obama in the search box take you?--Ykraps (talk) 07:33, 9 April 2013 (UTC)

Barack Obama is a very stable title regardless of the guideline section being discussed here. The connection between PT and BO is very weak, non-existent even. In practice, many people are well used to ignoring the actual text of PT. --SmokeyJoe (talk) 08:08, 9 April 2013 (UTC)

Barack Obama is a very stable title regardless of the guideline section being discussed here.

I agree that eliminating WP:PRIMARYTOPIC wouldn't result in a realistic possibility of the Barack Obama article being renamed. Our guidelines' text is descriptive (not prescriptive), and removing it wouldn't magically cause a convention that it describes to disappear. It would merely increase the difficulty experienced by editors seeking to learn about or explain said convention.

The connection between PT and BO is very weak, non-existent even.

On the contrary, WP:PRIMARYTOPIC documents ~~the~~ a principle on which the article's title is based. —David Levy 09:34, 9 April 2013 (UTC)

I know little of the Principles of PRIMARYTOPIC, but I think that's revisionism, and that it is much more likely that the original author of the article took the title from the title predominantly used in the sources at the time.

I think Meelar drew the information from his sources, which is what we should do.

I think it very unlikely that Meelar considered the likelihood of searches for the senator against searches for his father or other Obamas. --SmokeyJoe (talk) 11:25, 9 April 2013 (UTC)

I don't mean that the article's title was determined by consulting the section and applying its advice. I'm referring to the core principle that when a common name shared by multiple subjects is associated with one much more strongly than it is with all others combined, it's used as the title of that subject's article.

Had Meelar titled the article something other than "Barack Obama", it would have been moved to that title. Alternatively, if Misplaced Pages's standard procedure were to always use titular disambiguation in such cases, it would have been moved from "Barack Obama" to something else.

Of course, I should have written "a principle" (not the principle). The aforementioned WP:COMMONNAME obviously plays a major role. —David Levy 13:05, 9 April 2013 (UTC)

The descriptive not prescriptive mantra is not strictly true, unless using twisted definitions such as "policy describes ideal practice", where "ideal" means "what I would like you to do". In reality, much policy contains prescription mixed in, in this this case certainly. --SmokeyJoe (talk) 11:25, 9 April 2013 (UTC)

I don't think that you're fully understanding what's meant by "descriptive, not prescriptive".

Certainly, Misplaced Pages's guidelines are intended to advise users of how they should go about editing the encyclopedia. No one disputes that. We mean that the guidelines themselves don't dictate anything; they aren't like laws that take effect when enacted. They're intended to describe the practices on which the community has settled via consensus. We want people to do what's been written, but not because it's been written. It's been written because it's what we've decided to do. —David Levy 13:05, 9 April 2013 (UTC)

We don't have to move Barrack to get rid of PT.

That depends on what Ykraps means by "PT" — the guideline section or the underlying concept. My impression is that Ykraps means the latter. —David Levy 09:34, 9 April 2013 (UTC)

Why shouldn't we move him?

If you mean that the title "Barack Obama" no longer would lead to his article, that would be extremely unhelpful to readers. If you mean that the title "Barack Obama" would redirect to his article, such a change would provide no benefit.

Regardless, if titular disambiguation were called for, "Barrack Obama (44th President of the United States)" is excessively precise and inferior to various alternatives. —David Levy 09:34, 9 April 2013 (UTC)

We don't have to move Barrack to get rid of PT.

That depends on what Ykraps means by "PT" — the guideline section or the underlying concept. My impression is that Ykraps means the latter. —David Levy 09:34, 9 April 2013 (UTC)

Why shouldn't we move him?

Barack Obama is a highly distinctive name. What about the many people named John Adams, Thomas Jefferson, Andrew Jackson, Benjamin Harrison, and Jimmy Carter? bd2412 T 12:25, 9 April 2013 (UTC)

What about them? —David Levy 13:05, 9 April 2013 (UTC)

Under this proposition, it seems that they would all be moved to disambiguated titles, with the undisambiguated title being a disambiguation page. I think that would be equally unhelpful. bd2412 T 13:18, 9 April 2013 (UTC)

Agreed. —David Levy 13:45, 9 April 2013 (UTC)

Having Barack Obama go to Barack Obama only makes it easy for people who are looking for the 44th president of the United States. If you are looking for another Barack Obama, then it’s actually quite difficult. Entering Barack Obama (disambiguation), a perfectly logical thing to enter if you are aware of Primary topic, takes you to a page that doesn’t exist, although from here you can find a link to Obama (surname); I don’t know why, to me that is a step backwards, I’ve already made it clear that I’m looking for a Barack Obama. Once at the Obama (surname) disambiguation page, if you’re lucky, you can find a link to the article you’re looking for. This of course assumes you have spelled Barack with one ‘r’ and not the two I initially thought.

Yes we should make it easy to find the most likely topic but not to the point where it makes finding the less likely, difficult.

As far as naming stability is concerned, moving Barack Obama to Barack Obama (44th president of the United States) is surely more stable in the long run. There may be a more notable Barack Obama in the future or even another US president by that name, but there will never be another Barack Obama (44th president of the United States). I appreciate that Barack Obama is an uncommon name in the English speaking world and so this is unlikely but what about another George Washington or Richard Nixon. There are already two ex US presidents called John Adams.

My main point is however, if we want to do away with Primary topic, it is perfectly possible to do so. And all the time saved from writing and rewriting (again and again) the guidelines, and bickering about which article should occupy that slot, can be spent actually improving the encyclopaedia’s content. Ykraps (talk) 15:59, 9 April 2013 (UTC)

We then have the problem of incoming links. Thousands of links have already been made to Barack Obama, with the expectation that clicking on that link will take the reader to the article on the 44th President, not some menu of other possibilities. The same can be said of John Adams, Thomas Jefferson, Jesus, Europe, Flowers, Sheep, Bananas, China, Pork, and so on. If we do away with a primary topic, all of those becomes disambiguation pages because of their many possible meanings. Will you be the one to fix the millions of links that are created, so that readers can be taken to the expected articles that now sit at the primary topic titles for these pages? bd2412 T 16:45, 9 April 2013 (UTC)

Having Barack Obama go to Barack Obama only makes it easy for people who are looking for the 44th president of the United States.

(an overwhelming majority)

If you are looking for another Barack Obama, then it’s actually quite difficult.

No, it isn't. Simply visit the Barack Obama page. The only other Misplaced Pages article about a person by that name is linked at the top.

Under this setup, most users arriving at the Barack Obama page have reached the desired article, while the rest must follow a link. If a disambiguation page were to occupy the "Barack Obama" title, everyone arriving there would need to follow a link to the desired article. How would this be beneficial?

Entering Barack Obama (disambiguation), a perfectly logical thing to enter if you are aware of Primary topic, takes you to a page that doesn’t exist, although from here you can find a link to Obama (surname); I don’t know why, to me that is a step backwards, I’ve already made it clear that I’m looking for a Barack Obama.

It's a keyword search.

If you wish to create Barack Obama (disambiguation) as a redirect to Barack Obama, I don't object.

That logic applies to topics in general, irrespective of whether their names are ambiguous. Do you advocate that we preemptively disambiguate every article's title (just in case)?

My main point is however, if we want to do away with Primary topic, it is perfectly possible to do so.

Has someone asserted otherwise? —David Levy 16:47, 9 April 2013 (UTC)

David, yes an overwhelming majority probably are searching for the US president and I refer you to my earlier comment, “Yes we should make it easy to find the most likely topic but not to the point where it makes finding the less likely, difficult”. The fact that PT exists prompts me to search for a disambiguation. Why would I search for something I’m not looking for? I appreciate your simple solution but I really needed that information before I started my search. Also I erroneously (apparently) assumed it was a keywords (plural) search and I considered ‘Barack’ to be a key word.

I am advocating that titles should be precise enough to unambiguously define the topical scope of the article (as per wp:precise), yes.--Ykraps (talk) 19:52, 9 April 2013 (UTC)

I wasn't accusing you of searching for something that you weren't looking for. I was explaining your approach's inconsistency with our actual practices and describing a simpler means of accomplishing the task at hand.

As noted at WP:TWODABS (part of the same guideline whose existence prompted you to seek a disambiguation page), "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—it is sufficient to use a hatnote on the primary topic article, pointing to the other article. (This means that readers looking for the second topic are spared the extra navigational step of going through the disambiguation page.)" —David Levy 21:36, 9 April 2013 (UTC)

Suppose we were to move Barack Obama to Barrack Obama (44th President of the United States); but suppose that some tremendous calamity caused the United States to dissolve, but a century or two on, a group of South American countries decided to form a national union, and also called this new country, "the United States of America". Suppose, further, then two centuries later, this nation were to elect its 44th President, and his name happened to be "Barack Obama". Now we have two people who could be identified as Barrack Obama (44th President of the United States). Do we need to preemptively disambiguate the current one to Barrack Obama (44th President of the United States, North America)? How about Barrack Obama (44th President of the United States, born 1961)? Either name could still yield a scenario where the title is ambiguous - if for example the new United States were formed in North America instead of South America; or if some future order of man were to initiate a new date for counting years, so that thousands of years in the future there is a new 1961, and a new Barack Obama is born who is destined to become the 44th President of a new United States. In fact, I propose that we initiate a new system of counting years right now: Before Misplaced Pages and After Misplaced Pages, with it currently being the year 12 AW. Then we can move Barack Obama to Barrack Obama (44th President of the United States, born 40 BW). bd2412 T 17:26, 9 April 2013 (UTC)

Bd2412, I appreciate what you say about incoming links, a problem indeed. With regards to your second concern, I think we can safely leave that for our ancestors to worry about; or are you just using this opportunity to show everyone what a great wit you are?--Ykraps (talk) 19:52, 9 April 2013 (UTC)

I assume that you mean "descendants". Indeed, we can safely allow such hypothetical concerns to be addressed later. The same is true of those that might arise within our lifetime. This is a wiki. We base it upon the world as it exists today (including knowledge of what's to come), and if something changes tomorrow, we edit accordingly. —David Levy 21:36, 9 April 2013 (UTC)

You are quite correct, I did indeed mean descendants. Thank you.--Ykraps (talk) 22:34, 9 April 2013 (UTC)

An alternate wording

A topic is the primary topic if this topic is clearly more important to the Misplaced Pages project than any of the other topics with the same title. If no topic is clearly more important, it is better for the title to be a disambiguation page. This would replace the current guidance, not add to it. Perhaps the importance ratings assigned by Wikiprojects could help in this determinations - a top-importance article is usually primary topic, I'd assume. I'm not honestly sure if this is any better, though. Thoughts? Ego White Tray (talk) 01:42, 9 April 2013 (UTC)

I think that dictionary words should be assumed to be very important to the project, alongside the importance of actual articles.

I’d prefer that we err on the side of precision over conciseness, up to 50 characters, for sake of any reasonable but unusual reader being able to guess the subject before downloading the article. --SmokeyJoe (talk) 01:59, 9 April 2013 (UTC)

I think we need PRIMARYTOPIC, but that it's applied way more often than it should be. This rewrite doesn't seem to fix that. It doesn't even make sense, since no other articles have the same title. Think about Big or Brand New. Many will claim that the film and the band are the most important topics to WP, since there's not much else competing with them. These terms are too ambiguous to not be disambig pages. How can we rewrite PRIMARYTOPIC to not tempt editors to make so many primarytopic claims based on simple popularity when the title is so ambiguous? Dicklyon (talk) 02:08, 9 April 2013 (UTC)

I said no other topics, have the same title, not no other articles have the same title. Example: a planet, element and a god are three topics that have the same title, Mercury, but the articles don't have the same title. See the difference? Ego White Tray (talk) 02:43, 9 April 2013 (UTC)

I strongly oppose this idea, which would neither serve readers nor simplify the process.

Our goal is to make it as easy as possible for readers to find the articles desired, not to funnel them to articles about subjects that we've deemed "more important". And such a concept would fuel endless debates among editors (and the aforementioned WikiProjects) regarding which topics are "more important to the Misplaced Pages project" (itself poorly worded, as it could be interpreted to favor topics related to Misplaced Pages or those in which Wikipedians are personally interested). —David Levy 03:52, 9 April 2013 (UTC)

I agree with the statements here from User:Dicklyon and User:David Levy. We should not funnel readers to the most popular pages (google already does that). --SmokeyJoe (talk) 05:19, 9 April 2013 (UTC)

I think I also agree. I have this fear that I'm missing something if the solution could be so simple. Importance is clearly problematic in the application here since it is totally subjective. Vegaswikian (talk) 06:11, 9 April 2013 (UTC)

I think objective guidelines can be reached by consensus. From an encyclopedic standpoint, scientific significance should predominate over commerce and pop culture. Taking Mercury for an example, I would contend that the element, being a building block of the universe, is the most significant use for an encyclopedia, followed by the planet, and then the god for its historic value, and then the space program, and then the car (the record company doesn't enter into it, because it's name is Mercury Records). bd2412 T 12:15, 9 April 2013 (UTC)

Again, our goal is to make it as easy as possible for readers to access the articles that they wish to read.

Persons typing "Steve Reeves" probably seek information about the bodybuilder/actor. It would be unhelpful to send them to the article about the computer scientist (on the basis that "scientific significance should predominate over commerce and pop culture"). —David Levy 13:05, 9 April 2013 (UTC)

I was addressing the significance of concepts, not people. The comparable discussion for people would probably be Madonna, for whom there is a person, and a much older usage in history and art. I suppose if Freddie Mercury had gone by his adopted surname alone, then we would have to fit Mercury the musician in with the element and the planet, and a specific person would go near the bottom of that group. bd2412 T 15:09, 9 April 2013 (UTC)

Distinctions like that are arbitrary. There is no clean way to draw the line. It's much simpler and less contentious to simply look at which topic associated with a given term is most likely to be sought - and if one stands out, it's the primary topic for that term. Period. After all, if the significance is really there for a given topic, it will be reflected in its likelihood of being sought. --B2C 15:23, 9 April 2013 (UTC)

But that is also arbitrary. What readers are likely to be looking for depends on so many things, and can vary for the same reader over time. To imply that we have a way to determine what that is for most readers is a delusion. Vegaswikian (talk) 16:58, 9 April 2013 (UTC)

Measurements do not have to be (and cannot be) 100% precise before they can be useful. Using available measures (reductions in uncertainty) is not arbitrary. -- JHunterJ (talk) 17:15, 9 April 2013 (UTC)

Importance is a very subjective and inherently contentious attribute. This is not the intended sense of "primary" in "primary topic", neither literally nor historically in practice. Flawed as it is, likelihood of being the one that is sought is a much more objective attribute. --B2C 15:23, 9 April 2013 (UTC)

Indeed. But that is not the same thing as what most people would expect the article to be about. So if, for example, a manufactured band on a US TV show is given a name which already exists in some other context, their article will probably be the target of most searches for that name, but everyone else will be left confused and even frustrated to get there. There might be no reason why they would even have heard of the band. Whether that is the right solution for WP is a subjective judgement. --AJHingston (talk) 15:47, 9 April 2013 (UTC)

If the article about the manufactured band is clearly getting much more attention, then it should be okay to confuse and even frustrate "everyone else", which will be comprised of a relatively small minority, by definition. Besides, a hatnote link, not to mention informative Google results, should really be enough to limit the confusion and frustration to a minimum.

If the manufactured band is not clearly getting much more attention, then putting the dab page at the plain name is what WP:D indicates.

This is pretty much how it used to work up to a couple of years ago, and everything was much less contentious then. The watering down of primary topic by the addition of the long-term significance consideration has created a lot of ambiguity and contention. --B2C 19:33, 9 April 2013 (UTC)

But the US teenagers in this hypothetical case are not representative of anyone else. The same is true of any other special interest group. My point about objectivity v subjectivity is that the decision whether to stay 'in universe' (to pick on a previous discussion), to give precedence to US users because they are the largest single group (as has been argued in other WP contexts), or to judge the primary topic purely on a statistical analysis of visits over a period or anything else is a subjective decision, and it is only after that objective criteria come into play. That is why we keep going around this - we all begin with different values. --AJHingston (talk) 20:41, 9 April 2013 (UTC)

I have to agree with AJHingston, who seems to me to be saying in essence that Misplaced Pages is an encyclopedia, not a fansite where articles achieve primary status by the number of likes. What is the evidence for thinking that seekers of articles on pop stars and TV shows are stymied by the mounds of Elizabethans and philosophical terms and astral bodies in their search path? Cynwolfe (talk) 23:26, 9 April 2013 (UTC)

Alternatively, what is the evidence for thinking that seekers of articles on Elizabethans and philosophical terms and astral bodies are stymied by the mounds of pop stars and TV shows in their search path? -- ArglebargleIV (talk) 23:44, 9 April 2013 (UTC)

But that's the point: when in doubt, disambiguate, and stop battling over which is the primary topic. Cynwolfe (talk) 00:46, 10 April 2013 (UTC)

And how do we deal with the editors that see disambiguation as bad and that we need to have a primary topic? I believe that I agree with you. However in practice consensus does not support this alternative. Vegaswikian (talk) 01:56, 10 April 2013 (UTC)

Disambiguation is not inherently bad, but erroneous links to disambiguation pages are. Right now, we have over 337,000 such links needing to be fixed, and implementation of this proposal would bump this number back up into the millions. bd2412 T 11:08, 10 April 2013 (UTC)

And even more inherently bad are the unknown (vast) number of articles with titles like Foo (something) with no hatnote or dab page entry to link them from "Foo"! But that's another saga. PamD 11:22, 10 April 2013 (UTC)

@BD2412, 337,000 is a surprisingly small number considering the size of Misplaced Pages. It speaks to the effectiveness of the volunteers who patrol the pages listed on the reports. While I don't support the proposed alternative wording, I would support proposals to raise the bar for for primary topics and have a stronger default to disambiguation where there are substantive questions whether there is a primary topic. As it is, a page with 51% more traffic than all the other ambiguous pages combined can be a primary topic. Than means there could be a very large number of links mistakenly going to the primary topic that go unnoticed by the links to disambiguation page reports and only get fixed on the off chance that someone following the link feels inclined to go back to the page they left and edit the link. I'd say underreporting of incorrect links to ambiguously titled pages is a problem and that including more ambiguously titled pages to be patrolled is a good thing. older ≠ wiser 13:14, 10 April 2013 (UTC)

We have 337,000 today (instead of the 900,000+ we had three years ago) in large part because a handful of us have made hundreds of thousands of fixes. Were we to let up, the number would quickly shoot into the millions. Not a day goes by without some page being turned into a disambig with hundreds (occasionally thousands) of incoming links that go ignored by the editor who made the change. With respect to errant primary topic links, I fix these from time to time also. Apple and Mouse are usual suspects. I would propose that rather than make those titles disambiguation pages, we pipe links to them through unambiguous redirects like Apple (fruit) and Mouse (animal). That way, we can easily see which incoming links are unpiped, and can check those, and fix them if they are wrong. However, because most incoming links would be pointing to the right target, this would be a low priority compared to fixing links pointing to actual disambiguation pages, for which the incoming link is always wrong unless it is piped through a "foo (disambiguation)" redirect to the disambig. In short, I propose a third category. Pages like Barack Obama sit at their clear primary topic title and links go straight to the page with the presumption that they are intended to go there. Pages like Mouse sit at their primary topic title with links piped through a redirect for occasional checking. Pages like Bass disambiguate. Incidentally, there are a great many pages that should not be disambiguation pages at all because they are really just lists of variations of a subject (for example, this months most-linked-to list includes Bionic Commando, British poetry, Commoners in the United Kingdom, Law of the sea, Leaving Certificate, and Peace process, all of which should be general topic articles). bd2412 T 13:46, 10 April 2013 (UTC)

And as someone who has spent time going though that backlog, the links are indicative of the problem. A better place to review this is Category:Articles with links needing disambiguation. These are the articles that have links that other editors can not figure out what the target is. This is an indication of the problem we currently have when an article is in the name space and the title is ambiguous. These are links that from an article and other editors can not figure out what the correct target is. There is no simple way to find those inbound links when they go directly to an article. With a dab page at least those "bad" links are easy to find and can be corrected. So which is better, bad links that you can not find or links that you know need to be fixed? Vegaswikian (talk) 20:52, 10 April 2013 (UTC)

I believe that my proposition would address this problem. bd2412 T 02:09, 11 April 2013 (UTC)

Remove "long-term significance"

Many editors have suggested that the "long term significance" clause is problematic. This is a narrower issue that we might get an agreement on. Should it be removed? Ego White Tray (talk) 00:32, 10 April 2013 (UTC)

No. Long term significance is more important than usage. Long term significance is the most redeeming part of PRIMARYTOPIC, and the only part properly consistent with WP:COMMONNAME, which says that (contrary to the common interpretation of "common") we should use the name that is prevalent in reliable English-language sources. The usage clause invites examination of ghits and page visits and is inconsistent with reflecting the best sources. --SmokeyJoe (talk) 00:51, 10 April 2013 (UTC)

The interpretation of "commonly used" to mean "commonly used in reliable sources" is much better established (thanks primarily to PBS (talk · contribs)) than including "long term significance" as a direct consideration when evaluating primary topic (a relatively recent addition). Indeed, long term significance is inherently accounted for when looking at usage in reliable sources. Calling it out as a separate direct consideration is what the problem is, because it's a much more subjective consideration than looking at usage in reliable sources. Since looking at usage in reliable sources already accounts for "long term significance", giving it separate direct consideration is at best redundant, and often just provides a source of contention. --B2C 19:11, 10 April 2013 (UTC)

I believe that "enduring notability" is a valid consideration, but we're greatly overstating its role and muddying the waters with the highly subjective "educational value" language.

Rather than maintaining a separate "long-term significance" criterion (which seems to have triggered a great deal of conflict), we should qualify the longstanding "usage" criterion by explaining that it sometimes is reasonable to consider historical usage. We shouldn't necessarily focus solely on what usage predominates right now (which might prove ephemeral), but a notable subject's existence over a long period isn't a standalone factor to be weighed against common usage. (And the reference to "educational value" is largely useless.) —David Levy 01:00, 10 April 2013 (UTC)

I think this at best just slightly obscures the problem; it doesn't really address it. But your suggestion has inspired me to make the proposal below! --B2C 19:32, 10 April 2013 (UTC)

It would be better to make an AND instead of the OR. To declare a primary topic, a topic both be much more popular and have more long-term significance. Whenever these two points are in conflict, use a disambig page. The other problem is the phrase "more likely than all the other topics combined", which should be "much more likely than all the other topics combined". The way it is now, it has to be much more signifcant than any other single topic, but a bare majority over a wide ambiguity. Makes no sense. Dicklyon (talk) 03:47, 11 April 2013 (UTC)

It makes perfect sense as is. Given four uses, A,B,C and D, for A to be the PT, it needs be much more likely to be the one being sought than each of the others, and merely more likely than all 3 combined.

So if the likelihood of B C D is 20% each, then A at 40% is not the PT because 40% is not more likely than 60%. But if B C and D are 15% each, and A is 55% then it is the PT, because 55% is much more than 15%, and more than 45%, the combined likelihood of B C and D. --B2C 05:19, 11 April 2013 (UTC)

Also, if A is 55%, B is 35% and C and D are 5% each, A is not the PT because it is not MUCH more likely than B, though it is more likely than the other three combined (45% like in the previous example). --B2C 05:28, 11 April 2013 (UTC)

And if you had 55%, 5%, 5%, 5%, 5%, 5%, 5%, 5%, 5%, 5%, you'd make the first primary, even though it's a bare majority of accesses. Makes no sense (to me) to declare a primary topic for something that's the target in less than about 90% of the accesses. Even 90% is marginal, when with a disambig page we can more easily show people the landscape of what the term refers to and let them find the possibly several relevant articles that they may be looking for. Dicklyon (talk) 05:49, 11 April 2013 (UTC)

With a disambig page we can force them to look at the landscape first, yes. With a primary topic, we can more easily allow most of the readers to efficiently see the topic they sought, and use a hatnote to allow the smaller group of readers to see the landscape and let them find the relevant article they were looking for. -- JHunterJ (talk) 11:19, 11 April 2013 (UTC)

Agreed. That's why we should only use a primary topic when we are absolutely certain that the article in it is what most of the readers are looking for, even if they come from different backgrounds (i.e. without biasing it to just one single group of readers only because it's big - the biggest group may be favored, but only if it's heterogeneous). This means that the choice of primary topics should be made stricter than they currently are. Diego (talk) 12:12, 11 April 2013 (UTC)

That's sounds like "disagreed". We should use a primary topic when it serves the majority of a single group (the group is the readership as a whole). Trying to distinguish those majorities that count (different "backgrounds", whatever those are) from those majorities that don't count (for some reason a "single" group, even though the majority from different backgrounds is also a single group) would be a good cover for "primary topics when I'm in the majority, because obviously my majority is diverse, but no primary when I'm in the minority, because obviously that majority is just a single group". -- JHunterJ (talk) 12:54, 11 April 2013 (UTC)

I agree that primary topics give more exposure to the topic and that this is sometimes a good thing. But distinguishing which topics are likely to be sought shouldn't be done only from raw numbers of readers without any concern of who they are - what's missing in the current PRIMARY TOPIC is a recognition that people come to Misplaced Pages from very different perspectives, therefore only worldwide subjects are likely to be sought by any reader picked at random in the whole world. I like B2C's idea to give emphasis to reliable sources, it's a step in the right direction. I'm not suggesting to distinguish between "groups that count" and "groups that don't" - when there's a reasonable claim that two different groups would be best served by different articles, no primary topic should be used. I'm explicitly thinking of Madonna here, where the number of visits has been repeatedly rejected as the only criterion to be considered even if it's clearly overwhelming. Diego (talk) 14:00, 11 April 2013 (UTC)

I agree that it would be an improvement to require that a topic meet both criteria. Most of the debates seem to occur when the criteria each lead to different outcomes, as the guideline offers no real help on what to do in that case. The best option for the articles in almost all such cases is to have no primary topic, which settles the matter. It would also be an improvement to have primary topics much more likely than any other topic (i.e. around 90% of the usage, rather than 51%). I look at it as a gamble: does anyone need to check links to this title to make sure they will take readers to the right place? If 49% of the incoming links are actually meant for another article, we should be disambiguating the incoming links.--Trystan (talk) 13:50, 11 April 2013 (UTC)

Dicklyon, right. What this is really about is margin of error. In the case of A is 55%, and the others are 5% each, clearly there is no term that people are searching for more than topic A, and it's apparently more than half, so let's take them there. Why take everyone to the dab page?

In the case of A=55, B=45, C=5, D=5, the thinking is that it's not even all that clear that more people are looking for A than B (because of margin of error), so in that case we do go with the dab page. --B2C 16:50, 11 April 2013 (UTC)

So it is OK that 49% percent of people could be directed to the "wrong" page ... and that is acceptable. In what other field would a 49% failure rate be acceptable? older ≠ wiser 18:05, 11 April 2013 (UTC)

If 51% of people are looking for topic "A", and we instead direct them to a disambiguation page, then 51% have been taken to the "wrong" page. Is a 51% failure rate better than a 49% failure rate? Furthermore, is it a "failure" if instead of going to disambiguation page "Foo" that basically says, "you were probably looking for Foo A, but you may have been looking for Foo B, and by the way there is also Foo C, Foo D, and Foo E", the reader is taken directly to page Foo A, that has a note right at the top saying "You may have been looking for Foo B; for other meanings see Foo (disambiguation)"? For the majority of people who are looking for Foo A, there they are. For the large minority looking for Foo B, the link is right there, so they are no more inconvenienced than if they had been taken to a disambiguation page in the first place (one extra click either way); only the very small number looking for Foo C and beyond must face the relatively minor inconvenience of clicking the link to the disambiguation page. For an excellent example of this principle in practice, see Colorado. bd2412 T 21:40, 11 April 2013 (UTC)

Except that arriving to a disambiguation page is not nearly as problematic as arriving to the wrong article. There's the well known effect that navigating to an unexpected page will cause a disorientation that takes several seconds to recover from. A layout where all readers arrive to the DAB is more fair - everybody sees quickly that the page is a directory of links, and all need exactly one click to find their desired article; while when there's a primary topic, the worst case is that will need two clicks to find it, the first one of which is through the cumbersome format of a hatnote (you can visit WT:HATNOTE regularly to learn about the well-known problems it causes). The disambiguation page is inherently more recognizable, and easier to scan or read. Preferring the problems that the primary topic causes over a DAB is only justifiable when its prevalence is overwhelming. Diego (talk) 22:03, 11 April 2013 (UTC)

I don't think there is any consensus on the assertion that "arriving to a disambiguation page is not nearly as problematic as arriving to the wrong article", or that arriving on the wrong article causes disorientation. I, for one, have never experienced it. But this too is even indirectly addressed by primary topic. The idea is that it will only happen when the "right page" is a relatively obscure use of the term in question. For example, if someone looking for the relatively obscure "Paris" in Texas arrives at the article about the city in France, is that really a problem? At the very top it clearly states: This article is about the capital of France. For other uses, see Paris (disambiguation).

I don't see how it's a problem at all. --B2C 22:27, 11 April 2013 (UTC)

B2C, you've gone straight to the other end of the spectrum where it is likely that far far more than 51% of people will be expecting the city in France to be at Paris. For pages where the majority is much less certain, I agree with Diego that there is a disorientation involved in arriving at the "wrong" article rather than a disambiguation page. A disambiguation page at the base name is effectively signalling that the term is ambiguous and we (the editorial Illuminati) are not making any presumptions about what the reader might have been looking for. I'd like to see a rather higher bar for a primary topic than 51% and have a stronger default to disambiguation if there is no obvious primary topic. older ≠ wiser 23:29, 11 April 2013 (UTC)

I too would like to see a rather higher bar for a primary topic. I would also like it to be source-based, not pageview or ghit based or percent internal searches. A criterion like “whether a broad-audience international newspapers disambiguates the name when using it” could be good to include in the list of suggestions. A newspaper would not explain that the dog mentioned was an animal, but would indicate that the dog tool is a type of tool. Newspapers indicate whether mercury is an element, planet or god. Newspapers do not usually specify that Barack Obama, or Bush, or Bill Clinton is/was a US President, but I do think they do remind the reader who John Adams was (especially to an international audience). Therefore, Dog, Barack Obama, Bill Clinton are not disambiguated by us, but Mercury and and John Adams should be. This criterion would align the likelihood of bewilderment at arriving at the wrong article with the likelihood of bewilderment upon reading the newspaper. NB. This criteria is not catch-all. There is another reason that we disambiguate at George Bush, and that reason is not pageviews. --SmokeyJoe (talk) 01:31, 12 April 2013 (UTC)

I'm pretty much in agreement. My big concern is that when you have a topic that the news sources fail to disambiguate at all and use it for several uses, most notability raising a US mailing address to the equivalence of a place. One could argue that is is in fact a reason to have the dab page at the main name space. Vegaswikian (talk) 02:05, 12 April 2013 (UTC)

SmokeyJoe, the purpose of disambiguation in WP is disambiguation with other uses of the same name on WP, not to make the title recognizable to a broad audience. So a criterion like “whether a broad-audience international newspapers disambiguates the name when using it” is usually not relevant for primary topic determination.

Articles about barely notable topics are routinely placed at their generally unrecognizable names, when the name is unique or considered the primary topic. Countless examples can be found with SPECIAL:RANDOM (Southampton Street, APEN Agiou Athanasiou, Kottan ermittelt, Myeongbong Station). Thousands and thousands of titles not recognizable to a broad audience exist on WP. They're probably even in the majority of all articles. Is that a problem looking for a solution? I don't think so. So, generally, I don't see why whether a broad-audience international newspaper disambiguates the name when using it should be a consideration for determining primary topic, or any other reason.

However, that said, there is one exception. If one of the uses is generally not disambiguated by such newspapers, but all the others are, that's very strong evidence of that one use being the primary topic (even it's a fictional character and all the others are real person biographies). But, if all of the uses are disambiguated by newspapers, that doesn't mean there isn't a primary topic among them. --B2C 06:14, 12 April 2013 (UTC)

WP doesn’t make things “recognizable to a broad audience”, no, I’m saying we should use evidence of” recognizable to a broad audience” in disambiguation decisions. I think you don’t like this idea because you want to push/support simple titles for the most popular pages. Is that wrong/unfair? Most articles don’t face issues of disambiguation, and most subjects are not broadly recognisable. There’s no issue there. The issue here is with popular, short titles. If our sources disambiguate a topic title, so should we. We should always be guided by our sources. --SmokeyJoe (talk) 08:33, 12 April 2013 (UTC)

As noted earlier, a rather sizable gap exists between Misplaced Pages's actual consensus-backed article titling conventions and your personal preferences. This particular discussion's goal is to accurately and clearly describe the former, so I think that it would be more constructive to narrow the focus accordingly. I realize that you wish to effect change (regarding our actual practices, not merely our documentation thereof), but that's best discussed separately. —David Levy 09:55, 12 April 2013 (UTC)

Consensus can change, but I'm not even sure there ever has been consensus for how to deal with primary topics. I'm pretty sure there's no consensus right now for it, and the actual practices are diverse; so if we're going to craft a wording that will affect future practice, it better get feedback from as many editors' opinions as possible. I actually find “whether a broad-audience international newspapers disambiguates the name when using it” a really good criterion to add to the current batch of things we look for in topics with ambiguous titles; it could be a way to settle the current problems we're experiencing in consensus-building at RMs, although it probably won't be definitive. Diego (talk) 10:59, 12 April 2013 (UTC)

I find the idea confusing, but that's beside the point. Consensus certainly can change, and there's nothing wrong with proposing that we modify our conventions. But it's counterproductive to conflate such discussion with that of our current practices' documentation. —David Levy 11:34, 12 April 2013 (UTC)

David: errrrrr ... what???? Tony (talk) 12:41, 12 April 2013 (UTC)

I don't know what you're asking. —David Levy 13:02, 12 April 2013 (UTC)

If we're not going to get rid of the whole PT thing, then I think looking at the Misplaced Pages:Disambiguation#Determining_a_primary_topic guide is pretty sensible. Currently, it tells people to look at incoming links, page views and google. I think it very odd that it says nothing about sources. --SmokeyJoe (talk) 12:15, 12 April 2013 (UTC)

I think that we're again blurring the lines between WP:PRIMARYTOPIC and WP:COMMONNAME.

As noted above, I find this idea confusing. What, in your view, constitutes "disambiguation" on the part of reliable sources? Does their use of the phrase "President Barack Obama" or "US President Barack Obama" qualify? Are we to determine whether a description is included for the purpose of distinguishing the subject from others with the same or similar name? If so, how? —David Levy 12:35, 12 April 2013 (UTC)

That one's easy. If sources directed to a general audience always use the term "President Barack Obama" or "Barack Obama (the president, not his father)", that would be a telling sign that the name needs disambiguation (in there are several articles in Misplaced Pages using it). On the other hand, if sources for a general audience commonly refer to some "Barack Obama" without needing to distinguish him from other people, and expecting their readers to know who's the one they're referring to, that's strong evidence for the term to be used with a primary topic. Diego (talk) 13:41, 12 April 2013 (UTC)

It's normal to describe entities when writing/speaking of them, irrespective of whether they're well known or better known than other subjects with the same or similar names.

I would expect virtually all non-specialist publications covering the Rolling Stones to include a descriptive word or phrase ("band", "rock band", "musical group", "rockers", "recording artists", etc.), just as we do in our article's lead. This is standard writing/reporting. It doesn't mean that the sources seek to prevent confusion with the novel The Rolling Stones.

Obviously, that example is clear-cut. Other instances won't be. The above proposal is based upon an unreliable inference. —David Levy 14:58, 12 April 2013 (UTC)

And what's better: if sources directed to a general audience from different parts of the world use the same term without disambiguation for different topics, that would be almost definitive evidence that it can't be used with a primary topic. Diego (talk) 14:04, 12 April 2013 (UTC)

That makes sense to me, but it doesn't really deviate from our current practices or documentation thereof; if reliable, non-specialist, English-language sources frequently refer to different subjects by a particular name, a primary topic probably doesn't exist for that term, regardless of where these publications originate and whether descriptions are provided. —David Levy 14:58, 12 April 2013 (UTC)

Not necessarily. There are lots of instances where world-wide usage may be varied, but because of over-reliance on navel-gazing page traffic statistics one usage can be designated as primary. This is a type of systemic bias. older ≠ wiser 15:06, 12 April 2013 (UTC)

Can you cite some specific examples? —David Levy 15:13, 12 April 2013 (UTC)

These criteria being discussed, they would be suggestions to editors for points to inform the discussion, not factors feeding a formula, No? I wonder is David Levy is fearing a formulaic approach being pushed? --SmokeyJoe (talk) 01:51, 13 April 2013 (UTC)

As noted above, my main concern regarding your "If our sources disambiguate a topic title, so should we." proposal is that it seems to be is based upon the unreliable inference that such a description reflects a determination that confusion with similarly named subjects is likely. —David Levy 01:59, 13 April 2013 (UTC)

I don't see this as a major concern. I don't see the inference as necessarily unreliable, though odd things are always possible. I've mentioned major braod-audience newspapers as one type of source to look at. Probably, the sources cited in the article lede should be considered first. Why would you consider aberrant usage in these sources to be likely? Page view statistics can also produce aberrant results, yet the guideline advises their use without warnings. --SmokeyJoe (talk) 02:10, 13 April 2013 (UTC)

You appear to have misunderstood my concern. I'm not referring to aberrant usage. I'm saying that normal usage doesn't imply what you evidently believe it does. Please see my elaboration (including examples) above. —David Levy 02:18, 13 April 2013 (UTC)

I did read your posts and examples above. Your concerns read as something to consider in refinement of the actual wording inserted into the guideline so as to prevent these possible unintended readings. However, overall your concerns seem unfounded. Reference 1 of The Rolling Stones is a document titled "The Rolling Stones Biography", which is undisambiguated. Nowhere in the article is the least consideration of hillsides and round rocks. Everything I find on Dog (engineering) says something explicit about it being a tool, as if the reader might not have known. I think that everything you say is important, but not a problem to ensure it is avoided. On the other hand, your opposition to the mere mention of usage in reliable sources seems odd. I also don't understand you concern of blurring the lines between WP:PRIMARYTOPIC and WP:COMMONNAME. The two should be, could be, in complete harmony. Why should there be a dividing line? --SmokeyJoe (talk) 02:34, 13 April 2013 (UTC)

Your concerns read as something to consider in refinement of the actual wording inserted into the guideline so as to prevent these possible unintended readings.

I don't know what "unintended readings" you mean. You haven't answered my questions about what reading is intended.

However, overall your concerns seem unfounded.

They might be founded in an inaccurate understanding of what you've proposed. I'm not certain that we're on the same page.

Reference 1 of The Rolling Stones is a document titled "The Rolling Stones Biography", which is undisambiguated.

Firstly, the Rock and Roll Hall of Fame and Museum is a specialist source. Secondly, the Rolling Stones are described in the article as "a rock and roll band", "band" and "group". Perhaps that isn't the sort of "disambiguation" to which you intended to refer. It's precisely the type that I mentioned above, so it should be clear that it's what I had in mind.

Nowhere in the article is the least consideration of hillsides and round rocks.

Are you referring to explicit mentions of an alternative usage? Your following sentence seems to imply otherwise...

Everything I find on Dog (engineering) says something explicit about it being a tool, as if the reader might not have known.

This is what I mean. It's an inference on your part. You (correctly) regard that usage as much less common than another is, and this influences your perception of the sources' coverage. You see standard descriptions of the topic and assume that they constitute evidence that it's been deemed less familiar than another known by the same term. But you ignore comparable descriptions of the Rolling Stones — something explicit about them being a band. You cite the lack of "consideration of hillsides and round rocks", but you don't appear to demand that coverage of the tool type include explicit considerations of the mammalian subspecies referred to by the same name. You interpret the descriptions as evidence of what you already believe (and not evidence of what you don't believe).

I think that everything you say is important, but not a problem to ensure it is avoided.

I'll copy and paste a couple of my above questions:

Are we to determine whether a description is included for the purpose of distinguishing the subject from others with the same or similar name? If so, how?

On the other hand, your opposition to the mere mention of usage in reliable sources seems odd.

It's nonexistent. I find the assertion odd, given the fact that my suggested addition to WP:PRIMARYTOPIC is explicitly based upon "usage in reliable English-language sources".

I also don't understand you concern of blurring the lines between WP:PRIMARYTOPIC and WP:COMMONNAME. The two should be, could be, in complete harmony.

I believe that they are. Improvements are possible, but I see no conflict between the two.

Why should there be a dividing line?

They address distinct concepts. WP:COMMONNAME is about determining whether a subject is most commonly known as "x", "y" or "z". WP:PRIMARYTOPIC is about determining whether a topic referred to as "x" (irrespective of whether this is its most common name) is better known as "x" than other topics are.

For the type of tool discussed above, the common name apparently is "dog"; it isn't referred to as anything else with the same frequency (if at all). WP:COMMONNAME advises us to use that name instead of some other designation.

But for the term "dog", the primary topic (or whatever we're to call it) is Canis lupus familiaris, which is much better known as "dog" than any other subject is.

One determination has no bearing on the other, excepting the possibility that a subject has more than one common name, in which case the absence of a primary topic for one or more of these names might influence our article's title. (We prefer Cheque over Check (finance) and Apartment over Flat (domicile). The Vest article covers multiple meanings, whose individual articles reside at other common names.) —David Levy 09:55, 13 April 2013 (UTC)

The Rolling Stones (novel). None of the references that are immediately accessbile use the title in introduction. Instead, the title is introduced. For example, the external link reads: "First serialized in compressed form in Boys' Life Sep-Dec 1952 as Tramp Space Ship. This book makes reference to ". This is evidence against the title not needing disambiguation, vis a vis the other subjects. --SmokeyJoe (talk) 02:40, 13 April 2013 (UTC)

You're citing an entry from a database in which miscellaneous notes appear in that format. It doesn't imply what you perceive it to. I'm perplexed as to why you believe otherwise, given the fact that the entry begins with "Title: The Rolling Stones". —David Levy 09:55, 13 April 2013 (UTC)

You're persuading me that this is not as simple as I imaged. I appear to have blundered with the Rolling Stones. --SmokeyJoe (talk) 22:39, 13 April 2013 (UTC)

Make "long-term significance" implicit by clarifying reliable source qualification

I note that WP:COMMONNAME is clarified with the following important statement:

The most common name for a subject, as determined by its prevalence in reliable English-language sources, is often used as a title because it is recognizable and natural.

I propose we do something similar for primary topic:

We remove the long-term significance consideration bullet from WP:PRIMARYTOPIC
We clarify that when looking at usage to determine primary topic, we're looking at "prevalence in reliable English-language sources".

This will eliminate the inherent conflict and contention created by having two explicit potentially contradictory criterion of unspecified priority, and yet will continue to properly account for the long-term significance consideration because it is implicit in reliable source usage.

Specifically:

Change this:

There is no single criterion for defining a primary topic. However, there are two major aspects that are commonly discussed in connection with primary topics:

A topic is primary for a term, with respect to usage, if it is highly likely—much more likely than any other topic, and more likely than all the other topics combined—to be the topic sought when a reader searches for that term.

A topic is primary for a term, with respect to long-term significance, if it has substantially greater enduring notability and educational value than any other topic associated with that term.

In many cases, a topic that is primary with respect to usage is also primary with respect to long-term significance. 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 long-term significance. In such a case, consensus determines which article, if either, is the primary topic.

To this:

There is one single criterion to consider when determining a primary topic:

A topic is primary for a term, with respect to usage in reliable English-language sources, if it is highly likely—much more likely than any other topic, and more likely than all the other topics combined—to be the topic sought when a reader searches for that term. The likelihood that a topic is the one being sought for a given term is determined by the prevalence of usage of that term to refer to that topic in reliable English-language sources.

--B2C 19:32, 10 April 2013 (UTC) slight wording change --B2C 20:49, 10 April 2013 (UTC)

I agree that the separate "long-term significance" criterion is ill-advised, but I don't see why we shouldn't address the concern that led to its addition by explicitly noting the relevance of historical usage. I understand your point that it's implied, but there's no harm in spelling it out. —David Levy 21:02, 10 April 2013 (UTC)

Yes, there is harm in spelling it out. Spelling it out creates an explicit separate consideration, one that can be evaluated only subjectively. It can suggest, based on one's biases, a different title than the one suggested by usage in reliable sources. Actual usage in reliable sources is a consideration that can be reasonably evaluated much more objectively than "long-term significance".

Two different criteria potentially suggesting different titles is precisely what leads to much of the disagreement and contention regarding primary topic determination.

If someone thinks that a certain topic is or is not primary because of long-term significance, they need to prove it by showing how it is or isn't referenced accordingly in reliable sources. Usage in reliable sources is ultimately all that matters, and, so, that's all that should matter, but by spelling out "long-term significance" as an explicit consideration, that suggests otherwise, creating potential contention. Totally unnecessary. --B2C 21:33, 10 April 2013 (UTC)

I'm not sure that you understand my position. I agree that we shouldn't have two different criteria or describe "long-term significance" as a separate consideration. I mean that there's no harm in expressly stating that a term's long-term usage by reliable English-language sources can be relevant. It might be implicit, but it's far from obvious. That's why editors felt the need to add the "long-term significance" criterion in the first place. It wasn't a baseless change; it was overcompensation. I view your proposal in a similar light. —David Levy 22:06, 10 April 2013 (UTC)

Please provide some specific wording, so others can better understand what you mean. --B2C 22:50, 10 April 2013 (UTC)

I'll give it a shot, but please note that the following wording isn't part of a formal proposal. As demonstrated above, it can be a bit tricky to prevent the inference that "long-term significance" is being presented as a separate/competing consideration, so brainstorming may be required.

Both historical and current usage in reliable English-language sources may be relevant, depending on the circumstances. When examining these trends, editors should avoid assigning disproportionate weight to past usage whose prevalence has declined or recent usage that may be ephemeral.

—David Levy 11:48, 11 April 2013 (UTC)

I like the direction here, but propose this instead:

Per WP:COMMONNAME, the preferred title for an article is the name most usually used in the most reputable and reliable English-language sources, subject to a number caveats further explained in that policy. If all these conditions are met, this title is called the “Commonname title”.
Where multiple articles appear to claim the same Commonnname title, the competitiveness of each claim should be judged by the prevalence of use in the majority of sources, of the non-disambiguated, non-contextualised use of the title in introducing the subject, the reputation and reliability of the sources, and the number of sources.

Where multiple articles competitively claim the same Commonname title, disambiguation is needed at the Commonname title.

P1. WP:PRIMARYTOPIC (a term I would prefer to substitute, and so am avoiding) is directly subordinate to WP:COMMONNAME. I would want to avoid tangential paraphrasing of WP:COMMONNAME regarding inaccuracy or non-neutrality in sources, reasons to not use a Commonname title for reasons other than ambiguity.

P2. A slight tightening of the WP:COMMONNAME criteria, specifically directed to ranking the strengths of claims of different subjects to the same title, calling directly and solely for examination of use in sources. Google hits, page views, incidental mentions in social media, academic worthiness, etc, are not suggested as useful criteria. I consider that under P2, Dog makes a competitive claim to Dog, while no other nearly does so due to sources not using the simple term “dog” without disambiguation from another dog and without contextualisation to a very specific subject. I consider that multiple subjects are competitive in making a commonname title claim for Mercury.

P3. A simple conservative statement strongly reflecting current consensus, but phrase in the reverse of the current WP:PRIMARYTOPIC. (I believe that a dictionary definition should diminish a commonname title claim, but that is a separate debate). --SmokeyJoe (talk) 23:36, 10 April 2013 (UTC)

Let's not conflate primary topic and common name. COMMONNAME is about title selection; PRIMARYTOPIC is about disambiguation.

Recall that the same topic may be primary for multiple names, only one of which (or even none of which) is the most common name for that topic. For example, Barack Obama is the COMMONNAME for that article, but the topic at that article is primary for 122 other terms.

An article may also not be at its most common name, because it's not the primary use of its common name, and so it's at a disambiguated title, but at the same time may be primary for any number of other terms. For example, Portland, Oregon is disambiguated (it's not primary for Portland), but it's also the primary topic for its nickname, Little Beirut.

Primary topic considerations have to be made equally, regardless of whether the term in question is also the topic's COMMONNAME. Equal consideration in primary topic consideration must also be given regardless of whether the term is a candidate for the article's title, or is just being considered as a redirect or dab page entry to the article. Much of SmokeyJoe's proposed wording just above does not reflect any of this. It is reflected in the current wording of primary topic, and in what I have proposed above. This must be retained. --B2C 00:36, 11 April 2013 (UTC)

I reject several of B2C's views, and note that his answer illustrates the clumsiness of "primary" to explain what we do. --SmokeyJoe (talk) 01:52, 11 April 2013 (UTC)

Views? What views? What clumsiness? Please explain. --B2C 04:57, 11 April 2013 (UTC)

If I may answer, first, there's your view that primarytopic is a relevant concept where there is no ambiguity, even though this WP term of art is defined only in the context of ambiguity, starting with "Although a word, name or phrase may refer to more than one topic, it is sometimes the case that one of these topics is the primary topic." Then there's your opinion that "Primary topic considerations have to be made equally." Maybe so, maybe not. I don't see a guideline or policy that says that, and maybe it makes more sense to be more flexible. But I don't really like Smokey Joe's proposal either, as the case that "multiple articles competitively claim the same Commonname title" is not the only situation where ambiguity exists and disambiguation is needed. For one thing, Commonname is about names; we also have ambiguity with common non-name words, frequently; like Run, which is Commonname for some things for generic in others. Dicklyon (talk) 06:01, 11 April 2013 (UTC)

Is the multiplicity of reasons for disambiguation the biggest problem? On that point, I agree with you. --SmokeyJoe (talk) 07:15, 11 April 2013 (UTC)

Dicklyon, differences about those views are largely moot, don't you think? For example, what does it matter whether we refer to the capital of CA as the primary topic of Sacramento, or not? Per your view, it is the PT only if there are other uses of "Sacramento", right? Per my view, we call it the PT regardless of whether there are other uses. Yet, whether we call it the PT of "Sacramento" or not, Sacramento is the title of, or redirects to, the article about that topic.

And I didn't say PT considerations have to made equally, period. I said they had to made equally without regard to common name. That comes directly from the definition of primary topic, which makes no reference to common name. So why should common name be given any consideration during primary topic determination? --B2C 16:40, 11 April 2013 (UTC)

To B2C:

Your apparent views I don't accept:

PRIMARYTOPIC is independent of COMMONNAME. (My view: Usually, Primary topic only comes into play after COMMONNAME has pointed two articles to the same title.)

PRIMARYTOPIC is well defined.

Sentences like "the same topic may be primary for multiple names" make sense. You think "primary" is a function of a name?

Your view that it is good to optimise titles for the most frequent users. (I think, where this is at any expense of another user, it is bad)

Your view that precision is best avoided in titles.

Clumsy is the word "primary" as used here, especially by you. In every normal use of the word, it is decisive and objectively defined (although the definitions may be debated). Primary sources, for example, are primary or they are not primary. There are not degrees. There may be qualifications, such as source A is primary for point X, but it either is or it is not. One source is not more primary than another. Whether this source is primary is completely independent of all other sources. In colours, a source is perfectly primary, or it is not. Primary means "first". It does not mean approximately 1.0. I suspect that I could understand you much better if you rephrase all of the above without using the word "primary". This DAB page has created an unhelpful jargon word. I've looked at the history, and the evolution of the use of the term at WP:D. It happened in steps, and has got worse since introduced in ~2008. --SmokeyJoe (talk) 10:29, 11 April 2013 (UTC)

I agree with Born2cycle's assessment of your proposed wording, but I also understand your concern regarding the use of the word "primary". If switching to a different term would reduce confusion/misunderstanding, we should do that. I hope that we can set aside tradition (something that Wikipedians often seem reluctant to do) and consider the possibility.

"Preponderant" is more precise, but that isn't a particularly familiar word. "Predominant" is a more common synonym, but it also can convey traits (power and authority) that we don't seek to reference. The same is true of "principal" (which can pertain to importance and value) and "chief" (which shares the unintended meanings of "predominant" and "principal"). I'm tempted to suggest "prevalent", but its definition ("widespread; of wide extent or occurrence; in general use or acceptance") seems a bit too broad. (We want to communicate that a usage is more widespread and of widest occurrence/acceptance.) —David Levy 11:48, 11 April 2013 (UTC)

A word like prevalent would improve the logic of the sentences. A simple substitution highlights some troublesome, artificial definitions. --SmokeyJoe (talk) 13:05, 11 April 2013 (UTC)

Does "prevalent topic" make sense? I suppose we're talking about the topic that is referred to most prevalently by a given term... does that make it the "prevalent topic"? --B2C 16:42, 11 April 2013 (UTC)

That would be my question to you and others who see useful meaning in the concept. I find it confusing, see holes in it, and may not be the best person to answer this. Prevalent is the best of a few words suggested, but I think it would be best to speak of a prevailing topic, and to speak of measures of prevalence relating, measure using sources. --SmokeyJoe (talk) 21:39, 11 April 2013 (UTC)

The concept is simply this: Just because there are several topics with a given name, does not mean that name cannot be the title or redirect to one of them. If one of them prevails in terms of usage of that term in reliable sources, then that name should redirect to, or be the title of, the article of the topic that so prevails.

Is that confusing? Is it not useful? What are the holes? --B2C 22:04, 11 April 2013 (UTC)

That is not confusing or with holes. If that were the language we could discuss differences more efficiently. I would like to discuss problems with the prescriptive "then that name should redirect to". But if we could first change the whole section to straightforward standard English, it would be good. --SmokeyJoe (talk) 22:25, 11 April 2013 (UTC)

We can work on the wording, but you suggested the concept was a problem. Are we good with the concept? --B2C 05:50, 12 April 2013 (UTC)

I have problems with the concept yes. But the wording problem is first and worse. The wording problem causes actual communication problems at RMs when editors usethe same terms/links with different interpretations. I'm open to the possibility that the concept problem will go away if the wording is fixed.

I've been looking at your proposal leading this subthread, and have been looking to reword it without "primary". --SmokeyJoe (talk) 08:25, 12 April 2013 (UTC)

After thinking about this, I believe your view here is very reasonable. Commonname determines what would be the desired name for an article solely based on the subject of the article and how it is used in reliable sources and in the wild. We also have cases where the common name is not even considered for an article since there are naming conventions that dictate the use of another name. Primary topic only comes into play if we have multiple articles with a claim, no matter how small, to a specific title. The fact that existing guidelines and policies may dictate a different name does not drop those names from consideration in a discussion of primary topic. Vegaswikian (talk) 23:25, 17 April 2013 (UTC)

I don't have a problem with having two criteria; determining PT is a complex decision and I don't think it can be simplified to a single criterion. (The guideline could be improved by providing some guidance on what to do when the criteria point to different articles.) I would have much more difficult interpreting a guideline saying "the likelihood that a topic is the one being sought for a given term is determined by the prevalence of usage of that term to refer to that topic in reliable English-language sources," because it strikes me as articifally defined. There are much better metrics of determining which topic is likely being sought, like article traffic and incoming links. Usage in reliable sources is also a good PT criterion, but it's separate from determining the likelihood of a topic being sought.--Trystan (talk) 13:47, 12 April 2013 (UTC)

Good point. I agree the definition should continue to be based on the concept of likelihood of being sought, and that prevalence of usage in RS is but one criterion that can be used to help determine that. --B2C 16:30, 12 April 2013 (UTC)

I'm thinking the question of whether PrimaryTopic, however described or named, will need an RfC to gauge community consensus on whether it considers (1) actual usage in sources, or (2) data on searches/page views, or some balance. My sense that that (2) is favoured here and (1) will be prefered by the community at large. --SmokeyJoe (talk) 01:45, 13 April 2013 (UTC)

It depends on the situation, but in all cases we should be using these various tools to determine the likelihood a given topic is the one being sought for a search of a particular term, compared to the other uses of that term on WP. In some cases looking at usage in RS may make more sense, in other cases link counts, in still others page views. But most of the time we should get the same answer from all these tools. If the tools don't all indicate that one topic is primary, without good explanation for any discrepancy, there probably isn't a primary topic for that term. --B2C 22:36, 16 April 2013 (UTC)

We should not be using likelihood of search, because in optimising access for the majority we neglect the needs of a minority. While we are a top ten website with massive traffic, this is not our primary objective. We shouldn't pander to teenage pop culture at the expense of dry academic subjects. Our primary objective (See Misplaced Pages:Prime objective if you like) is to provide access to all information for all people. Ignoring rarer uses is not conducive ensuring the breadth of this. Instead we should ensure all searches lead efficiently to the right page. --SmokeyJoe (talk) 01:01, 17 April 2013 (UTC)

I'm all for minority rights, but when the issue is whether a minority should be inconvenienced to benefit the majority, or vice versa, I fail to understand why we should favor inconveniencing the majority to benefit a minority. --B2C 14:58, 17 April 2013 (UTC)

Further, your objection to using likelihood of search seems to be predicated on the assumption that taking someone to the wrong article is substantially more problematic than taking them to a dab page, an assumption for which I know of no evidence which supports it, including no evidence that this notion is accepted by consensus. To the contrary, the entire notion of primary topic is based on the goal to minimize the number of clicks users looking for pages have to make. Therefore, if by far most of the people searching with a given term are seeking one particular topic, we should take everyone searching with that term to the article about that topic. From there, the minority looking for another topic is given directions on how to get to their desired destination. The alternative is to send no one searching with that term to the place they seek, by taking everyone to a dab page. That's not what we want to do.

And the hysterical hypothetical about pandering to teenage pop culture is often cited, without a single actual example of that ever occurring, so far as I know. --B2C 16:57, 17 April 2013 (UTC)

"Number of clicks" is not the only measure relevant to browsing and should be not be taken as such. The load size of the target page is what causes the major delay in navigation; for every user that is wrongly sent to the "primary topic" article, the user will have to wait until the useless article fully loads, then the DAB page, and only then the desired article; this is costly for mobile browsing and for users without broadband or at places with bad connections; while DAB pages are commonly pretty small and without images, so they should load quickly for all users. Therefore erring on showing a DAB page is inherently safer than sending people to the wrong article - in particular for articles that are considered primary topics by popularity alone, which are likely to be large and contain various images. Diego (talk) 22:14, 17 April 2013 (UTC)

Page load time was already a non-issue when these discussions first occurred over 10 years ago. Mobile access is made through the abridged mobile interface, so even if there was an issue there, it's already greatly mitigated by the mobile interface.

To start considering page load time as a factor in deciding how and when to disambiguate at this stage in WP's life is plain silly. --B2C 23:18, 17 April 2013 (UTC)

By the way, when I turn off WIFI and 3g on my phone, it takes 4 seconds to load (the relatively large) List of M*A*S*H characters#Frank Burns, and it also takes 4 seconds to load the (very small) Frank Burns (disambiguation) dab page. --B2C 23:26, 17 April 2013 (UTC)

WIFI does not mean there is no noticeable delay. Also the mobile interface seems to be more linear so that you need to scroll through several screens to get to what appears at the top of the article with a normal browser. So users don't know if the comment on the first line applies since they don't know any of the content until later. Vegaswikian (talk) 23:32, 17 April 2013 (UTC)

"Page load time was already a non-issue when these discussions first occurred over 10 years ago". Nonsense, I tell you from personal experience. In recent years, using dialup, 56kbs, loading a large page would take minutes, especially if also loading other pages, which is tempting after the first 10 seconds of wait. The preference option to not load the whole page when loading a diff was for me the most fantastic discovery. I routinely load the history and then the diffs to avoid delays. I've also used nomimal 100Mbs that ran at a trickle at peak times. 3G is also highly variable, but behaves much more in waits and spurts, meaning that any page may take time, but then the whole lot comes at once. I found dialup, although slow, the most reliable access, although I gave it up when browsers stopped allowing me to disable loading of images.

I think the majority should be inconvenienced to being sent to a dab page if otherwise it means sending others to a page, such midway down as a long character list, that could be bewildering. The character is an old closed topic. The senator's page is probably suffering from lack of exposure by being buried.

I consider the inconvenience of being sent to a dab page very very small. I actually like to be more aware of similar topics. In fact, as said before, sometimes the information on the dab page is everything I wanted.

"predicated on the assumption that taking someone to the wrong article is substantially more problematic than taking them to a dab page". That is right.

"the entire notion of primary topic is based on the goal to minimize the number of clicks users looking for pages have to make." I understand that. I think that potential for bewildering minority users should be listed as one concern, as a possible reason for disambiguating regardless of a popular PT. --SmokeyJoe (talk) 05:47, 18 April 2013 (UTC)

The preference option to not load the whole page when loading a diff was for me the most fantastic discovery I pray, tell me where do you activate that setting? Diego (talk) 06:07, 18 April 2013 (UTC)

Preferences, Misc, top check button. --SmokeyJoe (talk) 08:59, 18 April 2013 (UTC)

Thanks! It's been a time since I checked Preferences. This does improve checking changes to articles when on mobile. Back in topic, we can't assume that a majority of readers will connect under ideal conditions - the group of users behind problematic connections is significant, and it's likely to be people without a thorough digital culture, so every navigation help is important. Diego (talk) 13:35, 18 April 2013 (UTC)

Let me put it this way. Even 10 years ago page load time was not enough of an issue for it to be raised, discussed and considered, much less listed as a factor to weigh in determining primary topic. To introduce it now would mean reevaluating thousands if not millions of decisions, taking this factor now into account, for the first time. I fail to see why it should be considered now when it never was before.

Every now and then people express a preference for the type of serendipitous discoveries that can be made on dab pages, but consensus seems to consistently prioritize getting people to the page they seek with fewer clicks above that. I am not aware of any reason to believe consensus may have changed on this point. --B2C 15:31, 18 April 2013 (UTC)

I don't think there was ever a very strong consensus for primary topic based on the "number of clicks" rationale. I think the foundation was more along the lines of a principle of least surprise. I can't speak for Diego or others, but I've always thought a simple majority (51%) was far too low of a threshold. There are subjects with ambiguous names, such as "Paris", for which there would be near universal agreement that readers would not be surprised to end up at the article on the French capitol when entering "Paris" into the search box, or clicking on a link to Paris. There are many other ambiguously titled subjects where there might be a regional recognition of one sense as primary, but which would not be readily recognized elsewhere. There are no clear-cut objective criteria that I'm aware of for determining this, other than discussion and consensus (which I'm sure means you'll reject it out-of-hand as leading to JDLI reasoning). I'd favor adding language stating that if there is principled disagreement as to whether there is primary topic, we should default to disambiguation. The qualification of "principled" is intended to deprecate disagreement based solely on subjective notions. older ≠ wiser 16:21, 18 April 2013 (UTC)

That's right. When users expect the topic that is found under the base name, it's not a big deal; readers looking for a different topic will know to add extra parameter to their search. The problem exists when an article is shown for users that wouldn't expect it; if that happens with any likelihood, that article is by definition not a primary topic and shouldn't be there. Consensus is that the best outcome for each case is decided within the most relevant parameters for it, with no single criterion valid in all situations. I'd support clarifying this in the policy, as well as defaulting to DAB when no local consensus exist. Since policies are not prescriptive, a change in it doesn't imply a need to have a sistematic review of all current pages; only new discussions would be informed by the additional clarifications. Diego (talk) 16:52, 18 April 2013 (UTC)

The Potter Stewart solution

Here's a crazy idea:

Misplaced Pages editors have struggled to define what makes something a primary topic but we know it when we see it. In cases where there is doubt, this will need to be discussed at the relevant talk page.

Ego White Tray (talk) 22:26, 10 April 2013 (UTC)

Prefer "In cases where there is doubt, there is no 'primary topic'". --SmokeyJoe (talk) 23:38, 10 April 2013 (UTC).

And how will this play out? Shall we determine consensus for doubt? Or is a single doubter sufficient -- JHunterJ (talk) 23:41, 10 April 2013 (UTC)

Yes. It's a crazy improvement. I more prefer my above solution that points to ranking each claim by use in sources. --SmokeyJoe (talk) 23:52, 10 April 2013 (UTC)

I don't know about "doubt", but the current scheme of "In such a case, consensus determines which article, if either, is the primary topic" doesn't work well when there's no consensus. It should default to no primary topic. Dicklyon (talk) 06:04, 11 April 2013 (UTC)

I concur. There are two improvements we could made right now without changing the wording of the current criteria (nor needing us to get them exactly right): make "no primary topic" the default when there's no consensus, and make it clear that any criteria listed are not final - they're descriptive, not prescriptive, so the final decision should be influenced only by whatever the local discussion agrees to by rough consensus. An explicit reminder of WP:NOTBUREAUCRACY and WP:IAR will help alleviate current discussions, where editors (myself included) often try to use WP:PRIMARYTOPIC as law even if its wording doesn't have a strong consensus supporting it. It should also make it easier to achieve consensus at corner cases, where the current criteria are not helpful and some other local considerations would decide that case better. Diego (talk) 12:24, 11 April 2013 (UTC)

I agree. I see non regular RM participant apparently assuming a PrimaryTopic exists, and that editors have to decide on one. --SmokeyJoe (talk) 21:44, 11 April 2013 (UTC)

The problem with making "no consensus" default to no primary topic is often we don't have a good measure of that particular question. That is, usually we're talking about an RM proposal where some of the arguments for or against moving the article to another title may be made in terms of the topic being or not being primary for the current or proposed title. But most participants are often silent on the primary topic question.

For example, look at the current discussion at Talk:Frank_Burns_(disambiguation)#Requested move. The proposal itself is ostensibly based on WP:PRIMARYTOPIC, but few supporters weigh in one way or another on that particular issue. Even those few who mention "primary" mostly don't really argue either way. Dicklyon, for example, simply asserts, without basis, much less any reference to policy/guidelines, that a fictional character should not be "primary". Similar, George Ho just asserts the topic is not primary. Even bd2412, who allows that a fictional character may be a primary topic, asserts that he doesn't believe this topic is, without saying why. No mention of search likelihoods, or any of the tools mentioned at WP:PRIMARYTOPIC. It goes on and on. I suggest it would be fair to characterize most of their positions as, "I don't know or care if it's the PT or not, I just don't like that title for this article". On the Oppose side there are at least several who argue the topic is primary topic for the current title, and reference what WP:PRIMARYTOPIC says as support. So, what is there consensus on the PT question in this case? If you look at the weight of the arguments, you'd have to say it is the primary topic, since there is no sound argument based on what the relevant guidelines actually say that it isn't. But, at least for now, the supporters greatly outnumber the opposers. --B2C 23:56, 12 April 2013 (UTC)

"no consensus" default to no primary topic is often we don't have a good measure of that particular question. We need good measure of that question, no?

The Frank Burns discussion shows that this guideline is not in harmony with community consensus. --SmokeyJoe (talk) 01:40, 13 April 2013 (UTC)

No discussion with so few participants, including this one and the Frank Burns one, says anything about community consensus. And when the predominant sentiment in such a discussion is so blatantly contrary to what policy and guidelines have long said, particularly when those expressing this sentiments show little knowledge and understanding of the underlying issues, or the implications of their position, as is the case at Frank Burns, it's likely to be an anomaly. --B2C 22:52, 15 April 2013 (UTC)

Maybe I have a skewed view, being in the habit of browsing the RM discussions from the back end. But doing so, time and time again, PrimaryTopic is being thrown around like an ill defined buzz term. In this case, if PrimaryTopic means that Frank Burns, the redirect to a mid list character now decades finished, should occupy the undisambiguated title while several other similarly named articles should be found via three clicks and the very large MASH character list, then this guideline is plain wrong. However, I don't think, unlike you, that this guideline really means to says that. --SmokeyJoe (talk) 01:10, 16 April 2013 (UTC) actually, given the character in question, I'm not sure there's not a joke at play

It may mean that or not. It depends entirely on the likelihood that people searching with the term "Frank Burns" are looking for the MASH character as opposed to other uses. If the likelihood is high enough to meet the PT criterion stated there, then it is the PT, by definition. It is that simple.

Looking at it backwards, by deciding the guideline is "plain wrong" if the result is that link redirects to that section about the MASH character in that article, without regard to said likelihoods relative to each other, is, well, backwards. It's also making decisions based on undocumented and unspecified bases that have no known consensus support. --B2C 02:00, 16 April 2013 (UTC)

Explaining the likely bewilderment for someone wanting Frank_Burns_(politician) is so very far from WP:JDLI, and is a concern to which this guideline is silent. --SmokeyJoe (talk) 04:24, 16 April 2013 (UTC)

The politician's page gets 2-6 views per day. The Frank Burns redirect gets 30-40 views per day. That's an order of magnitude difference. Let's get real. Limiting bewilderment is an implied concern managed by looking at likelihood; it's not such an overriding concern that it trumps overwhelming likelihood evidence. --B2C 18:08, 16 April 2013 (UTC)

Matters for resolution pertaining to the Frank Burns (disambiguation) page

A discussion germane to the one above, regarding criteria to be applied in the choice of WP:PRIMARYTOPIC in human name disambiguation pages, as well as the specificity of the disambiguating qualifier to be appended to a fictional character who is a member of the cast of a TV series, is currently active at Talk:Frank Burns (disambiguation)#Requested move. —Roman Spinner (talk)(contribs) 13:48, 10 April 2013 (UTC)

Talk:South Pier, Blackpool

No other article South Pier, so according to WP:DISAMBIGUATION, is "Blackpool" unneccesary disambiguation or not? In ictu oculi (talk) 03:57, 12 April 2013 (UTC)

Similarly, North Pier is currently a redirect to North Pier, Blackpool. 92.40.233.132 (talk) 12:11, 12 April 2013 (UTC)

Likewise Central Ferry Piers redirects to Central Ferry Piers, Hong Kong - see Talk:Central Ferry Piers, Hong Kong#Move. 92.40.233.132 (talk) 23:24, 12 April 2013 (UTC)

South Pier (Blackpool) is unnecessary disambiguation. If the correct title per naming conventions uses the "natural qualfier", the disambiguation project does not contradict that. -- JHunterJ (talk) 17:57, 12 April 2013 (UTC)

Are you saying South Pier (Blackpool) is disambiguation of South Pier, but South Pier, Blackpool is not disambiguation of South Pier? If so, I disagree, and point out that they are examples of the two forms of disambiguation: parenthetic and natural, and neither is necessary (thus both are unnecessary disambiguation). --B2C 18:36, 12 April 2013 (UTC)

I'm saying that the naming conventions might specify the longer version even when no ambiguity exists, and that does not contradict the disambiguation guidelines. If the naming conventions would normally use the short version, the natural qualifier might be preferred over the parenthesized qualifier. Of the set of titles with commas separating hierarchical levels, some are examples of natural disambiguation and some are examples of naming convention preferences. Akron redirects to Akron, Ohio, for instance. -- JHunterJ (talk) 19:55, 12 April 2013 (UTC)

If a naming convention specifies the longer version even when no ambiguity exists, the disambiguation is still unnecessary for the purposes of disambiguation. Yes, unnecessary disambiguation may be in compliance with disambiguation guideline wording, due to superfluous special-case rules created to rationalize certain overly precise titles, but it's still unnecessary disambiguation.

But this is all probably moot as I'm pretty sure there are no special naming conventions for naming piers, which would be unnecessary too! --B2C 21:04, 12 April 2013 (UTC)

There is no unnecessary disambiguation in "Akron, Ohio" (or, say, Prineville, Oregon). If the common name of or naming convention for "Prineville" is "Prineville, Oregon", then that's the undisambiguated title that should be used. If there is no ambiguity, then disambiguation (unnecessary or otherwise) is impossible. There is no need to describe the naming conventions as rationalizations or special-case, as if those qualities were faults. All naming conventions are rationalized, and they exist to handle special cases as well. -- JHunterJ (talk) 12:01, 13 April 2013 (UTC)

JHunterJ's comments on natural qualifiers, parenthetical or otherwise, are very interesting. It sounds natural and good. Is this concept documented elsewhere, or just what there is on thei project page? --SmokeyJoe (talk) 01:37, 13 April 2013 (UTC)
Yes, and I think I agree that a more natural way to add needed precision is better. Blackpool South Pier seems to be a very common, and perhaps official, name for this one, and most editors who put any value at all on precision recognize that precision as necessary to indicate the topic of the article, even if B2C doesn't. In that respect JHunterJ is correct that "a naming convention specifies the longer version", though I don't agree that it's a situation "when no ambiguity exists". It's terribly ambiguous, with so many places having a North and South Pier. Dicklyon (talk) 06:02, 13 April 2013 (UTC)
Happy we agree in the important points. Two notes on the side topics: (a) Parenthetical qualifiers are never natural. If parentheses are in a title, it's because the topic itself includes parentheses or because a qualifier was needed and a natural one wasn't identified or found suitable. (b) In this case, I took the opening comment that there is no other article "South Pier", so indeed no Misplaced Pages ambiguity exists if that's true. -- JHunterJ (talk) 11:51, 13 April 2013 (UTC)

My first instinct on seeing this discussion was that I have a hard time believing that Misplaced Pages only has one article that could reasonably be called South pier. A Wiki search confirmed this - there are places with names similar enough for confusion in Singapore, Michigan and a couple other places. I'll make a disambig. Ego White Tray (talk) 15:36, 13 April 2013 (UTC)

South pier and North Pier (disambiguation) created. Ego White Tray (talk) 16:39, 13 April 2013 (UTC)

Prefer natural disambiguation over paranthetical disambiguation

Should we prefer Mavia, Queen of the Tanukh over Mavia (queen)? Or Queen Mavia? --SmokeyJoe (talk) 08:54, 14 April 2013 (UTC)

I'm not seeing evidence JHunterJ's assertion of our preference for non-parenthetical natural disambiguation. The entries at Chris_Young_(disambiguation) show a natural tendency to parenthetical disambiguation. --SmokeyJoe (talk) 10:47, 16 April 2013 (UTC)

You misspelled "WP:PRECISION's assertion of our preference" there. -- JHunterJ (talk) 11:12, 16 April 2013 (UTC)

Don't you mean WP:NATURAL? Thanks for the pointer anyway. Why doesn't this guideline point to that policy section? Speaking of that policy section, it is not obvious that the three general methods are listed on order of preference. Looking at Chris Young for example, it looks like stub writers prefer parenthetical disambiguation. I would guess that writers of new articles look at existing dab pages for guidance, and I don't think anything here, or WP:AT, communicates a preference against parenthetical or comma disambiguation. --SmokeyJoe (talk) 12:09, 16 April 2013 (UTC)

A crucial, and sometimes overlooked, aspect of natural disambiguation is that it is an alternative name that the subject is also commonly called in English. In cases where the subject is not commonly known by an alternative name, parenthetical disambiguation may be preferred. To me, the criteria is something like this: is there another commonly used name that can be used and linked inline without appearing awkward or unnatural? If yes, use that, if not, use parenthetical disambiguation, except for some types of place names or nobility. For example, Chris Young (American football) could be titled as Christopher Lamont Young, but if he is not commonly referred to by that name, there is no benefit to readers. older ≠ wiser 12:26, 16 April 2013 (UTC)

so the following would be true? "The most common name for a subject, as determined by its prevalence in reliable English-language sources, is often used as a title because it is recognizable and natural. However, the second or third most common name may be used if the most common name is ambiguous with another topic, but the second or third is not. ". --SmokeyJoe (talk) 13:15, 16 April 2013 (UTC)

"However, another common name (less common, but still common) may be used if the most common name is ambiguous but the other common name is not." It doesn't have to win, place, or show, so long as it's common. -- JHunterJ (talk) 13:22, 16 April 2013 (UTC)

Right. The second most commonly used name would not be appropriate if it's only very rarely used to refer to the topic in question. --B2C 19:57, 16 April 2013 (UTC)

SmokeyJoe, I don't know, did you mean WP:NATURAL instead of mis-attributing it to me? (Looks like WP:NATURAL is a recent addition, from October 2012; I've been using WP:PRECISION as the shortcut for much longer.) This guideline doesn't specify which unique titles to select, as I have also said many, many times; the naming conventions and content projects specify those, as they like. The disambiguation guidelines specify how to name disambiguation pages. (I have no preference for natural disambiguation over parenthetical disambiguation, but I am happy to accept the consensus.) -- JHunterJ (talk) 12:59, 16 April 2013 (UTC)

Need advice/help with disambiguation page

It all started by searching for a medieval papal legate, which I knew only by his ecclesiastical name (in latin) "Pelagius". In the disambiguation page, his article was not there, so I assumed it was not created. I looked around the fifth crusade article and finally found his article under his native name of "Pelagio Galvani" (Pelagio is the Leonese version of "Pelagius" kind of like "Theodore" being an english version of the Greek "Theodoros".

So, I found out that under the disambiguation page of "Pelagius" there was a see also for "Pelagio" for those named as such. I am wondering what are the guidelines for disambiguation pages for cognates? For instance, should there just be one disambiguation page that redirects from "Pelagius" Pelagios" "Pelagio" and every other cognate, then listing the various people in sections by whichever name they are most known for? Because I almost was unable to find an article just because of there being three disambiguations for a single name.

I was thinking of even putting the Pelagius Galvani into the "Pelagius" disambiguation, thinking there would not be much harm in it being in both the disambiguation page for "Pelagius" and Pelagio". but I did not want to do this without seeking input from others. So, anyone have any comments/advice/ideas?75.73.114.111 (talk) 23:11, 13 April 2013 (UTC)

As it sounds helpful, I advise just doing it. If there were some reason to not do it, which I think is unlikely, then someone will say so. --SmokeyJoe (talk) 01:14, 14 April 2013 (UTC)
A line in Pelagius (disambiguation) like the below might work.

Pelagius or Pelagio Galvani (c. 1165—1230), Leonese cardinal, canon lawyer, papal legate and leader of the Fifth Crusade

Thanks, SchreiberBike (talk) 02:47, 14 April 2013 (UTC)

Sounds like a good plan, thank you for your help!75.73.114.111 (talk) 23:52, 15 April 2013 (UTC)

Agios dab page

I just came across Agios, which is tagged with {{disambiguation}} but does look much like anything described at WP:MOSDAB. An older version is perhaps a bit closer, though really it looks like most everything is a partial title match. Thoughts? older ≠ wiser 19:36, 16 April 2013 (UTC)

I usually just clean up such pages (and speedy or prod them as needed), or tag them for cleanup. If someone disagrees with the cleanup, then remove the dab tag. -- JHunterJ (talk) 19:46, 16 April 2013 (UTC)

Heh, I was just about to try to figure out/fix the Agios situation myself. It's certainly not a dab page currently. Theoldsparkle (talk) 14:07, 17 April 2013 (UTC)

First, the section title “Is this a reasonable dab page?” was silly, and I changed it. Second, there are many users who do not understand the difference between Misplaced Pages and Wiktionary: one part of them push dictionary explanations to Misplaced Pages; another one accuse perfectly encyclopedic topics in perceived “dictionaryness”. I erased abominations from there, and also from “agioi”. Third, I advise to post such cases to WT:WikiProject Disambiguation, not here. Incnis Mrsi (talk) 15:01, 17 April 2013 (UTC)

I'm not sure I see the changes made by User:Incnis Mrsi as a significant improvement. Look: are there any topics covered in any Misplaced Pages article that would reasonably be referred to as just "agios"? Sacred doesn't use the word, and I don't think we need a disambiguation page just to say what Foo means in another language. Agio is probably reasonable. If it were up to me I'd probably just delete the page and let anyone who's searching for the plural of agio find it at the top of the search results. Theoldsparkle (talk) 18:31, 17 April 2013 (UTC)

I still can’t realize this bizarre custom of en.WP to link a user’s nickname within few inches of his/her own signature. Incnis Mrsi (talk) 10:22, 18 April 2013 (UTC)

I would just delete it. It is a disambiguation page that doesn't disambiguate anything. bd2412 T 21:29, 17 April 2013 (UTC)

Although I do not insist strongly on keeping, IMHO this page can serve as a useful entry point to Wiktionary. You may initiate an AfD, but this is not a WP:CSD#G6 if only because of this discussion. Incnis Mrsi (talk) 10:22, 18 April 2013 (UTC)

Trudeau clan

Trudeau has six all in the same clan in Canada. Pierre is the most famous so far but one son may pass him. Should we give them all their own 'clan' section on the dab page under people? We may have to add any mountains or space craft possibly named after them in the future as well. The Kennedy dab may need a similar sub group but that is too far south for me.--Canoe1967 (talk) 02:42, 21 April 2013 (UTC)

Clan is certainly an odd choice of words - in this case, I would say it would be justified to create an article called Trudeau family, similar to the US articles, Kennedy family and Romney family. It would then look something like below:

Trudeau family, a notable Canadian family known in politics, business and film, including:
- Pierre Trudeau fifteenth Prime Minister of Canada
- Margaret Trudeau, his wife

etc...
Ego White Tray (talk) 02:49, 21 April 2013 (UTC)

I've changed the page from a disambiguation page to an anthroponymy list article. (No bearing on whether the Trudeau family article might still be created.) -- JHunterJ (talk) 03:00, 21 April 2013 (UTC)

Done--Canoe1967 (talk) 03:09, 21 April 2013 (UTC)

RFC-birth date format conformity when used to disambiguate

I have closed my own discussion at Misplaced Pages talk:Naming conventions (people)#RFC-birth date format conformity when used to disambiguate. Please feel free to take any action necessary to modify the closure of my own discussion to make it appear more Kosher.--TonyTheTiger (T/C/BIO/WP:CHICAGO/WP:FOUR) 21:59, 21 April 2013 (UTC)

Is there a primary topic for Limerick?

Talk:Limerick#Requested move 2013

--B2C 20:04, 22 April 2013 (UTC)

Proposal that designation of a primary topic should be inherited by subtopics.

Tibet is an article, and I have seen no effort to dispute that. However, up until I redirected it earlier this afternoon, Demographics of Tibet was a disambiguation page, pointing to the demographics section of Tibet, and the demographics section of Tibet Autonomous Region. It does not make sense to me that if "Foo" is a primary topic, then "Bar of Foo" should be anything but the primary topic, "Bar" of "Foo". There may be many places named "Paris", but Paris is primary, and therefore "Cuisine of Paris" should mean "Cuisine of Paris", and should not be a disambiguation page for the cuisines of various places coincidentally called Paris. If there are many cuisine articles for places by that name, they should be at a Cuisine of Paris (disambiguation). There are a number of places called "Florida" but Florida is primary, and therefore List of toll roads in Florida is about toll roads in Florida. Conversely, of course, if "Foo" is a disambiguation page, then "Cuisine of Foo" should also be a disambiguation page, provided that there are multiple uses of "Foo" for which a "Cuisine of" exists. For example, Georgia is a disambiguation page, and therefore Supreme Court of Georgia is appropriately a disambiguation page. Note that there is no page on the cuisine of the U.S. state of Georgia, so Cuisine of Georgia redirects to Georgian cuisine; there is no article on the lower courts of the country, Georgia, so Courts of Georgia redirects to Government of Georgia (U.S. state)#Judiciary.

I therefore propose that we establish as a formal rule that if "Foo" is a primary topic, "Bar of Foo" should be a primary topic page for "Bar" of "Foo", and not a disambiguation page for "Bar" of various possible meanings of "Foo", unless there is no "Bar of Foo" for the primary topic; and that if "Foo" is a disambiguation page, "Bar of Foo" should be a disambiguation page for "Bar" of various possible meanings of "Foo", unless there is only one "Bar of Foo" article to point to. bd2412 T 21:50, 22 April 2013 (UTC)

Category:

WikiProject Disambiguation pages