Wikipedia talk:Disambiguation: Difference between revisions
Content deleted Content added
→A current case in point: I agree "policy is at least a little but busted". I have long argued that there is a lack of logic in WP:PT, as written, and as practiced. It has more history and faithful adherents than underlying logic. I don't agree with repealing it entirely. I agree with User:Certes 12:20, 15 October 2018 "... we still need primary topics, though I agree that they should be used less often". |
|||
| Line 1,029: | Line 1,029: | ||
**'''Support''' "Titles should be precise enough to unambiguously define the scope of the article". But wouldn't this bring it into direct conflict with WP:P T, and '''perfectly''' into agreement with [[User:Andrewa/Primary Topic RfC]]? Because, that's '''exactly''' what I'm proposing there, with a plan of action to achieve this with a minimum of effort and inconvenience. Or that was my intent, anyway. |
**'''Support''' "Titles should be precise enough to unambiguously define the scope of the article". But wouldn't this bring it into direct conflict with WP:P T, and '''perfectly''' into agreement with [[User:Andrewa/Primary Topic RfC]]? Because, that's '''exactly''' what I'm proposing there, with a plan of action to achieve this with a minimum of effort and inconvenience. Or that was my intent, anyway. |
||
**''Move'' was a good close (in theory... it was bungled in practice, I'm discussing that with the page closer). There's no grounds for [[wp:MR|MR]]. None of this was raised in the discussion, I considered whether it was [[wp:pointy|pointy]] of me not to chime in but even if I had there was consensus to move IMO. The move was based on policy, I'd have needed to invoke IAR to oppose and IMO I can't validly do that until I've demonstrated some consensus for my concerns with WP:P T. And I don't think that has happened yet, but I'm hopeful it's not too far away. [[User:Andrewa|Andrewa]] ([[User talk:Andrewa|talk]]) 00:23, 16 October 2018 (UTC) |
**''Move'' was a good close (in theory... it was bungled in practice, I'm discussing that with the page closer). There's no grounds for [[wp:MR|MR]]. None of this was raised in the discussion, I considered whether it was [[wp:pointy|pointy]] of me not to chime in but even if I had there was consensus to move IMO. The move was based on policy, I'd have needed to invoke IAR to oppose and IMO I can't validly do that until I've demonstrated some consensus for my concerns with WP:P T. And I don't think that has happened yet, but I'm hopeful it's not too far away. [[User:Andrewa|Andrewa]] ([[User talk:Andrewa|talk]]) 00:23, 16 October 2018 (UTC) |
||
::I just searched Google and got both articles, as the third and fourth items. My understanding is article titles have little to no effect on Google results. [[User:Station1|Station1]] ([[User talk:Station1|talk]]) 06:40, 16 October 2018 (UTC) |
|||
'''A second challenge''': In terms of [[wp:silence|silence]] I think there's rough consensus here that this move was a very bad idea indeed. But I also think the move was completely justified in terms of policy. |
'''A second challenge''': In terms of [[wp:silence|silence]] I think there's rough consensus here that this move was a very bad idea indeed. But I also think the move was completely justified in terms of policy. |
||
Revision as of 06:40, 16 October 2018
| This is the talk page for discussing improvements to the Disambiguation page. |
|
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, 56Auto-archiving period: 60 days  |
 | The contentious topics procedure applies to this page. This page is related to the English Wikipedia article titles policy and Manual of Style, which has been designated as a contentious topic. Editors who repeatedly or seriously fail to adhere to the purpose of Wikipedia, any expected standards of behaviour, or any normal editorial process may be blocked or restricted by an administrator. Editors are advised to familiarise themselves with the contentious topics procedures before editing this page. |
 | Disambiguation | |||
| ||||
 | Threads older than 60 days may be archived by Lowercase sigmabot III. |
RfC: INTDAB links to non-dab pages
- The following discussion is an archived record of a request for comment. Please do not modify it. No further edits should be made to this discussion. A summary of the conclusions reached follows.
Despite the amount written , this RFC was only about whether redirects that end in "(disambiguation)" should be allowed to target non-disambiguation pages. While numerically there is no clear "winner", many of those opposing based their arguments solely on the fact that set-index articles or name lists are not disambiguation pages and that this might be confusing to some readers to be redirected as such. However, when challenged whether readers have previously indicated such confusion, there was no further evidence provided. Hence, the argument of "misleading", "surprise" or not "least astonishment" was not as convincing. The discussion got sidetracked quite a bit about the difference between SIA and DAB pages but without any real reason since no one actually doubted that DAB and SIA are different. Furthermore, despite multiple challenges, these opposers have not demonstrated any actual need to remove such redirects.
In the end, when considering the policy-base of the arguments for or against such redirects, one has to also remember that redirects are governed by the WP:R guideline. Judging the arguments based on WP:R#DELETE and WP:R#KEEP, the main argument made in favor of deletion could be summarized under DELETE #2, potential for confusion. However, as pointed out multiple times, the risk of confusion is minimal because the distinction between DAB, SIA and name-lists is one most readers are unlikely to know about and even if they do, confusion could only happen if they are redirected from a "(disambiguation)" page to a SIA instead of the correct DAB page. But that can be dealt with by retargeting. On the other hand, a number of users have argued that these redirects are useful, which is a reason to keep them per KEEP #5.
So to answer the question of this RFC: Consensus is (slightly) in favor of having redirects that end in "(disambiguation)" that target SIA or name lists in general, although single examples might exist where the redirect should be deleted. The question whether such redirects should be created was not part of this RFC, although there was no consensus for that amongst those who debated it nonetheless. Consensus might change if and when those opposing such redirects can provide actual examples of users being confused by such redirects. Regards SoWhy 18:02, 23 September 2018 (UTC)Intro and !votes
Should there be Foo (disambiguation) redirects that target non-disambiguation pages that contain a list of articles, for instance where Foo is a {{Surname}} page, or {{Set index article}}, or a list article? Shhhnotsoloud (talk) 07:01, 26 June 2018 (UTC)
Extended content
| ||
|---|---|---|
Questions and answersThere is a lot of repetition and misunderstanding above, this is an attempt to clarify what is actually being proposed:
Thryduulf (talk) 14:23, 29 June 2018 (UTC)
Maybe some have been proposed before, I don't know. Widefox; talk 10:10, 2 July 2018 (UTC)
Proposed mitigation 2If the answer to this RfC is Yes then I propose the following guidance:
This mitigation is offered in the spirit of compromise, taking into account that some contributors to the discussion think users find such redirects useful, but others think they are not necessary and their existence creates confusion. Short support/oppose comments (perhaps with improvements) are welcome, without re-running the arguments above. Shhhnotsoloud (talk) 07:35, 9 July 2018 (UTC)
Proposed mitigation 3I propose the following guidance:
|
The discussion above is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.
Without an article
Is a disambiguation page necessary if it consists of multiple entries without a Wikipedia article that is directly related to the disambiguated term? E.g. High on Life consists of multiple entries but none of which has a Wikipedia article. The editor whose username is Z0 06:51, 27 July 2018 (UTC)
- Suppose you remember a song title called “High on Life”, but you don’t remember who made it or when. That page would present you with possibilities to explore, even if there is no stand-alone article on the song you’re thinking of. That’s a perfectly fine disambiguation page, as long as there’s a blue-linked article for each entry that at least mentions the disambiguated term. — Gorthian (talk) 07:33, 27 July 2018 (UTC)
- @Gorthian: Sounds reasonable but that could lead to the addition of non-notable (or promotional) content. I mean if it's okay to add anything without a WP article, that's going to create a mess. In the RFD for High on Life, I brought it up but editors seem to agree with you that disambiguation pages could be created even when the entries don't consist of a correlated article. The editor whose username is Z0 09:49, 27 July 2018 (UTC)
- It's not a disambiguation page, it's a redirect that has been hijacked. The redirect is under discussion at Wikipedia:Redirects for discussion/Log/2018 July 25#High on Life, please respect WP:MULTI. --Redrose64 🌹 (talk) 11:17, 27 July 2018 (UTC)
- And please see MOS:DABMENTION, the part that covers those entries while prohibiting non-notable ones. -- JHunterJ (talk) 12:39, 27 July 2018 (UTC)
- Is it possible for someone to clarify in the guideline page that creating a disambiguation page based on just 2-3 entries that do not have a WIkipedia article matching the term, is acceptable? Courtesy ping @Ss112: @Hayman30: @Station1: The editor whose username is Z0 08:14, 2 August 2018 (UTC)
- WP:DABSTYLE: "Each bulleted entry should have a navigable (blue) link, normally as the entry itself (see the previous bullet), or in the description if the entry is red-linked or unlinked.
- Rarely should a bulleted entry have more than one navigable link; including more than one link can confuse the reader." I believe that covers it. Nobody's saying or going to promote the idea that creating a disambiguation page where there is nothing navigable for any of the entries is acceptable. Ss112 08:20, 2 August 2018 (UTC)
- Some editors think there should not be a dab page if there are zero articles on WP for a term, which indicates the guideline isn't clear enough. It should be amended to include a list of requirements for creating a disambiguation page just for clarity. The editor whose username is Z0 09:56, 2 August 2018 (UTC)
- Yeah, that would probably be harmless and likely helpful. I would watch out that the summary does not subtly but substantively change the requirements, though (intentionally or otherwise). — SMcCandlish ☏ ¢ 😼 04:17, 3 August 2018 (UTC)
- Some editors think there should not be a dab page if there are zero articles on WP for a term, which indicates the guideline isn't clear enough. It should be amended to include a list of requirements for creating a disambiguation page just for clarity. The editor whose username is Z0 09:56, 2 August 2018 (UTC)
- Is it possible for someone to clarify in the guideline page that creating a disambiguation page based on just 2-3 entries that do not have a WIkipedia article matching the term, is acceptable? Courtesy ping @Ss112: @Hayman30: @Station1: The editor whose username is Z0 08:14, 2 August 2018 (UTC)
- @Gorthian: Sounds reasonable but that could lead to the addition of non-notable (or promotional) content. I mean if it's okay to add anything without a WP article, that's going to create a mess. In the RFD for High on Life, I brought it up but editors seem to agree with you that disambiguation pages could be created even when the entries don't consist of a correlated article. The editor whose username is Z0 09:49, 27 July 2018 (UTC)
RFC
- I propose adding the following (in bold) to Wikipedia:Disambiguation#Disambiguation_pages: "...focused on getting the reader to their desired article. A disambiguation page is warranted when there are two or more blue-linked entries and there is no primary topic." The editor whose username is Z0 07:58, 3 August 2018 (UTC)
- Comment Should we amend the wording to state explicitly that the blue link may occur anywhere in the entry? That would clarify that items such as
- High on Life, an album by QFX
- on dab High on Life do count as blue-linked entries, even though the initial words aren't linked. — Certes (talk) 12:27, 3 August 2018 (UTC)
- Sounds reasonable. The editor whose username is Z0 15:00, 8 August 2018 (UTC)
- Yes; that definitely qualifies. — SMcCandlish ☏ ¢ 😼 15:51, 3 August 2018 (UTC)
- But this questions is already treated, with much more detail, in the section Wikipedia:Disambiguation#Disambiguation page or hatnotes? which is further up the page. – Uanfala (talk) 12:53, 3 August 2018 (UTC)
- That section is about primary topic and suggests a disambiguation page may be created when there is no primary topic, but it does not say exactly what is needed to warrant a disambiguation page. Hence, causing some editors (1) to think a dab shouldn't be created if two or more title-matching WP articles do not exist. In some way I agree with their sentiment that non-notable entries shouldn't be added but then it could be helpful, so why not? @Uanfala: The editor whose username is Z0 14:58, 8 August 2018 (UTC)
- May be better to cross-reference that section, though it doesn't actually hurt to reiterate the central matter. We have to remember that people enter these page sections via shortcut links and do not read them top-to-bottom. So, something like: "... is warranted when there are two or more entries with blue links and there is no primary topic ." No need for boldfacing. — SMcCandlish ☏ ¢ 😼 15:51, 3 August 2018 (UTC)
- That could work. The editor whose username is Z0 15:01, 8 August 2018 (UTC)
- "Disambiguation in Wikipedia is the process of resolving conflicts that arise when a potential article title is ambiguous..." Non-notable topics are not potential article titles, and therefore should not appear on dab pages. Dab pages were never intended to be indexes or search engines. In fact, a bad dab page can get in the way of the search engine. DABMENTION always used to be understood to be for cases where a notable topic was covered in another article. It's only recently that a few editors have been adding non-notable songs that often are barely mentioned at all in the article, sometimes with no more info than what appears on the dab page itself. If we start adding non-notable entries, we would need to add hatnotes or dab pages for every spouse of a state legislator who happens to be mentioned in passing and shares a name with a notable person, not to mention every Main Street and every bit actor and fictional character who shows up in a cast list. I recommend we change "topic" to "notable topic" in DABMENTION. Station1 (talk) 19:27, 4 August 2018 (UTC)
- That's one interpretation. But the other (and mine) is that any non-notable information should be removed from the article that mentions it. Disambiguation pages are not the place to make the value judgment of when a mention in the encyclopedia is notable and when it isn't. If any given song title were unambiguous, would it be a possible {{R from song}}? Yes. So if ambiguity exists instead, it should be listed on the disambiguation page. If it's truly non-notable, it shouldn't be mentioned (or "noted") in the encyclopedia. -- JHunterJ (talk) 12:16, 10 August 2018 (UTC)
- And there is a third view, the one that's practiced by the majority of wikipedians: whether to create a redirect or add an entry to a dab page is a decision whose basis lies somewhere between the two extremes of notability ( = deserving a separate article) and noteworthiness ( = deserving a mention within an article). If we have meaningful encyclopedic content on a topic, then it doesn't matter whether the topic is treated in an article of its own, or together with other topics as part of an artile – this warrants having an entry on a dab page. Conversely, every reasonably developed article is likely to make passing mentions of a large number of entities, about which we will never have any content apart from these passing mentions – almost no-one would dream of adding these to dab pages. – Uanfala (talk) 13:07, 10 August 2018 (UTC)
- I suppose that's effectively my view as well. If it could have been a redirect except for the ambiguity, then it can be on the disambiguation page with the the ambiguity, because that's the technical problem disambiguation pages exist for in the first place. The passing mentions that no one would think to add to dab pages (or think to put into the Search Wikipedia box) wouldn't be redirects or disambiguation page entries. But song on albums would be things that people with think to make redirects for (and think to put into the Search Wikipedia box), so would be things people would think to put on disambiguation pages. Certainly if no one thinks to put it on a disambiguation page, the question of whether to remove it from that disambiguation page won't come up. -- JHunterJ (talk) 15:49, 10 August 2018 (UTC)
- I agree that we should use the same criterion for creating redirects as for adding dab entries. The disagreement, I guess, comes from the application: I don't think that a song that is merely mentioned in an article about an album warrants a redirect. If we don't have any meaningful content about the song, and if we can't expect out article to ever say anything about it other than the fact that it was on a certain album, then this is not the kind of thing that readers will be putting in the search box and it is not the kind of thing we would want to integrate in the encyclopedia's navigational system. – Uanfala (talk) 20:34, 10 August 2018 (UTC)
- Do such redirects, when created, get deleted through RfD? (I don't know, I don't follow RfD closely, but the existence of {{R from song}} indicates to me that they're viable without expectation of going anywhere else.) -- JHunterJ (talk) 20:44, 10 August 2018 (UTC)
- WP:R#PLA seems to indicate that that there should generally be content at the target but unlike WP:DABMENTION doesn't appear to mandate it. I'd suggest treating the criteria for a redirect as the same as DABMENTION. Crouch, Swale (talk) 12:57, 29 August 2018 (UTC)
- Do such redirects, when created, get deleted through RfD? (I don't know, I don't follow RfD closely, but the existence of {{R from song}} indicates to me that they're viable without expectation of going anywhere else.) -- JHunterJ (talk) 20:44, 10 August 2018 (UTC)
- I agree that we should use the same criterion for creating redirects as for adding dab entries. The disagreement, I guess, comes from the application: I don't think that a song that is merely mentioned in an article about an album warrants a redirect. If we don't have any meaningful content about the song, and if we can't expect out article to ever say anything about it other than the fact that it was on a certain album, then this is not the kind of thing that readers will be putting in the search box and it is not the kind of thing we would want to integrate in the encyclopedia's navigational system. – Uanfala (talk) 20:34, 10 August 2018 (UTC)
- I suppose that's effectively my view as well. If it could have been a redirect except for the ambiguity, then it can be on the disambiguation page with the the ambiguity, because that's the technical problem disambiguation pages exist for in the first place. The passing mentions that no one would think to add to dab pages (or think to put into the Search Wikipedia box) wouldn't be redirects or disambiguation page entries. But song on albums would be things that people with think to make redirects for (and think to put into the Search Wikipedia box), so would be things people would think to put on disambiguation pages. Certainly if no one thinks to put it on a disambiguation page, the question of whether to remove it from that disambiguation page won't come up. -- JHunterJ (talk) 15:49, 10 August 2018 (UTC)
- And there is a third view, the one that's practiced by the majority of wikipedians: whether to create a redirect or add an entry to a dab page is a decision whose basis lies somewhere between the two extremes of notability ( = deserving a separate article) and noteworthiness ( = deserving a mention within an article). If we have meaningful encyclopedic content on a topic, then it doesn't matter whether the topic is treated in an article of its own, or together with other topics as part of an artile – this warrants having an entry on a dab page. Conversely, every reasonably developed article is likely to make passing mentions of a large number of entities, about which we will never have any content apart from these passing mentions – almost no-one would dream of adding these to dab pages. – Uanfala (talk) 13:07, 10 August 2018 (UTC)
- That's one interpretation. But the other (and mine) is that any non-notable information should be removed from the article that mentions it. Disambiguation pages are not the place to make the value judgment of when a mention in the encyclopedia is notable and when it isn't. If any given song title were unambiguous, would it be a possible {{R from song}}? Yes. So if ambiguity exists instead, it should be listed on the disambiguation page. If it's truly non-notable, it shouldn't be mentioned (or "noted") in the encyclopedia. -- JHunterJ (talk) 12:16, 10 August 2018 (UTC)
- Disambiguation pages when there are only two pages is bloat. Totally unnecessary, even if there is no clear primary topic. Choose the one that was created first if you have to, but put one at the base name, and a dab link from it to the other one at disambiguated title. A dab page with two entries serves no purpose. This is what WP:TWODABS used to say, and that guidance should be restored. --В²C ☎ 18:33, 10 August 2018 (UTC)
- A dab page with only two entries when neither of them is primary serves the purpose of disambiguating two topics when neither of them is primary. WP:TWODABS, when it said that, did not have consensus. -- JHunterJ (talk) 19:04, 10 August 2018 (UTC)
- But User:Born2cycle is right though. It's bloat because instead of OneTopicX and OtherTopixX (2 pages) we have TopicX, TopicX (disambiguation) (an INTDAB redirect), OneTopicX and OtherTopixX (4 pages). Not only are there now twice as many pages to maintain, but every searcher is a loser. If a user searches for "TopicX" when they wanted OneTopicX, they are diverted via the dab page. If a user searches for "TopicX" when they wanted OtherTopicX, they are diverted via the dab page. 100% of searchers have to click through a dab page. Any mechanism to determine a primary topic (page views, Google hits, page creation date, toss of a coin) and the consequent use of hatnotes is better than a 2 dab page with no primary topic. (That's clearly my opinion, not what the guidance says). Shhhnotsoloud (talk) 10:20, 11 August 2018 (UTC)
- But that's wrong, because it's another click but a needed click because neither topic is primary. There are twice as many pages to "maintain", but of course "maintaining" a two-entry dab and its redirect is pretty light work (until there are 3+, and then it's no longer the subject of this discussion). And every disambiguation page without a primary topic means that every searcher arriving is a "loser" in your terms; should we randomly or otherwise compel a primary topic for every title? Since we're not ready to abolish base-name disambiguation pages in the general case, there's no reason to abolish them in the two-entry case. If one of the two entries is primary, great; otherwise, there's no problem here. -- JHunterJ (talk) 12:19, 13 August 2018 (UTC)
- Borderline primary topics already inconvenience readers when wikilinks to the wrong targets creep in. (For example, I recently diverted several soldiers from a company to a company (military unit), and species from a family to a family (biology).) Such errors would be more widespread if we put ambiguous topics at the base name just because there is only one other meaning. Certes (talk) 14:43, 13 August 2018 (UTC)
- JHunterJ, when there are only two topics and neither is clearly primary than both are close to being primary. That is, about half the people searching with that name are seeking one or the other. An article with a hat link to the other article serves as a disambiguation function just as well as a dab page for disambiguating. The distinction between a primary topic and a nearly-primary topic is, by definition, minor. So treating either nearly-primary topic as primary should not be a big deal. But, in the general case, with more than two entries, the odds of landing someone on the sought page drops from about 50% rapidly to around 33%, 25% and lower, and now the "losers" are usually redirected to a dab page from which they have to click again to get to their desired page. That's why TWODABS was originally treated as a special case. I don't think most participants understood and appreciated this critical distinction when it was changed. Certes, yes, topics that are nearly-primary have the same problem as topics that are actually primary - the possibility of linking to the wrong topic exists when you have an article rather than a dab page at a base name. I've seen others use this as an argument to not have primary topics at all. I think the nearly-primary topic in a TWODABS case having this issue is just a minor side issue. If we tolerate it for all genuine primary topics, tolerating it for nearly primary topics as well is relatively minor. --В²C ☎ 18:16, 13 August 2018 (UTC)
- Born2cycle, when there are only two topics and neither is primary (clearly primary is just primary), then both are far from being primary. That is, neither is close to being much more likely than any other single topic (the other single topic), since we just stipulated that they are close to each other. The distinction between being a primary topic and a non-primary topic (there is no such thing as "nearly primary") is, by definition, major. -- JHunterJ (talk) 18:48, 13 August 2018 (UTC)
- JHunterJ, okay, point taken, but I think we need to reasonably interpret that wording in a 2-dab situation. In a 3-dab situation we could easily have a primary topic having 51% of the likelihood being hit (e.g. 51/25/24). But in a 2-dab 51/49 situation the same 51% likelihood is not good enough to be PT? I don't see how it makes sense that a topic with 51% likelihood can be primary topic if it's the only use, or one of three or more uses, but not if there are exactly two uses. Seems absurd to me.
Okay, I know, in a 3-dab 51/48/1 situation the 51 also wouldn't technically meet the PT criteria, but I really don't get the point. I mean, the whole point of primary topics is to take people directly to the article they are likely to be seeking, and to minimize those who are not immediately taken to the article they are seeking. I guess it comes down to whether one thinks being sent to a different article (with a hatnote to the article being sought) is less desirable than being taken to a dab page. I see those two results as being functionally equivalent in terms of desirability. I personally don't mind being taken to the wrong article if that article has a hat link to the article I am seeking; I mean, that's no worse than being taken to a dab page. Or, if you prefer, they are equally annoying. So I don't get the motivation to prefer sending everyone to a dab page with links to their sought articles rather than taking about half to the article they seek and the other half to an article with links to their sought articles (hatnote macros support links to multiple alternatives and thus can serve as mini dab pages at the top of any article at a base name). --В²C ☎ 20:46, 13 August 2018 (UTC)
- "wouldn't technically meet the PT critera" = "wouldn't be a primary topic". "Technically" is unneeded there. See Certes' note of 14:43, 13 August 2018 (UTC) above for part of the point. -- JHunterJ (talk) 12:20, 14 August 2018 (UTC)
- Yes, but we can still treat a nearly-primary topic as if it was primary and put it at the base name. As to Certes' point, I addressed that already, just above, in the latter part of my 18:16, 13 August 2018 comment. I, for one, think serving readers better is more important than making it easier to identify some very small percentage of imperfect links. Taking about half of those searching with a given term directly to the article they seek is better than taking none to the article they seek, even if it also means taking the half not seeking that article there. That half is no worse off than they would be had they been taking to a dab page; either way they are just one easy click to the article they seek, while the first half is clearly worse off at the dab page than if taken directly to the article they sought. Similarly, for the imperfect links that take a user to the wrong one of two nearly-primary topics, that user is also only one hatnote click from the right article, which is why I refer to the original link as being merely "imperfect".
More generally, why the affinity for dab pages? Any search that lands a user on a dab page is a failure in our system. Note that Google searches don't take users to dab pages because they know that's not what users are seeking. For example, enter "obama presidential campaign" at Google.com and click on feeling lucky (their equivalent of our "Go"). You don't get taken to our TWODABS dab page; you get taken to Barack Obama presidential campaign, 2012. But in WP search hitting Go on the same search string takes us to the TWODABS dab page at Barack Obama presidential campaign. Putting one of two nearly-primary topics at the ambiguous base name helps us avoid sending users unnecessarily to dab pages like this - that's a good thing. --В²C ☎ 17:29, 14 August 2018 (UTC)
- There are two things that are good to avoid:
- Unnecessarily sending users to disambiguation pages
- Unnecessarily sending users to articles about topics they weren't seeking
- The primary topic guidelines, including those for base-name two-element disambiguation pages, balance those two. Landing on a two-element disambiguation page when neither is primary is not unnecessarily landing there. -- JHunterJ (talk) 19:52, 14 August 2018 (UTC)
- Landing on a two-element dab page when neither is primary is unnecessarily landing there, for at least about half of those that land there. It's unnecessarily landing there for them because if we put the one of the two articles they are seeking at the basename, they will land on the article they are seeking. As to the second half, the only alternative is also "good to avoid", unnecessarily landing on an article they're not seeking, which is no more important to avoid than unnecessarily landing on a dab page, so it's a wash from their perspective. That is, the second half is unnecessarily landing on one or the other; it's unnecessarily either way because no matter which it is, it could be the other one. However, it is necessary to land the second half on the article they are not seeking in order to improve the experience of the users in the first half, while not compromising the experience of the second half. When we can reduce the number of users unnecessarily landing somewhere good to avoid by about half of those in question, that seems like an obvious improvement to the encyclopedia user experience. After all, the users in the second half for a given topic might very well be in the first half when searching with other terms with two uses. On these encyclopedia-improving grounds, I think a strong WP:IAR argument can be made for ignoring the current wording at WP:PRIMARYTOPIC and WP:TWODABS in these two-near-primary-topic situations. Of course, it would be better to make explicit changes in the relevant policy and guidelines accordingly. --В²C ☎ 21:13, 14 August 2018 (UTC)
- I suggest that sending a reader to the wrong page is worse than sending them to a dab. They may not realise that they're on the wrong page, even with a hatnote (which won't appear on all devices). Articles also tend to take more data and time to load than dabs. No one can quantify this difference and say that sending one reader to the wrong page is as bad as sending exactly 3.142 readers to a dab, but there will be some such ratio and I suggest that . Let be the proportion of readers seeking the most popular topic (best candidate for PT). Then a primary topic page will cause no inconvenience to of the readers, and times more inconvenience than a dab for the other . So we should prefer a dab where , i.e. . For example, if you think that (a number I plucked out of thin air) then we should have a dab page when we estimate that fewer than of readers seek the primary topic. Of course, there may be other factors to consider when making this editorial judgement, but that logic may be useful for simple cases. Certes (talk) 00:07, 15 August 2018 (UTC)
- Great analysis. Thank you! We disagree about sending a user to the wrong page being significantly worse than sending them to a dab (assuming the "wrong page" has a hat note link to the "right" page). Or, if you prefer, I think or so. But as long as then placing the article for either of two topics at the base name is at worst a wash if both of their values are near , which is the situation we're addressing, and is progressively better than a wash the lower below the value of is, and the higher above the value of is for the article placed at the base name in a given case (which it will be in most cases - both topics being exactly is highly unlikely). --В²C ☎ 17:07, 15 August 2018 (UTC)
- I suggest that sending a reader to the wrong page is worse than sending them to a dab. They may not realise that they're on the wrong page, even with a hatnote (which won't appear on all devices). Articles also tend to take more data and time to load than dabs. No one can quantify this difference and say that sending one reader to the wrong page is as bad as sending exactly 3.142 readers to a dab, but there will be some such ratio and I suggest that . Let be the proportion of readers seeking the most popular topic (best candidate for PT). Then a primary topic page will cause no inconvenience to of the readers, and times more inconvenience than a dab for the other . So we should prefer a dab where , i.e. . For example, if you think that (a number I plucked out of thin air) then we should have a dab page when we estimate that fewer than of readers seek the primary topic. Of course, there may be other factors to consider when making this editorial judgement, but that logic may be useful for simple cases. Certes (talk) 00:07, 15 August 2018 (UTC)
- Landing on a two-element dab page when neither is primary is unnecessarily landing there, for at least about half of those that land there. It's unnecessarily landing there for them because if we put the one of the two articles they are seeking at the basename, they will land on the article they are seeking. As to the second half, the only alternative is also "good to avoid", unnecessarily landing on an article they're not seeking, which is no more important to avoid than unnecessarily landing on a dab page, so it's a wash from their perspective. That is, the second half is unnecessarily landing on one or the other; it's unnecessarily either way because no matter which it is, it could be the other one. However, it is necessary to land the second half on the article they are not seeking in order to improve the experience of the users in the first half, while not compromising the experience of the second half. When we can reduce the number of users unnecessarily landing somewhere good to avoid by about half of those in question, that seems like an obvious improvement to the encyclopedia user experience. After all, the users in the second half for a given topic might very well be in the first half when searching with other terms with two uses. On these encyclopedia-improving grounds, I think a strong WP:IAR argument can be made for ignoring the current wording at WP:PRIMARYTOPIC and WP:TWODABS in these two-near-primary-topic situations. Of course, it would be better to make explicit changes in the relevant policy and guidelines accordingly. --В²C ☎ 21:13, 14 August 2018 (UTC)
- There are two things that are good to avoid:
- Yes, but we can still treat a nearly-primary topic as if it was primary and put it at the base name. As to Certes' point, I addressed that already, just above, in the latter part of my 18:16, 13 August 2018 comment. I, for one, think serving readers better is more important than making it easier to identify some very small percentage of imperfect links. Taking about half of those searching with a given term directly to the article they seek is better than taking none to the article they seek, even if it also means taking the half not seeking that article there. That half is no worse off than they would be had they been taking to a dab page; either way they are just one easy click to the article they seek, while the first half is clearly worse off at the dab page than if taken directly to the article they sought. Similarly, for the imperfect links that take a user to the wrong one of two nearly-primary topics, that user is also only one hatnote click from the right article, which is why I refer to the original link as being merely "imperfect".
- "wouldn't technically meet the PT critera" = "wouldn't be a primary topic". "Technically" is unneeded there. See Certes' note of 14:43, 13 August 2018 (UTC) above for part of the point. -- JHunterJ (talk) 12:20, 14 August 2018 (UTC)
- JHunterJ, okay, point taken, but I think we need to reasonably interpret that wording in a 2-dab situation. In a 3-dab situation we could easily have a primary topic having 51% of the likelihood being hit (e.g. 51/25/24). But in a 2-dab 51/49 situation the same 51% likelihood is not good enough to be PT? I don't see how it makes sense that a topic with 51% likelihood can be primary topic if it's the only use, or one of three or more uses, but not if there are exactly two uses. Seems absurd to me.
- Born2cycle, when there are only two topics and neither is primary (clearly primary is just primary), then both are far from being primary. That is, neither is close to being much more likely than any other single topic (the other single topic), since we just stipulated that they are close to each other. The distinction between being a primary topic and a non-primary topic (there is no such thing as "nearly primary") is, by definition, major. -- JHunterJ (talk) 18:48, 13 August 2018 (UTC)
- JHunterJ, when there are only two topics and neither is clearly primary than both are close to being primary. That is, about half the people searching with that name are seeking one or the other. An article with a hat link to the other article serves as a disambiguation function just as well as a dab page for disambiguating. The distinction between a primary topic and a nearly-primary topic is, by definition, minor. So treating either nearly-primary topic as primary should not be a big deal. But, in the general case, with more than two entries, the odds of landing someone on the sought page drops from about 50% rapidly to around 33%, 25% and lower, and now the "losers" are usually redirected to a dab page from which they have to click again to get to their desired page. That's why TWODABS was originally treated as a special case. I don't think most participants understood and appreciated this critical distinction when it was changed. Certes, yes, topics that are nearly-primary have the same problem as topics that are actually primary - the possibility of linking to the wrong topic exists when you have an article rather than a dab page at a base name. I've seen others use this as an argument to not have primary topics at all. I think the nearly-primary topic in a TWODABS case having this issue is just a minor side issue. If we tolerate it for all genuine primary topics, tolerating it for nearly primary topics as well is relatively minor. --В²C ☎ 18:16, 13 August 2018 (UTC)
- But User:Born2cycle is right though. It's bloat because instead of OneTopicX and OtherTopixX (2 pages) we have TopicX, TopicX (disambiguation) (an INTDAB redirect), OneTopicX and OtherTopixX (4 pages). Not only are there now twice as many pages to maintain, but every searcher is a loser. If a user searches for "TopicX" when they wanted OneTopicX, they are diverted via the dab page. If a user searches for "TopicX" when they wanted OtherTopicX, they are diverted via the dab page. 100% of searchers have to click through a dab page. Any mechanism to determine a primary topic (page views, Google hits, page creation date, toss of a coin) and the consequent use of hatnotes is better than a 2 dab page with no primary topic. (That's clearly my opinion, not what the guidance says). Shhhnotsoloud (talk) 10:20, 11 August 2018 (UTC)
- A dab page with only two entries when neither of them is primary serves the purpose of disambiguating two topics when neither of them is primary. WP:TWODABS, when it said that, did not have consensus. -- JHunterJ (talk) 19:04, 10 August 2018 (UTC)
two dabs history
- FWIW, WP:TWODABS "used to say" exactly this, that when neither topic was considered primary, a disambiguation page is used.[8] If someone else wants to identify whatever other language was put in there after what it used to say, we can go look to see if there was any consensus for that change. I believe there was not. -- JHunterJ (talk) 18:54, 13 August 2018 (UTC)
- This change and this discussion seem relevant. Certes (talk) 19:17, 13 August 2018 (UTC)
- This is what I mean by what it used to say, and what I think should be restored: "If there are two topics for a term but neither is considered the primary topic, then a disambiguation page may be used; an alternative is to set up a redirect from the term to one of the topics, and use disambiguation links only." [9] --В²C ☎ 20:58, 13 August 2018 (UTC)
- Let's highlight that appropriately: "If there are two topics for a term but neither is considered the primary topic, then a disambiguation page may be used; an alternative is to set up a redirect from the term to one of the topics, and use disambiguation links only." -- JHunterJ (talk) 12:20, 14 August 2018 (UTC)
- Yes, that's the situation we are talking about: two topics neither of which is primary. What TWODABS used to say was that a dab page may (not must) be used. --В²C ☎ 16:48, 14 August 2018 (UTC)
- Hmm, the concept was apparently originally introduced at MOSDABS on July 21, 2007, and there it clearly applied only when one of the two was primary[10]. --В²C ☎ 17:00, 14 August 2018 (UTC)
- Let's highlight that appropriately: "If there are two topics for a term but neither is considered the primary topic, then a disambiguation page may be used; an alternative is to set up a redirect from the term to one of the topics, and use disambiguation links only." -- JHunterJ (talk) 12:20, 14 August 2018 (UTC)
- This is what I mean by what it used to say, and what I think should be restored: "If there are two topics for a term but neither is considered the primary topic, then a disambiguation page may be used; an alternative is to set up a redirect from the term to one of the topics, and use disambiguation links only." [9] --В²C ☎ 20:58, 13 August 2018 (UTC)
- Personally per WP:2DABPRIMARY I would give a lower threshold to PT status since the DAB means readers seeking either have a click, rather than just those looking for the alternative topic. However I would note that DAB pages are easier to load and probably should be used in situations where its unclear which topic is primary. Crouch, Swale (talk) 07:30, 28 August 2018 (UTC)
1
Is there consensus to make the proposed change? The editor whose username is Z0 04:49, 26 August 2018 (UTC)
Redlinks
What is the position on redlinks on Dab pages?
The WP:DABNOT section seems to suggest that titles shouldn't be included if there isn't an article to disambiguate from: OTOH the Urney example at WP:PRIMARYTOPIC, and WP:DABRED, say that they are fine. (Though the Urney page looks more like a set index; is there a better example?). It came up in conversation here, and I've had redlinks removed from dab pages before, “as they don't belong”, so I'm asking; are they generally OK, or are they not really a good idea? Moonraker12 (talk) 19:42, 13 August 2018 (UTC)
- Moonraker12, details of when red links are allowed are at WP:DABRL. Basically, a red link is okay as long as there is also a blue link in the line to an article where it is also a red link, and (we hope) some information about it. — Gorthian (talk) 19:50, 13 August 2018 (UTC)
- Gorthian: So, no problem, then? fair enough, I'd got the impression they were a bit frowned upon. Thanks for replying so quickly, Moonraker12 (talk) 19:56, 13 August 2018 (UTC)
- Moonraker12: Actually,I answered a little hastily above, not realizing we were citing the same section. WP:DABRELATED says
Include articles only if the term being disambiguated is actually described in the target article.
. There doesn’t need to be a whole article about it, just somewhere it is “described”. (But in practice, sometimes “described” is interpreted as “mentioned”, as in songs on an album’s track list.) — Gorthian (talk) 20:03, 13 August 2018 (UTC)- Gorthian: Ah! So we are on the same page! Moonraker12 (talk) 20:18, 13 August 2018 (UTC)
- Moonraker12: Actually,I answered a little hastily above, not realizing we were citing the same section. WP:DABRELATED says
- Gorthian: So, no problem, then? fair enough, I'd got the impression they were a bit frowned upon. Thanks for replying so quickly, Moonraker12 (talk) 19:56, 13 August 2018 (UTC)
Yeah, as the guidelines linked above say, redlinks are fine as long as we've got the topic covered somewhere. The emphasis here is on "topic" and "covered", rather than "term" and "mentioned": in my opinion (with which a small number of dab regulars disagree) is that if the given term is only mentioned in a target article and there isn't likelihood there will ever be any content about it beside this mention, then the term does not belong on the dab page. On the other hand – and I suspect more people will disagree with me on this point – there are situations in which we ought to add an entry even if it's not described or mentioned anywhere. For example, imagine that you're looking for an Eastern European village named "Bukovets". If you look at this version of the dab page, which lists the three villages for which we have articles, all of them in Bulgaria, you might conclude that your village is among those three and that it is located in Bulgaria. However, it happens to be the case that there are several villages with the same name in Ukraine, and even though we don't seem to have either articles or "mentions" about them anywhere yet, it nevertheless makes sense to include them on the dab page (as in this version), to make readers aware that the village they're looking for might not be among the ones we currently happen to have content about. – Uanfala (talk) 20:41, 13 August 2018 (UTC)
- Since dab pages are meant to guide readers to an article, a red link with no accompanying blue link is useless. (Just the same as a hatnote pointing to a red link is useless.) That’s why the MOS is so specific about that. — Gorthian (talk) 21:36, 13 August 2018 (UTC)
- What (if anything) should we do about pages full of red links which resemble a set index article as much as a dab? Maoping is a typical case but there are far worse examples. Certes (talk) 21:53, 13 August 2018 (UTC)
- If the topics are notable (as they are in the case of Maoping), my preferred course of action is to take no action: the read sea will eventually turn blue. – Uanfala (talk) 21:58, 13 August 2018 (UTC)
- MOS:DABRL says:
A link to a non-existent article (a "red link") should only be included on a disambiguation page when a linked article (not just other disambiguation pages) also includes that red link.
So the main criterion is the existence of inward links to the redlinked page. Each redlinked entry on Maoping should be considered individually. Looking at this version of the page and ignoring the bluelinks:- Maoping, Meitan County has no inward links from mainspace
- Maoping, Qiandongnan has none
- Maoping, Chengbu has three, other than the one from Maoping
- Maoping, Longshan has six
- Maoping, Baihe County has one
- Maoping, Yang County has one
- Maoping Township, Hunan has none
- So the three that have been removed from the page were valid removals, the other four were correctly retained. --Redrose64 🌹 (talk) 12:11, 14 August 2018 (UTC)
- The incoming links criterion is useful in weeding out potentially non-notable topics, but I don't think it's relevant here, as the notability of these populated places is inherently given. – Uanfala (talk) —Preceding undated comment added 12:32, 14 August 2018 (UTC)
- Wait, what? No, incoming links to red links is relevant there. If they don't have them, we don't use the red link on disambiguation pages. The populated place might still be included on the disambiguation page if there is a blue link to an article that MOS:DABMENTIONs it. -- JHunterJ (talk) 13:49, 14 August 2018 (UTC)
- Redrose64, that’s one way to find articles for blue links to match the red ones. Luckily, the editors working on Chinese places tend to be consistent with the title disambiguations, and it’s easy to check. But in other situations, one editor has chosen one disambiguated title and another editor has chosen a different one, and sometimes neither one is the one that ends up on the dab page. There was one I ran across yesterday: in one article, it was Armando González (singer), but in another it was DJ Armando González. I couldn’t even be sure it was the same person, so I finally decided against including a link on Armando González (disambiguation). In other cases, I’ll try to make all the red links consistent in all the articles. — Gorthian (talk) 20:23, 14 August 2018 (UTC)
- Thank you ( one and all ) for commenting. I can see the value of having all the redlinks in on the Bukovets, Maoping and even the Urney pages: OTOH I can also see that it goes against our guidelines on the subject. But as those pages would work well (better?) as set indexes (indices?) why not simply change them (and those like them) from dab pages to SI's? Moonraker12 (talk) 22:44, 22 August 2018 (UTC)
- @Moonraker12: This didn't work to notify me, so it won't have notified Uanfala, Gorthian, Certes or JHunterJ either. You need to add the user links and signature in the same post; moreover, it must be a new line, not a modification to an existing line. --Redrose64 🌹 (talk) 19:13, 23 August 2018 (UTC)
- Putting a set index at the base name instead of a disambiguation page is fine, as long as all the items are elements of the same set being indexed. Alternatively, creating a "side" SIA like List of places named Bukovets would also work; the Bukovets disambiguation page would simply link to it in the See also section, and we'd keep all of the appropriate entries in both places. Either way is fine. -- JHunterJ (talk) 19:50, 23 August 2018 (UTC)
- Having both a dab page and a set index covering the same topic is confusing for readers and adds unnecessarily to the maintenance burden. On the other hand, reformatting the dab page itself into a set index article is probably an option, although I don't see the point (should we turn the SIA back into a dab when all the redlinked articles get created?). I guess I would disagree with the whole premise here – that the presence of entries without links goes against the guidelines. Does it really? Not necessarily: of all guidelines, MOS:DAB is the only one that has a section dedicated to reiterating the principle "ignore all rules": Wikipedia:Manual_of_Style/Disambiguation_pages#When_to_break_Wikipedia_rules – this section is there for a reason. – Uanfala (talk) 23:07, 23 August 2018 (UTC)
- Putting a set index at the base name instead of a disambiguation page is fine, as long as all the items are elements of the same set being indexed. Alternatively, creating a "side" SIA like List of places named Bukovets would also work; the Bukovets disambiguation page would simply link to it in the See also section, and we'd keep all of the appropriate entries in both places. Either way is fine. -- JHunterJ (talk) 19:50, 23 August 2018 (UTC)
- @Moonraker12: This didn't work to notify me, so it won't have notified Uanfala, Gorthian, Certes or JHunterJ either. You need to add the user links and signature in the same post; moreover, it must be a new line, not a modification to an existing line. --Redrose64 🌹 (talk) 19:13, 23 August 2018 (UTC)
- Thank you ( one and all ) for commenting. I can see the value of having all the redlinks in on the Bukovets, Maoping and even the Urney pages: OTOH I can also see that it goes against our guidelines on the subject. But as those pages would work well (better?) as set indexes (indices?) why not simply change them (and those like them) from dab pages to SI's? Moonraker12 (talk) 22:44, 22 August 2018 (UTC)
- The incoming links criterion is useful in weeding out potentially non-notable topics, but I don't think it's relevant here, as the notability of these populated places is inherently given. – Uanfala (talk) —Preceding undated comment added 12:32, 14 August 2018 (UTC)
- MOS:DABRL says:
- If the topics are notable (as they are in the case of Maoping), my preferred course of action is to take no action: the read sea will eventually turn blue. – Uanfala (talk) 21:58, 13 August 2018 (UTC)
- I think redlinks are fine, as long as they turn blue with a stub or an article section link within 24 hours. I think an article section is the minimum, DAB pages shouldn’t send readers to mere mentions. —SmokeyJoe (talk) 23:17, 22 August 2018 (UTC)
- "mere mention" is just "mention". A mention, not an article section, is the minimum. Disambiguation pages should provide navigational aid to readers seeking things that are mentioned on Wikipedia. The existence of a page (article or disambiguation page) in the place of search results shouldn't hinder the reader's navigation. If the mention truly is too "mere" to need navigation, then it is too mere to be encyclopedic and should be removed from the mentioning article. Red links are fine as long as they also exist in articles; they don't have to turn blue within any timeframe. MOS:DABRL. -- JHunterJ (talk) 19:50, 23 August 2018 (UTC)
- JHunterJ, I disagree that "mere mention" is just "mention", though I am happy for you to wordsmith a better explanation, as I think we basically agree. I think the defining line is whether the mention provides some level of definition. If the mention of "John Doe" is "On the third expedition the party was accompanied by Jane and John Doe", it is unworthy of a link, as it says nothing of any substance about John Doe, and such links would often be erroneous anyway. Any reader looking for these low level mentions should use the search function. If the mention includes an introduction, or definition, then it is more than a mere mention. I agree that an article section is above the minimum required for a DAB page link. A paragraph is sufficient, and here on discussing "mere" mentions, we are discussion the depth of mention within possibly just a single sentence. --SmokeyJoe (talk) 01:12, 8 October 2018 (UTC)
- If the mention is too mere (whatever that means), then it should be removed entirely from the article. If it's present in the article with any utility at all (and so not delete-able from the article), then it is include-able on the disambiguation page. I do not think we basically agree, so I can't wordsmith it any better. A song listed on an article with no paragraph is a mention enough for inclusion on a disambiguation page. Any mention (whether within a sentence, the subject of a whole paragraph, the heading of a section, or any points in between), along with an editor who then links to the mentioning article from the disambiguation page, is good enough. -- JHunterJ (talk) 11:44, 8 October 2018 (UTC)
- A song listed in the article on its album, that would be an appropriate DAB page link. A song that featured in the description of a timeline during an event, in an article on the event, that would not be an appropriate DAB page link. In the second case, this song, non-notable and not mentioned in any other article, I would not agree to removing its mention if sources mention it. SmokeyJoe (talk) 12:01, 8 October 2018 (UTC)
- If the mention is too mere (whatever that means), then it should be removed entirely from the article. If it's present in the article with any utility at all (and so not delete-able from the article), then it is include-able on the disambiguation page. I do not think we basically agree, so I can't wordsmith it any better. A song listed on an article with no paragraph is a mention enough for inclusion on a disambiguation page. Any mention (whether within a sentence, the subject of a whole paragraph, the heading of a section, or any points in between), along with an editor who then links to the mentioning article from the disambiguation page, is good enough. -- JHunterJ (talk) 11:44, 8 October 2018 (UTC)
- JHunterJ, I disagree that "mere mention" is just "mention", though I am happy for you to wordsmith a better explanation, as I think we basically agree. I think the defining line is whether the mention provides some level of definition. If the mention of "John Doe" is "On the third expedition the party was accompanied by Jane and John Doe", it is unworthy of a link, as it says nothing of any substance about John Doe, and such links would often be erroneous anyway. Any reader looking for these low level mentions should use the search function. If the mention includes an introduction, or definition, then it is more than a mere mention. I agree that an article section is above the minimum required for a DAB page link. A paragraph is sufficient, and here on discussing "mere" mentions, we are discussion the depth of mention within possibly just a single sentence. --SmokeyJoe (talk) 01:12, 8 October 2018 (UTC)
- "mere mention" is just "mention". A mention, not an article section, is the minimum. Disambiguation pages should provide navigational aid to readers seeking things that are mentioned on Wikipedia. The existence of a page (article or disambiguation page) in the place of search results shouldn't hinder the reader's navigation. If the mention truly is too "mere" to need navigation, then it is too mere to be encyclopedic and should be removed from the mentioning article. Red links are fine as long as they also exist in articles; they don't have to turn blue within any timeframe. MOS:DABRL. -- JHunterJ (talk) 19:50, 23 August 2018 (UTC)
What if we determined primary topic totally differently... solely from Google?
This is not a proposal, just throwing this idea out there for thought and discussion. What if we determined primary topic solely from Google?
Specifically, for a given search term, you would go to google.com and enter:
site:en.wikipedia.org search term
and click on "I Feel Lucky". Whatever page it puts you on, that's our primary topic for that term. Sound crazy? Try it out! Here are some examples:
- site:en.wikipedia.org mercury
- site:en.wikipedia.org barack obama presidential campaign
- site:en.wikipedia.org paris
- site:en.wikipedia.org portland
Why do this?
- Leverage Google's billion dollar AI technology to determine what users are mostly likely looking for on WP for a given term.
- Increase how often users are taken to the page they are seeking.
- No more debates about what the primary topic is.
Thoughts? --В²C ☎ 00:26, 15 August 2018 (UTC)
- Horrible. Stupid. Wikipedia is not Google. Article titles should not be sought to match google behaviours. --SmokeyJoe (talk) 10:52, 15 August 2018 (UTC)
- If taking users to where they are most likely trying to go when searching with a given term is a priority, then that's exactly what this approach facilitates. If your priority is to take users to the article or dab page you think they should be taken for a given search term based on your personal sensibilities, then this approach is one you would not favor. --В²C ☎ 16:50, 15 August 2018 (UTC)
- You confuse typing in a url with using a search engine. A good title is not necessarily a common search term. Common search terms derived from google are not reliable sources, no matter how often coincidental. —SmokeyJoe (talk) 08:12, 16 August 2018 (UTC)
- If taking users to where they are most likely trying to go when searching with a given term is a priority, then that's exactly what this approach facilitates. If your priority is to take users to the article or dab page you think they should be taken for a given search term based on your personal sensibilities, then this approach is one you would not favor. --В²C ☎ 16:50, 15 August 2018 (UTC)
- Bad idea. We already know that Wikipedia is not a reliable source. Why would you want to restrict the search domain to Wikipedia pages? Mathglot (talk) 11:07, 15 August 2018 (UTC)
- This approaches leverages Google's AI technology to identify what topic users are most likely to be seeking with a given search term as it applies to WP articles. It has nothing to do with our reliability. --В²C ☎ 16:50, 15 August 2018 (UTC)
- Do you realize that Google's algorithms are heavily weighted towards following current and local trends, so what they show varies greatly between places, persons and moments? "Google's AI technology" does not provide one topic which is the most sought for a given term. Diego (talk) 14:32, 16 August 2018 (UTC)
- This approaches leverages Google's AI technology to identify what topic users are most likely to be seeking with a given search term as it applies to WP articles. It has nothing to do with our reliability. --В²C ☎ 16:50, 15 August 2018 (UTC)
- Yeah, let's see how well that works.
- site:en.wikipedia.org Thor's "I'm feeling lucky" leads to Thor: The Dark World
- site:en.wikipedia.org Britannia is Britannia (TV series)
- site:en.wikipedia.org Mosaic is Mosaic (murder mystery)
- site:en.wikipedia.org Rampage is Rampage (2018 film)
- site:en.wikipedia.org Nostalgia is Nostalgia (2018 film)
I don't know you, but I'm getting a lot of recentism right there. I'd dare to say that Google doesn't have the same interests than us in deciding which topic to show first. Not to mention that, in a few months, those searches will likely lead to different articles, so there's no stability at all. Diego (talk) 11:34, 15 August 2018 (UTC)
- Hmm, let's try a couple more examples. I'm Feeling Lucky.
- site:en.wikipedia.org Apple -> Apple Inc.. In flat contradiction of the "long-term significance" guideline at WP:PTOPIC
- site:en.wikipedia.org Luther -> Luther (TV series). So much for the Reformation, a triviality compared to a 2010 UK TV series
- site:en.wikipedia.org Mercury -> Mercury (planet). So much for chemistry and mythology
- site:en.wikipedia.org Phoenix -> Phoenix (mythology). Tough luck if you live in Arizona, your state capital is insignificant
- site:en.wikipedia.org Washington -> Washington, the DAB page. I finally got lucky
- I have other worries also:
- (1) I dislike PTOPICs unless the selection is clear beyond reasonable doubt. Even then, the PTOPIC will slowly but surely accumulate bad links and no bot report will ever find them. Unhelpful to readers, and bad for the project. (I check the incoming links to Tetrahedron every few months; I usually find a few which were intended for Tetrahedron.)
- (2) WP:RECENTISM.
- (3) Wikipedia is not WP:RS; Google search results are even less so. There are armies of SEO spammers out there trying to drive the sites which they are being paid to promote up the Google rankings. That's a cottage industry in India and Vietnam. As a former spamhunter/moderator on another site, I saw cases where spammers had got their sites up to the top Google hit. (It was standard practice to open such sites only in a sandboxed browser. I remember one which went all Tasmanian devil on me – it was trying to escape, but my sandbox and AV wouldn't let it.)
- (4) Anyone who wants to find out what Google calculates as the primary meaning of a word or phrase has a very simple way to do so. It doesn't involve Wikipedia at all.
- In conclusion: No. Just: No. Narky Blert (talk) 19:17, 17 August 2018 (UTC)
- Narky Blert, advertisers are excluded and WP is involved because the Google search is limited to the English WP with site:en.wikipedia.org. That said, much of this depends on one's opinion about the relative benefits and risks of landing someone on a sometimes "wrong" article page vs on an always wrong dab page. You dislike PTOPICs; well, I dislike dab pages, and that's probably the biggest difference. In ambiguous cases I'd much rather sometimes be taken to the wrong but most-likely-to-be-sought article, with a hatnote link to my sought article, then always be taken to a wrong (dab) page. As to RECENTISM, as I noted below, I've always felt that's mainly to address temporary spikes in popularity of a few days or weeks, not sustained popularity that is at least many months long, if not years. I mean, if people searching with a given term are most likely looking for some article for a recent movie, and that's likely to be sustained for at least a few months, I think that's where we should take them. If the popularity is only going to spike for a few days, or a couple of weeks, yeah, that is RECENTISM, and we shouldn't bother. It all comes down to what you believe best serves our users. Getting them to the articles they are seeking through the fewest clicks, or delivering to them whatever we think they should be seeing for a given search. Personally, I'm repulsed by that latter goal. It reeks of elitism and snobbery, which is my objection to the whole "historical significance" concept. --В²C ☎ 00:02, 18 August 2018 (UTC)
- Landing on a DAB page is never wrong. What exactly is the problem with one more click?
- I try to look at around 200 bad links to DAB pages every day. I can usually fix about 180 of them. That is about one-third of the bad links which User:DPL bot reports every single day. The WP:DPL team has got the headline number down from 65,000 four years ago to around 5,000. If we're holding that number, it may be because I'm putting in one-third of the effort.
- No bot can spot bad links to WP:PTOPIC pages. I see and fix 5 or 10 of those every day; usually when I've seen a possible bad link near a bad link to a DAB page, and because I'm inquisitive and could be bothered to look.
- At a rough estimate, based on what I have seen and fixed, English Wikipedia contains several thousands or tens of thousands of links which are completely wrong.
- Put in a shift or two at Disambiguation pages with links. Unless you do, I will continue to dismiss your armchair opinions, Narky Blert (talk) 01:08, 18 August 2018 (UTC)
- Yes, there are far too many bad links to WP:PTOPIC pages: primates celebrating mass, fees paid in Guineas, songs by Prince and Nirvana… I fixed over 1,000 today and that's the tip of the iceberg. They would be so much easier to find if we had dabs at those base names. Certes (talk) 01:37, 18 August 2018 (UTC)
- By the same token I submit that anyone who works with cleaning up bad links is going to be biased to favor making finding "bad" links easier over getting users to their sought pages more efficiently. I'll put in some time there, and give it some more thought. Thanks. --В²C ☎ 01:52, 18 August 2018 (UTC)
- Yes, there are far too many bad links to WP:PTOPIC pages: primates celebrating mass, fees paid in Guineas, songs by Prince and Nirvana… I fixed over 1,000 today and that's the tip of the iceberg. They would be so much easier to find if we had dabs at those base names. Certes (talk) 01:37, 18 August 2018 (UTC)
- Narky Blert, advertisers are excluded and WP is involved because the Google search is limited to the English WP with site:en.wikipedia.org. That said, much of this depends on one's opinion about the relative benefits and risks of landing someone on a sometimes "wrong" article page vs on an always wrong dab page. You dislike PTOPICs; well, I dislike dab pages, and that's probably the biggest difference. In ambiguous cases I'd much rather sometimes be taken to the wrong but most-likely-to-be-sought article, with a hatnote link to my sought article, then always be taken to a wrong (dab) page. As to RECENTISM, as I noted below, I've always felt that's mainly to address temporary spikes in popularity of a few days or weeks, not sustained popularity that is at least many months long, if not years. I mean, if people searching with a given term are most likely looking for some article for a recent movie, and that's likely to be sustained for at least a few months, I think that's where we should take them. If the popularity is only going to spike for a few days, or a couple of weeks, yeah, that is RECENTISM, and we shouldn't bother. It all comes down to what you believe best serves our users. Getting them to the articles they are seeking through the fewest clicks, or delivering to them whatever we think they should be seeing for a given search. Personally, I'm repulsed by that latter goal. It reeks of elitism and snobbery, which is my objection to the whole "historical significance" concept. --В²C ☎ 00:02, 18 August 2018 (UTC)
- Hmm, let's try a couple more examples. I'm Feeling Lucky.
- That's an interesting idea but I can't support it for several reasons. It does not address the question of whether a primary topic exists at all. Google shows different results to different people: whose lucky pick do we take? Its balance of current usage and lasting importance may not be best for Wikipedia. Finally, Google is influenced by advertisers; we shouldn't move Mars bar to Mars at the whim of the chocolate firm's marketing agency. Certes (talk) 11:39, 15 August 2018 (UTC)
- All good points.
- The point here is that there is always a "primary topic" when defined in this new way. It means being okay with sending users to the wrong page that is not a dab page (more on that point coming in our discussion above).
- There's a tag you can add to the URL (&pws=0) to de-personalize results, especially if used in an incognito session. I haven't figured out how to combine it with "I Feel Lucky", but you can always go with the first result in a regular Search.
- I've long held that lasting importance, to the extent that it's significant, is already inherent in how likely a topic is to be sought. No separate consideration is needed.
- I don't think advertisers influence searches limited to a particular site. Apple, Ford, and Oracle do go to the respective pages on corporations, but Trek does not. I think this accurately reflects the likely sought pages respectively.
- --В²C ☎ 16:50, 15 August 2018 (UTC)
- All good points.
- We might usefully employ the "Google test" in a related area. Topping the search results isn't sufficient to make a topic primary, but it is an argument against any other topic being primary for that term. If Wikipedia's normal process and search engines suggest different primary topics, then there may be enough confusion in the reader's mind to justify a dab. Certes (talk) 12:48, 16 August 2018 (UTC)
- Of course what I'm suggesting is a radical redefinition of what it means to be a "primary topic". Topping the search results would not only be sufficient to make a topic primary, that's what would make it primary by (the new) definition, and every term would have a primary topic. Entering a term and selecting Go would take you to the article about the topic you're most likely seeking (the primary topic for that term). But just like with Google's "I'm Feeling Lucky" button, you would always have the Search alternative to see and pick from a list of results (including Search Term (disambiguation)). --В²C ☎ 20:13, 16 August 2018 (UTC)
- But then what would we do for aggravation?? Herostratus (talk) 23:30, 16 August 2018 (UTC)
- Move Aggravation to Aggravation (disambiguation) over redirect, and move Aggravation (law) → Aggravation? Certes (talk) 00:15, 17 August 2018 (UTC)
- Google is showing Aggravation (board game) first in the results for "site:en.wikipedia.org aggravation", so per that we would move that article to Aggravation. The board games is what most WP users are seeking when searching with "Aggravation", according to Google AI tech... --В²C ☎ 01:37, 17 August 2018 (UTC)
- That's the same as what page views show. I bet those two metrics line up way more often than not. Station1 (talk) 02:37, 17 August 2018 (UTC)
- What happens when something like Aggravation (2021 film) is released a few years down the line and gets the lion's share of searches? Do we keep looking how Google searches evolve, and shuffling all our article titles accordingly Diego (talk) 05:16, 17 August 2018 (UTC)
- Oh dear. I believe that Herostratus' comment was not entirely serious, and I was replying in the same spirit. I don't think any of those pages should be moved. Certes (talk) 10:32, 17 August 2018 (UTC)
- Serious or not, it's a good example of how this change would affect titles. We would be sending more people more often to the articles they are seeking. But yes, as popularity changes, we would realign accordingly, taking into account WP:RECENTISM, which says not to react to brief changes in the popularity winds. I've always taken that to mainly refer to news stories about single events that create sudden spikes in popularity. But yeah, if it's a new movie that shows sustained popularity in what our users are searching for, we should adjust our article titles accordingly. That's the price we pay for not having dynamic intelligent search processing algorithms like Google does. And, yes, Station1, Google results confined to English Wikipedia should align very well with our own page view counts. Good point. --В²C ☎ 16:00, 17 August 2018 (UTC)
- Google is showing Aggravation (board game) first in the results for "site:en.wikipedia.org aggravation", so per that we would move that article to Aggravation. The board games is what most WP users are seeking when searching with "Aggravation", according to Google AI tech... --В²C ☎ 01:37, 17 August 2018 (UTC)
- Move Aggravation to Aggravation (disambiguation) over redirect, and move Aggravation (law) → Aggravation? Certes (talk) 00:15, 17 August 2018 (UTC)
- But then what would we do for aggravation?? Herostratus (talk) 23:30, 16 August 2018 (UTC)
- Of course what I'm suggesting is a radical redefinition of what it means to be a "primary topic". Topping the search results would not only be sufficient to make a topic primary, that's what would make it primary by (the new) definition, and every term would have a primary topic. Entering a term and selecting Go would take you to the article about the topic you're most likely seeking (the primary topic for that term). But just like with Google's "I'm Feeling Lucky" button, you would always have the Search alternative to see and pick from a list of results (including Search Term (disambiguation)). --В²C ☎ 20:13, 16 August 2018 (UTC)
- I think it's a very interesting proposal. But have a look at User:Andrewa/Primary Topic RfC#Some of this is highly counter intuitive. If I'm right, it pulls the rug out from under much of the above discussion. Discuss at User talk:Andrewa/Primary Topic RfC if you think you can shed some light on it. Andrewa (talk) 01:31, 22 August 2018 (UTC)
PRIMARYTOPIC vs NOTADICT
From time to time, including currently at Talk:Nosedive_(disambiguation), I see the following sentiment expressed in RM discussions: a relatively obscure topic should not be at its basename if the basename is a commonly known word even if we don't have a topic about that word (WP:NOTADICT) nor any other topic with that name. What some people seem to prefer in these cases is that a dab page be at the basename, with links to the article about the obscure topic and to an article that is ostensibly somewhat about the common word for which we have no specific article. I'm baffled by the propensity for sending users to dab pages, but maybe I dislike them more than most.
I'm wondering how others here feel about this issue, and whether some clarification at PRIMARYTOPIC wouldn't help. I know we already have Wikipedia:Disambiguation#Not_"what_first_comes_to_(your)_mind", but that doesn't seem to be enough, perhaps because the focus there seems to be to avoid personal bias regarding "what first comes to mind". --В²C ☎ 17:28, 20 August 2018 (UTC)
- A primary topic needs long term significance, not just pageviews and click bait. This is because Wikipedia is an encyclopedia, which means it is an historiographical document. Here, as with the thread above more-so, you dislike the long term significance part, we know. Pageviews and click-counting statistics are interesting, illuminating, informing, but they cannot override long term significance. If there are no pageviews, log term significance works fine. If there is no long term significance, who cares?
- The long term significance of a title is unquestionably impacted by the long term existence of a word, especially where topics are derivatively named from the word. There is nothing wrong with the WP:NOTADICT policy; Wikipedia should not attempt to host dictionary entries overlapping with the purpose of Wiktionary. However, readers cannot be expected to instinctively know the WP:NOTADICT policy and that mere simple definitions cannot have Wikipedia articles.
- You may dislike dab pages more than average, but you have a point. I agree, readers should know when they are going unwantingly to a dab page. Similarly, readers wanting a dab page (it happens!) should know which is the dab page. The solution is to recognize that WP:MALPLACED was a randomly made up bad idea and should be overturned. Overturned one page at a time, not wholesale. If a page is a disambiguation page, it should *always* be suffixed with "(disambiguation)". This would make things so much easier using the "search" drop down menu that appears as you type, as well as urls that you may check before following, and mouse-over hovertext. I believe that there are number of other advantages, and no disadvantages. There is no primary topic for "nosedive", all topics derive from the word. The dab page should be at Nosedive (disambiguation), and nosedive should redirect to the disambiguation page. The disambiguation page is the natural place to find the wiktionary link, which suits perfectly the reader intent on going for the word.
- The way to avoid personal bias is to look to the best quality sources, go to those sources, and see how they introduce the topic. Attempts to use the myriads of available statistics opens more doors for unconscious bias. --SmokeyJoe (talk) 07:09, 21 August 2018 (UTC)
- Perfect analysis by SmokeyJoe. I agree that having disambiguation pages at base names is a culprit of making them unintuitive; if they were unambiguously labeled as navigation pages, they would be more homogeneous and simpler to understand - after using the search box, no one is surprised to find a navigation page with links to existing articles; but I agree that it may feel weird to follow an entry to a base name in the preview results of the drop-down search box, and not find an article there. Diego (talk) 07:44, 21 August 2018 (UTC)
- Just a note re "malplaced". Currently, every dab page at the base title ought to have (and almost every one does have) a redirect from the corresponding title with "(disambiguation)". And if all primary topic dab pages were moved, then there would still be redirects at the base titles – in either case, if you type "X" in the search box then the drop-down suggestions will include both "X" and "X (disambiguation)". As for urls and mouse-over text of links, assuming these are within wikipedia, then they are already piped via the "(disambiguation)" redirects. – Uanfala (talk) 09:46, 21 August 2018 (UTC)
- It's true that selecting the base title would still lead the user to the DAB page. But there's an important difference, which is that the target page would have a highly visible and huge header at the top saying ARTICLE TITLE (disambiguation), instead of the current DAB page title which is just ARTICLE TITLE. That alone should make it clear to the reader that they're not seeing an article about the topic. Diego (talk) 10:05, 21 August 2018 (UTC)
- Do you think this makes a difference? Given that dab pages have an immediately recognisable layout that visually differentiates them from the articles, I'm not sure readers who land on a dab page will need to see the title of the page to know that they're on a dab page. – Uanfala (talk) 10:10, 21 August 2018 (UTC)
- (edit conflict) P.S. And maybe, just maybe, it would also help if we changed MOS:DCAT to enforce placing the {{disambiguation}} template at the top, rather than the bottom of the page, so that it can actually be seen on long DAB pages. Diego (talk) 10:12, 21 August 2018 (UTC)
- I was writing the above P.S. comment and I edit-conflicted, and I think it somewhat answers your question. Of course it helps to make special pages as visually differentiated as possible. I remember this particular episode where there was a very high-profile ongoing event (I can't remember which one right now, something about American elections I think) and readers were coming in droves to the disambiguation page that came naturally from the search term. The DAB page started with a short paragraph describing the event; and the talk page was having comments like "why is there so few content about this event at Wikipedia?" (!) Many people were arriving to the page and not following through to any of the several linked articles, which had all the details about the topic in different years; they were thinking the DAB page was the actual, very short article.
- Not all readers are as used to the layout as we are. In particular, I'm not so sure that there's a clear visual difference between DAB pages and list articles, not one that an untrained eye would spot at first glance. Consider this vs this as a quick example. Diego (talk) 10:25, 21 August 2018 (UTC)
- A page that begins with a paragraph describing the event should probably be a broad concept article rather than a dab. Certes (talk) 10:48, 21 August 2018 (UTC)
- Yes, probably. It didn't help to the thousands of readers we were having at that DAB page that particular morning, though, which had to use what we had right then. The point is that they definitely weren't noticing that the page was a disambiguation page rather than an article, so the distinction by layout alone is not that clear to many readers. Diego (talk) 11:02, 21 August 2018 (UTC)
- A page that begins with a paragraph describing the event should probably be a broad concept article rather than a dab. Certes (talk) 10:48, 21 August 2018 (UTC)
- What do you think of it:Madonna or pl:Madonna? Certes (talk) 10:48, 21 August 2018 (UTC)
- Both have something that Madonna doesn't have: a message above the fold saying that it is a disambiguation page. I like that. Their visual design (and the icons they use) stand out a bit because it's different than ours, but we wouldn't have that problem (I mean, the icon at our DAB pages is a subtle grey). Diego (talk) 11:02, 21 August 2018 (UTC)
- I've made (and then reverted) this fast edit for comparison, to give us an idea how it would look with the template at the top. Or, we could have it like this, after the lead sentence with the most prominent articles. Diego (talk) 11:12, 21 August 2018 (UTC)
- It's true that selecting the base title would still lead the user to the DAB page. But there's an important difference, which is that the target page would have a highly visible and huge header at the top saying ARTICLE TITLE (disambiguation), instead of the current DAB page title which is just ARTICLE TITLE. That alone should make it clear to the reader that they're not seeing an article about the topic. Diego (talk) 10:05, 21 August 2018 (UTC)
- Just a note re "malplaced". Currently, every dab page at the base title ought to have (and almost every one does have) a redirect from the corresponding title with "(disambiguation)". And if all primary topic dab pages were moved, then there would still be redirects at the base titles – in either case, if you type "X" in the search box then the drop-down suggestions will include both "X" and "X (disambiguation)". As for urls and mouse-over text of links, assuming these are within wikipedia, then they are already piped via the "(disambiguation)" redirects. – Uanfala (talk) 09:46, 21 August 2018 (UTC)
- A primary topic needs long term significance? How about Anne Hathaway, an example from WP:D? Long-term significance is a factor to consider, and one that I believe is implicitly and more than adequately accounted for in page view counts and by the Google test, but it's not the only consideration. I don't see why topics not covered on WP, like the dictionary meaning of nosedive, should be considered when determining whether a given title is ambiguous, much less for determining what the primary topic is for a given title. --В²C ☎ 17:06, 21 August 2018 (UTC)
- A primary topic needs long term significance? - No, but it needs that there's no other long-term-significant topic competing for primacy. Anne Hathaway is a case where consensus has determined that usage (what readers were looking for) was the main factor in determining the primary topic, and the historical significance of Shakespeare's wife was not enough to override that.
- I don't see why topics not covered on WP, like the dictionary meaning of nosedive, should be considered - Because disambiguation (including primary topics) is all about navigation, i.e. leading readers to the content they're looking for, not about content structure; we have categories for that second purpose. To create effective navigation, you need to take into account what readers know, not just what content we have.
- Sometimes a hint saying sorry, we don't have anything about that, but you might try the shop next door is as important to provide adequate service as breadcrumbs to existing content, as it prevents the patron from going around, lost; a good librarian should know when to guide users out and stop looking for something that is not here. Diego (talk) 09:01, 24 August 2018 (UTC)
- That would make sense if we were talking about topics that users reasonably and regularly would seek here. But if we identify any topics like that, then we should create articles for them. This isn't the case of articles we should have but just haven't created yet, in which case your approach would make sense. We're talking about non-encyclopedic topics, like simple word definitions, that nobody would reasonably even look for here. Nobody is going to wonder what the history of a nosedive is. It doesn't belong here. It's not a topic. We can and should ignore it, as if it does not exist at all, because, for our intents and purposes, it doesn't exist. We should not clutter WP with dictionary content. That's the point of WP:NOTADICT. --В²C ☎ 07:13, 25 August 2018 (UTC)
- How do you know what topics users "reasonably and regularly" would seek? A major problem is that we don't have good tools to understand how our readers use the website. Until you try it, you don't know whether a word has an associated article at Wikipedia or not (being a non-English native, I definitely had to look here what a "nosedive" is). Also, your reasoning would explain why we don't have an article, but not why we shouldn't make it easy finding an external link to a sister site. Diego (talk) 17:35, 25 August 2018 (UTC)
- That would make sense if we were talking about topics that users reasonably and regularly would seek here. But if we identify any topics like that, then we should create articles for them. This isn't the case of articles we should have but just haven't created yet, in which case your approach would make sense. We're talking about non-encyclopedic topics, like simple word definitions, that nobody would reasonably even look for here. Nobody is going to wonder what the history of a nosedive is. It doesn't belong here. It's not a topic. We can and should ignore it, as if it does not exist at all, because, for our intents and purposes, it doesn't exist. We should not clutter WP with dictionary content. That's the point of WP:NOTADICT. --В²C ☎ 07:13, 25 August 2018 (UTC)
- Perfect analysis by SmokeyJoe. I agree that having disambiguation pages at base names is a culprit of making them unintuitive; if they were unambiguously labeled as navigation pages, they would be more homogeneous and simpler to understand - after using the search box, no one is surprised to find a navigation page with links to existing articles; but I agree that it may feel weird to follow an entry to a base name in the preview results of the drop-down search box, and not find an article there. Diego (talk) 07:44, 21 August 2018 (UTC)
BTW, we do have the Descent a article which describes the history of nosedives, and I think it would be a much better target than the current obscure episode of a recent TV series. It certainly would have served me much better than the current PRIMARYTOPIC disposition, even though I'm a regular who fully understands how to navigate our website. I dread to think how a user with lesser knowledge of English, and who hadn't heard of the series, could have found about that content and understand the concept. Diego (talk) 18:06, 25 August 2018 (UTC)
- Diego Moya, you make some good points. But there is
a hint saying “sorry, we don't have anything about that, but you might try the shop next door”
: that’s what the {{Wiktionary}} template is for. It should be on any dab page where the ambiguous term is defined in Wiktionary. Then the dab page can concentrate on guiding readers to what Wikipedia has. — Gorthian (talk) 21:10, 25 August 2018 (UTC)- Yes, that's what I was getting at. That's a good reason to place the DAB page at a prominent place more often than not, which is what Born2cycle wants to avoid. Arriving at a DAB page when looking for the bare word shows the Wiktionary link, while placing a PRIMARYTOPIC article does not, so it's much easier in the first case for the reader to find the definition of the word, or any article related to the meaning of the word but which is placed at an article with a different name (like the nosedive case). Diego (talk) 22:25, 25 August 2018 (UTC)
- @Diego Moya:, the process to determine which topics users seek "reasonably and regularly" is not perfect but nothing novel. It's about the same as determining whether we should have an article about a topic, which WPians have been doing since WP launched. It's okay to leave pointers to external sources where appropriate, but we should not rearrange our titles to make it easier to find external sources at the cost of adding more clicks to users trying to get to topics we do cover, which is exactly what considering external uses in determining primary topic does. --В²C ☎ 19:56, 27 August 2018 (UTC)
- -
- Also, the page view stats at Descent (aeronautics)#dives are dwarfed Nosedive stats. All of the evidence indicates the interest in the episode is far higher than in any other use. --В²C ☎ 21:05, 27 August 2018 (UTC)
- That's just if you consider exclusively "recent usage" as "all of the evidence". However if you consider content published in reliable sources (which is what we do to determine notability), the aviation maneuver is prevalent. The consensus is to consider all the information published over time and not just spikes of current trendiness, because mere transient popularity is not a good proxy of overall interest, and it's too prone to bias favoring a subset of readers that would leave all the rest out of the picture (in this case, spectators of the TV series, which is targeted at a very specific demografics of western technologically educated audiences). If you cater to that subset because they're currently actively looking for the chapter and temporarily appear higher in search trends, you're making a disservice to all the other classes of readers, who will be interested in the topic with the majority of historic coverage.
- We determine which topics to cover by looking at what other people has written about in a reliable way, and then excluding all the sources that we dont't think are providing an encyclopedic treatment and choosing the subset of all that information that we want to cover. But readers care about finding information, they don't care about what we want or don't want to include in the project. If we only assess the types of content that we want to include in the encyclopedia for determining primary topics, we are distorting the amount of coverage by third parties by attending to just those that write in a style we can use, and we are re-arranging our articles in a way that is detrimental to what readers want to know about, by establishing navigation according to content curated to our interests, not those of the external world.
- Therefore, of course we should make it easier to find such external content that exist but we've decided not to use, yet is probably as interesting to readers as what we happen to collect; it's only fair that, if we decide to exclude some types of information from our project, at least we let our users know about that decision and guide them to other places who do cover that information. Diego (talk) 22:45, 27 August 2018 (UTC)
- -
- That said, we can accomodate the fact that there's a significant amount of people looking for the series article. We do that by establishing that such high interest means that there's no primary topic, because two different concepts are competing for mindshare; therefore placing a useful disambiguation page that let readers find their way, instead of just having the historical concept at the base title, as would have happened without the spike of interest for the TV show. Diego (talk) 22:51, 27 August 2018 (UTC)
Arbitrary break
- Born2cycle is back on his drive to minimize title length for frequently viewed articles. It has been his sole ambition for his entire Wikipedian career. It necessarily means doing away with "long term significance" in favor of usage as measured by page-views. It is entirely a misguided ambition, as it has no basis in making a better product for the readers, and it is at odds with the function of an encyclopedia, which is to collect all information and to organise it in a logical fashion.
- With nosedive, he has cherry picked an extreme example of contrast between long term significance and page-views. One is the highly angled forward descent of an aircraft, covered both in a section of an aeronautical article and at Wiktionary, the other is a single episode of a popular TV show. Long term significance should trump because every fan of the TV episode knows what a nosedive is, the knowledge of the meaning of the word is a prerequisite to understand the them of the episode, but of the set of people interested in the aeronautical topic only a very small proportion would be fans of the TV series or otherwise interested in the single isolate episode.
- Accordingly, I think we need to define "Primary Meaning" alongside "Primary Topic", where Primary Meaning includes generic concepts and common words regardless of there being a current article on the topic. It is important to prevent astonishment to the reader when they think they are following the generic meaning of a word and are taken to an unexpected niche, albeit high page view, article. "Primary Topic" should have an exclusion where the word that would be used as the title does not match the Primary Meaning of that word.
- Further, as I have frequently noted in RM discussions, without objections except by those who argue that the rules include no mention of this, commercial topics should have to meet a higher bar to be afforded Primary Topic status. The driver for this is the importance of WP:NOTPROMOTION. Commercial products are typically named with a strategy of seeking a catchy name. Commercial topics also come along with insidious Search engine optimization tricks, increasingly if not already totally, automated tricks. Wikipedia should be resistant to this. Largely, largely undocumented, it already is. It has been explicitly consider in many cases. Windows passes, and Apple fails, for example. Nosedive should definitely fail. It is a commercial product, realized with a flurry of promotion, lead episode of a series that the Netflix needs to advertise to get subscriptions, and it surely has artificially boosted ghits, millions, which accompany a burst of low brow fan hype in popular media, nothing scholarly.
- I also think User:Born2cycle needs to be topic banned to prevent continued relentless agitation at obscure articles, seeking to advance his battle for algorithmic titling decisions. It is disruption. It distracts other editors from doing useful things. It is not in the interested of improving the encyclopedia.
- --SmokeyJoe (talk) 01:56, 28 August 2018 (UTC)
- I'm not going to engage with you, except to say you have grossly misrepresented my motivations and behavior and that commenting about such matters on a policy/guideline talk page here is a blatant violation of WP:NPA. --В²C ☎ 02:15, 28 August 2018 (UTC)
- I have studied Born2cycle's actions and articulated motivations for many years, and consider much, especially User:Born2cycle#Persistence_pays, to be unashamed disruption seeking to achieve an objective through wearing down everyone else. Perhaps there is a deeper motivation underlying the algorithmic titling and title minimisation objectives that I don't understand, but I have read his userpage and usersubpages, and don't find one. This is now quite topical, PRIMARYTOPIC vs NOTADICT is a worthy discussion point, all I posted applies. It needs a discussion here, but what it does not need is petty skirmishing in isolated RM discussions. We know he believes in "change at the article level contrary to guidelines", and I submit that such an approach, litigating titling policy opinions in isolated RMs on cherry picked articles is very disruptive to editors genuinely interested in those articles. These questions, IF they are important, should be approached through an RfC, not through RM skirmishes. --SmokeyJoe (talk) 03:46, 28 August 2018 (UTC)
- I don't see anything disruptive about B2C, they (like me) are obviously very interested in titles, consider this comment which is linked from their user page. However with NOTDICT, I'd consider if the subject is properly covered by WP under any title, not just a word than can be used to refer to that topic, for example Size doesn't really cover the concept of "Big" but Hurricane is an alternative word for Tropical cyclone. Crouch, Swale (talk) 07:20, 28 August 2018 (UTC)
- Are you familiar with "big"? Have you read Talk:Big (film) & Wikipedia:Move_review/Log/2012_June#Big_(film)? It was a lot of fuss, and precedent setting for a big film with big page views not being the Primary Meaning of "big". Born2cycle was not happy then, and in starting this thread he is still wanting to litigate exactly the same issue. After five years, an RfC is reasonable, but carrying the battles on endlessly in RM discussions at obscure articles is not, that would be disruptive to editors of those articles. --SmokeyJoe (talk) 08:30, 28 August 2018 (UTC)
- Yes I read the discussions a few years ago. The major problem was that WP doesn't really cover the concept of "Big". However since that an article on Size has been created which doesn't really cover it but is better than before. There are a number of other topic so I'd say disambiguation makes sense. As you noted Bell was opposed, though I think that is a clear PT. Crouch, Swale (talk) 08:35, 28 August 2018 (UTC)
- Are you familiar with "big"? Have you read Talk:Big (film) & Wikipedia:Move_review/Log/2012_June#Big_(film)? It was a lot of fuss, and precedent setting for a big film with big page views not being the Primary Meaning of "big". Born2cycle was not happy then, and in starting this thread he is still wanting to litigate exactly the same issue. After five years, an RfC is reasonable, but carrying the battles on endlessly in RM discussions at obscure articles is not, that would be disruptive to editors of those articles. --SmokeyJoe (talk) 08:30, 28 August 2018 (UTC)
- I did not start the Nosedive RM. I just weighed in like everyone else. Then I raised the larger issue on this much broader forum, merely using Nosedive as an example because it happens to be current and is the discussion which sparked me to raise this issue here. As my user page, FAQ and each of my comments on titles should make clear, my larger motivation is article title stability, the opposite of the disruption caused by having rules that are more ambiguous than necessary and creating pointless debate where none is warranted. A point of apparent difference I have with SmokeyJoe is that I reject the notion that more descriptive titles are helpful to users by any significant degree. The name of the topic is always enough for all intents and purposes of titles on WP, including for user needs, unless disambiguation is necessary. The moment you allow adding description beyond the name of the topic you're opening a can of worms, because there is no objective means by which to determine what is "descriptive enough". Allowing for term uses not covered on WP (or covered by an obscure section on a relatively rarely visited article and therefore a relatively rarely sought topic) in deciding primary topic to disambiguate the title of the clearly most likely sought topic is not helpful to users anyway. And the result of tolerance for unnecessary description/disambiguation in titles is the disruptive never-ending RM backlog. And, yes, changes on WP tend to be made one article at a time. For better or worse, WP works bottom up, not top down. If a consensus of participants agrees to move an article against guidelines, then it is moved, and that move itself, perhaps with a few other similar moves, becomes sufficient reason to update the guideline to better reflect actual consensus. Don't shoot the messenger. --В²C ☎ 16:09, 28 August 2018 (UTC)
- B2C, if you read about Information scent, Tree testing and other topics in Information architecture, you'll learn that there are objective methods to assess whether titles are descriptive enough; namely by observing people read them and counting the number and types of errors they make when trying to navigate to their targets. It's a different thing that we don't have the tools to put those methods in practice... but I don't want to start a digression; just to point out that, according to the professional experts in how to properly lay out a website, they think that merely looking at the content you have is an awful practice, and you instead must look at what people intend to do with it (which quite often might involve looking for information that is not here at all). Diego (talk) 17:18, 28 August 2018 (UTC)
- I get all that. WP is unique for a number of reasons. First of all, our title decisions really amount to choosing between Just The Name and The Name (plus some description). Whichever one we choose for the title of the article, the other typically redirects to it. Another decision we make is whether to put one of several ambiguous topics at a given title, or put a dab page listing all of the ambiguous topics at that title. These are very specific contexts in which to consider how people react and respond. And we do have tools that provide considerable information about people's behavior. Page view counts tell us very accurately how often people are looking at one topic relative to others. The Google test, especially when the search is narrowed to en.wp, leverages the observational tools Google has to tell us which of our pages is mostly likely being sought for a given search term. The bottom line is that all of the tools we do have at our disposal tell us quite clearly that people searching for "nosedive" are overwhelmingly looking for the TV show episode. I've never even seen the show, much less that episode. I try to be as unbiased as I can in making title decisions, going as much as possible by what these tools tell us about actual user behavior. And there's no evidence, anywhere, that users are coming to WP to search for dictionary definitions. --В²C ☎ 18:31, 28 August 2018 (UTC)
- The thing is, we don't have no evidence that users want definitions because the tools we have are inadequate to provide such evidence. Neither Google nor page views can tell us anything about how many users search for a word looking for its meaning, arrive to a disambiguation page and then go to to Wiktionary to learn the word's meaning. Page views provide a measure of popularity exclusively, but as I've said, exclusively catering to what's trendy is a terrible way to design the navigation structure without also taking into consideration other elements defining the meaning and cultural context of the term. We're not a media company, we try to be an encyclopedia, and the treatment we give to people outside of the mainstream latest craze matters; it's not OK to offer a dreadful experience to those looking for the educational yet less popular topic.
- I'll confess, I know first hand that people look for definitions at the encyclopedia, because that's something I often do myself, having found that Wikipedia is a very good tool for that purpose. I use the search bar for learning the meaning of English words that I've never encountered before; and I believe it's likely that other non-native English speakers will be interested in the same usage, as well as English elementary and secondary students. Wikipedia usually provides a very good first approach to concepts that are completely unknown, by placing them in context with categories or the introductory catchphrase "In FIELD X, a NAME is...", and then the lede proceeds to illustrate the concept with examples and usage notes (which you don't get in a dictionary).
- When using the encyclopedia in this way, it's a really awful experience to find the target page hijacked by some obscure pop-culture item that I've never heard before, but which happens to be slightly more popular with people across the pond than all the other minutiae sharing the same name. It's understandable and even beneficial when the pop item is really a well known landmark of the English culture, i.e. when it's really the primary topic that people undoubtedly identifies with the term even if it's not its original meaning; but not for every random music album, TV episode, low-budget movie or cleaning product brand, which is getting a slightly higher attention than the rest because of its recentism. We already have a reputation of being disseminators of products for the contemporary media and technological conglomerates; let's make sure that the only popular items occupying the base names and displacing the meanings of English words are those that deserve it. Diego (talk) 20:42, 28 August 2018 (UTC)
- "Deserve" is another idea that floats around primary topic discussions. Primary-topic-ness isn't an award to deserving topics; it's a navigation service to the readers. -- JHunterJ (talk) 11:53, 29 August 2018 (UTC)
- DAB pages are navigation services. PrimaryTopic status, where overzealously applied, is the opposite. Even when justified, it doesn’t help with navigation, google doesn’t rely on titles, people don’t type urls, wikilinks don’t care. PrimaryTopics, by removing clarity, precision, make the java search box autocomplete options more difficult. —SmokeyJoe (talk) 12:27, 29 August 2018 (UTC)
- And of course no one is recommending overzealous application. Primary topic status is a navigation service too; when justified, it helps with navigation for the readership, which is why we do it. Putting readers at Earth (disambiguation) or Shakespeare (disambiguation) or Banana (disambiguation) instead of Earth, Shakespeare, or Banana would be the navigation disservice. -- JHunterJ (talk) 12:43, 29 August 2018 (UTC)
- I think everyone opposing at Talk:Nosedive (disambiguation) is overzealously pushing for a PrimaryTopic status of a narrowly popular topic. —SmokeyJoe (talk) 13:11, 29 August 2018 (UTC)
- Right. And I thing everyone supporting there is overzealously pushing for removing the primary status from the topic that would best serve the encyclopedia readership in favor of treating Wikipedia like a dictionary. -- JHunterJ (talk) 13:14, 29 August 2018 (UTC)
- So which is it that makes you think Nosedive (Black Mirror) has primary status: you don't think that Dive (aviation) is a topic identified by the ambiguous term "nosedive", or you don't think it has historical weight that makes it compete with the popularity of the TV episode? I'm not as interested in your opinion as I am in the policy ramifications of your position; what I get from your position is that apparently, the topic which correspond to the common meaning of a word has its significance somehow devalued because WP:NOTDICT would apply in some way. Diego (talk) 13:53, 29 August 2018 (UTC)
- You've got it backwards. Its significance in primary topic determination is not inflated because of its common meaning as a dictionary word. That is irrelevant in primary topic determination. What is relevant, especially in the context of this discussion about navigation aids for our users, is likelihood of being sought by users searching with "nosedive". --В²C ☎ 16:45, 29 August 2018 (UTC)
- So which is it that makes you think Nosedive (Black Mirror) has primary status: you don't think that Dive (aviation) is a topic identified by the ambiguous term "nosedive", or you don't think it has historical weight that makes it compete with the popularity of the TV episode? I'm not as interested in your opinion as I am in the policy ramifications of your position; what I get from your position is that apparently, the topic which correspond to the common meaning of a word has its significance somehow devalued because WP:NOTDICT would apply in some way. Diego (talk) 13:53, 29 August 2018 (UTC)
- Right. And I thing everyone supporting there is overzealously pushing for removing the primary status from the topic that would best serve the encyclopedia readership in favor of treating Wikipedia like a dictionary. -- JHunterJ (talk) 13:14, 29 August 2018 (UTC)
- I think everyone opposing at Talk:Nosedive (disambiguation) is overzealously pushing for a PrimaryTopic status of a narrowly popular topic. —SmokeyJoe (talk) 13:11, 29 August 2018 (UTC)
- And of course no one is recommending overzealous application. Primary topic status is a navigation service too; when justified, it helps with navigation for the readership, which is why we do it. Putting readers at Earth (disambiguation) or Shakespeare (disambiguation) or Banana (disambiguation) instead of Earth, Shakespeare, or Banana would be the navigation disservice. -- JHunterJ (talk) 12:43, 29 August 2018 (UTC)
- DAB pages are navigation services. PrimaryTopic status, where overzealously applied, is the opposite. Even when justified, it doesn’t help with navigation, google doesn’t rely on titles, people don’t type urls, wikilinks don’t care. PrimaryTopics, by removing clarity, precision, make the java search box autocomplete options more difficult. —SmokeyJoe (talk) 12:27, 29 August 2018 (UTC)
- "Deserve" is another idea that floats around primary topic discussions. Primary-topic-ness isn't an award to deserving topics; it's a navigation service to the readers. -- JHunterJ (talk) 11:53, 29 August 2018 (UTC)
- I get all that. WP is unique for a number of reasons. First of all, our title decisions really amount to choosing between Just The Name and The Name (plus some description). Whichever one we choose for the title of the article, the other typically redirects to it. Another decision we make is whether to put one of several ambiguous topics at a given title, or put a dab page listing all of the ambiguous topics at that title. These are very specific contexts in which to consider how people react and respond. And we do have tools that provide considerable information about people's behavior. Page view counts tell us very accurately how often people are looking at one topic relative to others. The Google test, especially when the search is narrowed to en.wp, leverages the observational tools Google has to tell us which of our pages is mostly likely being sought for a given search term. The bottom line is that all of the tools we do have at our disposal tell us quite clearly that people searching for "nosedive" are overwhelmingly looking for the TV show episode. I've never even seen the show, much less that episode. I try to be as unbiased as I can in making title decisions, going as much as possible by what these tools tell us about actual user behavior. And there's no evidence, anywhere, that users are coming to WP to search for dictionary definitions. --В²C ☎ 18:31, 28 August 2018 (UTC)
- B2C, if you read about Information scent, Tree testing and other topics in Information architecture, you'll learn that there are objective methods to assess whether titles are descriptive enough; namely by observing people read them and counting the number and types of errors they make when trying to navigate to their targets. It's a different thing that we don't have the tools to put those methods in practice... but I don't want to start a digression; just to point out that, according to the professional experts in how to properly lay out a website, they think that merely looking at the content you have is an awful practice, and you instead must look at what people intend to do with it (which quite often might involve looking for information that is not here at all). Diego (talk) 17:18, 28 August 2018 (UTC)
- I don't see anything disruptive about B2C, they (like me) are obviously very interested in titles, consider this comment which is linked from their user page. However with NOTDICT, I'd consider if the subject is properly covered by WP under any title, not just a word than can be used to refer to that topic, for example Size doesn't really cover the concept of "Big" but Hurricane is an alternative word for Tropical cyclone. Crouch, Swale (talk) 07:20, 28 August 2018 (UTC)
- I have studied Born2cycle's actions and articulated motivations for many years, and consider much, especially User:Born2cycle#Persistence_pays, to be unashamed disruption seeking to achieve an objective through wearing down everyone else. Perhaps there is a deeper motivation underlying the algorithmic titling and title minimisation objectives that I don't understand, but I have read his userpage and usersubpages, and don't find one. This is now quite topical, PRIMARYTOPIC vs NOTADICT is a worthy discussion point, all I posted applies. It needs a discussion here, but what it does not need is petty skirmishing in isolated RM discussions. We know he believes in "change at the article level contrary to guidelines", and I submit that such an approach, litigating titling policy opinions in isolated RMs on cherry picked articles is very disruptive to editors genuinely interested in those articles. These questions, IF they are important, should be approached through an RfC, not through RM skirmishes. --SmokeyJoe (talk) 03:46, 28 August 2018 (UTC)
- I'm not going to engage with you, except to say you have grossly misrepresented my motivations and behavior and that commenting about such matters on a policy/guideline talk page here is a blatant violation of WP:NPA. --В²C ☎ 02:15, 28 August 2018 (UTC)
Meaning of words in disambiguation
SmokeyJoe above makes an interesting point that I think merits being explored. When determining primary topic, often there are cases where either:
- the base name is a common word with some well-known meaning (what SmokeyJoe calls "primary meaning"),
- or the name of the topic with the most usage is a derivative from the topic with the most long-term significance.
As we often analyze these aspects in move discussions, I think we should acknowledge this fact and mention them in the guideline, adding them to the "aspects that editors commonly consider" (or at least the "Tools that may help to support the determination of a primary topic but are not considered absolute determining factors"); we already do this in a limited way for the second criterion. We could add some guidance on how they are used in discussions, and maybe explain the outcomes of when they've been found relevant, and how they affect decisions over what is the primary topic. Diego (talk) 09:19, 28 August 2018 (UTC)
- What would people say the primary meanings of Bell, Settle, Steep, Wells, Bury, Unlikely and Bray are? Crouch, Swale (talk) 11:11, 28 August 2018 (UTC)
- I don't know, but as a non-native English speaker, they are immediately useful to me, and at first sight they seem well placed. Bell is an obvious primary topic, being the direct meaning of one of the 2000 most common English words. Those which are not nouns are difficult to make primary; so if they don't have an obvious primary topic, a disambiguation page allows me to learn their meaning and overlook the list of articles we have about the term. I particularly like how Unlikely, which doesn't have a disambiguation page, gets to have the Wiktionary link right there at the top, even if we are at article space (although it makes me wonder if the album is primary over the comic, since Jeffrey Brown (cartoonist) gets more daily visits; isn't that a WP:TWODAB situation?)
- And contrast Bray (a disambiguation page) with Hee Haw. In the second case we have an article with many more visits than the other articles sharing the title, and for which we don't have an article covering the common meaning; that looks like a good candidate for primary topic, unless we find a section that covers that particular onomatopoeia elsewhere. Diego (talk) 11:45, 28 August 2018 (UTC)
- "Primary meaning" is a neologism for a concept for which there is no direct bearing on primary topic determination. The primary topic for a term may or may not reflect the "primary meaning" of the term. To the extent that primary meaning influences how likely users are to be looking for that meaning when searching with a given term is already accounted for in normal primary topic determination. No special consideration for "primary meaning" needs to be given. I see no reason to clutter the section with words about this. --В²C ☎ 18:18, 29 August 2018 (UTC)
- But the common meaning of a word surely has historical significance, which is one of the criteria for primary topic, right? The meaning of a word as it's understood in English has historical significance unless it's a neologism. Therefore, topics covered by Wikipedia that match the meaning of dictionary words are notable topics with long-term significance, and they shouldn't be dismissed from PT assessment merely because their meaning is also found in a dictionary. Diego (talk) 18:44, 29 August 2018 (UTC)
- The meaning has history, not historical significance (unless there's some encyclopedic coverage of the historical significance of etymology, philology, epistemology of the word. The topic of that meaning may or may not have historical significance, which is the criterion. Being a dictionary word is great for dictionaries, but Wikipedia is still not a dictionary. -- JHunterJ (talk) 19:25, 29 August 2018 (UTC)
- And no one is suggesting that Wikipedia is a dictionary, so I still don't see how you want to make the WP:NOTDICT policy relevant to primary topic discussions. But as you've just agreed, if a meaning has a long-term history, the topic covered in Wikipedia which corresponds to that meaning may have historical significance, in special when it's a notable topic on its own; and we should assess how that history is relevant - because it makes long-term significance more likely than for topics which don't have such historicity. Diego (talk) 21:58, 29 August 2018 (UTC)
- This is why I don't understand the point of long-term significance. The whole point of identifying primary topics is to arrange our titles in a manner that aids user navigation. If long-term significance causes us to send users to a page they're not seeking, we're hindering user navigation, aren't we? --В²C ☎ 01:21, 30 August 2018 (UTC)
- Is it? The point of long term significance is the logic of titles. A short title implies generic coverage. For example, Science should cover the general concept, not something very specific called "science", like Science (journal). The point of titles is not to aid navigation in the ways of search engines, simulating primitive search engine functionality will work to a point but will necessarily be limited like a primitive search engine, as well as producing bad titles. Fundamentally, a title is the biggest text at the top of the page that describes the content below. If users are being sent (by what) to a page they are not seeking, then most likely they are using a poor search engine, or title are misleading, or the disambiguation page should be more prominent. Disambiguation pages would be more prominent in search engines and other lists if all disambiguation pages were suffixed with "(disambiguation)" and ambiguous basenames redirected to disambiguation pages. WP:MALPLACED was and remains a mistake, as discussed previously. --SmokeyJoe (talk) 01:41, 30 August 2018 (UTC)
- WP Search works like this: You enter a term, click on Go, and, if there is an article (or dab page) with that title, it takes you there. The point of primary topics is to ensure that when users enter a term in that WP Search box and hit Go, that they're taken to the article about the topic they are most likely seeking, if the search term has one. If the search term has such a "most likely to be sought" topic, and we take them to an article about some other "historically significant" topic, or to a dab page, aren't we hindering the very user navigation we're trying to improve? If the topic they're most likely seeking is a TV episode, then it makes sense to have the name of that episode at the top of that article when we take them there, even if that title also corresponds to a common word or a topic with greater historical significance. The first few words of the intro handle that. --В²C ☎ 01:51, 30 August 2018 (UTC)
- No, that is not how search works. Documentation is here: Help:Searching. I'm seeing the problem. In the default skin, there is a box containing the grey text "search Wikipedia", but it primaryily serves to autocomplete, not search. The autocomplete behaviors can be found in your account preferences under the (mistitled!) "search" tab. The Default (recommended) is: "Corrects up to two typos. Resolves close redirects"
- This autocomplete is I understand a Java thing, which depending on device and connection, can be very slow and clunky. Often it is for me, so I rarely use it.
- As you type, a list of possible matches drops down, initially quite a long list. Only at the very bottom, below a horizontal line, below the grey text "containing ..." can you, without explanation, invoke the real search.
- I think this hiding / obscuring of the real search is really a bad idea.
- WP Search works like this: You enter a term, click on Go, and, if there is an article (or dab page) with that title, it takes you there. The point of primary topics is to ensure that when users enter a term in that WP Search box and hit Go, that they're taken to the article about the topic they are most likely seeking, if the search term has one. If the search term has such a "most likely to be sought" topic, and we take them to an article about some other "historically significant" topic, or to a dab page, aren't we hindering the very user navigation we're trying to improve? If the topic they're most likely seeking is a TV episode, then it makes sense to have the name of that episode at the top of that article when we take them there, even if that title also corresponds to a common word or a topic with greater historical significance. The first few words of the intro handle that. --В²C ☎ 01:51, 30 August 2018 (UTC)
- Is it? The point of long term significance is the logic of titles. A short title implies generic coverage. For example, Science should cover the general concept, not something very specific called "science", like Science (journal). The point of titles is not to aid navigation in the ways of search engines, simulating primitive search engine functionality will work to a point but will necessarily be limited like a primitive search engine, as well as producing bad titles. Fundamentally, a title is the biggest text at the top of the page that describes the content below. If users are being sent (by what) to a page they are not seeking, then most likely they are using a poor search engine, or title are misleading, or the disambiguation page should be more prominent. Disambiguation pages would be more prominent in search engines and other lists if all disambiguation pages were suffixed with "(disambiguation)" and ambiguous basenames redirected to disambiguation pages. WP:MALPLACED was and remains a mistake, as discussed previously. --SmokeyJoe (talk) 01:41, 30 August 2018 (UTC)
- This is why I don't understand the point of long-term significance. The whole point of identifying primary topics is to arrange our titles in a manner that aids user navigation. If long-term significance causes us to send users to a page they're not seeking, we're hindering user navigation, aren't we? --В²C ☎ 01:21, 30 August 2018 (UTC)
- And no one is suggesting that Wikipedia is a dictionary, so I still don't see how you want to make the WP:NOTDICT policy relevant to primary topic discussions. But as you've just agreed, if a meaning has a long-term history, the topic covered in Wikipedia which corresponds to that meaning may have historical significance, in special when it's a notable topic on its own; and we should assess how that history is relevant - because it makes long-term significance more likely than for topics which don't have such historicity. Diego (talk) 21:58, 29 August 2018 (UTC)
- The meaning has history, not historical significance (unless there's some encyclopedic coverage of the historical significance of etymology, philology, epistemology of the word. The topic of that meaning may or may not have historical significance, which is the criterion. Being a dictionary word is great for dictionaries, but Wikipedia is still not a dictionary. -- JHunterJ (talk) 19:25, 29 August 2018 (UTC)
- But the common meaning of a word surely has historical significance, which is one of the criteria for primary topic, right? The meaning of a word as it's understood in English has historical significance unless it's a neologism. Therefore, topics covered by Wikipedia that match the meaning of dictionary words are notable topics with long-term significance, and they shouldn't be dismissed from PT assessment merely because their meaning is also found in a dictionary. Diego (talk) 18:44, 29 August 2018 (UTC)
- But anyway, if you use the default autocomplete, it is better to have precise titles over most likely topics at basenames, because then the precise title will appear.
- Nosedive, for example. If the episode were at "Nosedive (Black Mirror)", then that entry would appear in the list, even at the top or second I think. This would best serve all users. --SmokeyJoe (talk) 02:34, 30 August 2018 (UTC)
- This is why I don't understand the point of long-term significance. The point is that we don't put an article at the base name merely because it gets more searches/it's more popular. It also needs to be the topic that is commonly associated with the term, with no other topic competing for the meaning, even if people are not actively seeking the other meanings at this time. That's why Madonna is a DAB, and consensus about this view is not going to change soon.
- As for how this improves navigation: it's best for those seeking the historic meaning, it makes incoming wikilinks harder to miss the mark and easier to fix, and it doesn't hurt those looking for the popular topic as much as you say it does - these can nagivate it directly using Google or selecting the disambiguated title in the search box dropdown, which gives more hints as to which on of the several significant meanings is the one covered by the article. Diego (talk) 05:57, 30 August 2018 (UTC)
- I think there are hints of acceptance of the notion that Madonna should be moved to Madonna (disambiguation), and Madonna redirecting to it. This would mean the the unqualified "Madonna" no longer appears in the drop down list, "(disambiguation)" would be displayed in the title line (currently whitespace), and readers won't have to guess between "Madonna" and "Madonna (entertainer)" when they are looking for the singer. --SmokeyJoe (talk) 06:19, 30 August 2018 (UTC)
Dictionary words can impact PRIMARYTOPIC decisions
I think there is a a consensus that dictionary words *can* impact PRIMARYTOPIC decisions. NOTADICT is not a blanket rule for excluding dictionary meanings from consideration.
I think this soft wording is agreeable. Just because there is a dictionary word, it doesn't mean that there can't be a PrimaryTopic for a topic which is not the meaning of that word.
Examples from https://randomwordgenerator.com/
- Dictionary words that have a primary topic (is the title of an article or redirects to an article): reddening, urine, pottery, hall, brick, banquet, salt, shiver, This group is fits the majority of returned random words.
- Many words are not extant topics, but are defined concepts with multiple uses and derivatives, and should be disambiguation pages, eg: refund, exception, circumstance, brilliance, loop, stall. For these, the appearance of a new popular creative work, like a song, or a book, should not easily displace the DAB page.
Cases where I might support a title change:
- Examples of word searches that I found astonishing, I think they should be DAB pages, were collect, abridge,
- Examples of DAB pages where I think there is a PrimaryTopic, or the DAB page should become a DABCONCEPT: conference
--SmokeyJoe (talk) 01:54, 5 September 2018 (UTC)
- All kind of crazy shit can and does impact PRIMARYTOPIC decisions. What's important however is what should affect PRIMARYTOPIC decisions, which goes back to why we even have PRIMARYTOPIC, which is of course related to why we have disambiguation and how we choose our titles. I note that SmokeyJoe cites no policy or guideline basis for his wish that dictionary words impact PRIMARYTOPIC decisions.
- It's important to distinguish dictionary words for which there are encyclopedic topics considered sufficiently notable to have an article on WP from those that don't have coverage on WP. Smokey's first list of examples (reddening, et al), are mostly of dictionary words with meanings that have WP-covered topics. There is no dispute that words with dictionary meanings covered on WP should be given due consideration per the two PT criteria just like any other use on WP, assuming they are even ambiguous.
- Speaking of ambiguity, most of the words in his first list are not primary topics as he claims, as they are not even ambiguous (primary topic by definition applies to ambiguous terms). Regardless, even if a dictionary word is ambiguous with other notable uses, of course its use is considered in PT determination of that term. There is no dispute about that. The dispute is about whether the dictionary word should be given weight due to being a dictionary word. In other words, should we have a third PT criteria for that? Or should we just stick with the current two?
- Further, Smokey claims "words are not extant topics, but are defined concepts with multiple uses and derivatives". What does this mean? Of course words are not topics. Words refer to topics. The literal word "key" is not a topic - it refers to a variety of topics, none of which are considered primary. "Key" is not a single "defined concept". There are several defined concepts (aka topics) to which key refers, and each of those is a separate use of "key".
- The bottom line is that that dictionary-word meaning of a given ambiguous term is just another use to consider in deciding which if any one of them is primary. It should be given equal consideration under the two PT criteria relative to the other use or uses. We seem to have consensus about that. --В²C ☎ 17:50, 7 September 2018 (UTC)
- "I note that SmokeyJoe cites no policy or guideline basis for his wish that dictionary words impact PRIMARYTOPIC decisions."
- I note again that you seem blinkered, limited, by the wording the text on this guideline, and that this guideline has some long standing logical flaws. As a consequence, your logic is flawed.
- You admit that words with associated topics count for reasons for displacing a primary topic. Good. Now consider that a reasonable reader does not know the exact line between these words, and other word that may, but do not, have coverage somewhere in some article. It is even unreasonable to expect that readers know Wikipedia should not be used for looking words. It is more reasonable to have common words, for which there may be a multitude of associated topics and subtopics, to redirect to DAB pages that include all the covered topics and subtopics plus the consistently and prominently located Wiktionary box.
- Dictionary words should be given weight according to the rationalizations of editors. The guideline should reflect what editors usually decide. The guideline should not be used to contain reasonable arguments not directly couched from the guideline. Guidelines are imperfect, and this guideline has serious logical flaws.
- "The bottom line is that that dictionary-word meaning of a given ambiguous term is just another use to consider in deciding which if any one of them is primary."
- To my surprise, this is completely agreeable. It implies that NOTDICT is not a reason to discount another's argument, which is the reason for this thread, because that is what you have been trying to do.
- "It should be given equal consideration"
- Cut "equal". Human judgement is needed on levels of consideration.
- "under the two PT criteria"
- This is a subtly flawed reading of the guideline and its intent. The guideline does not list the criteria exhaustively. There are undocumented criteria that should and frequently do swing discussions. These include: Whether the audience of one topic is aware of the other topic; Whether one topic is etymologically derived from the other. --SmokeyJoe (talk) 01:11, 10 October 2018 (UTC)
“John Collins” and “John H. Collins”
Anyone is welcome to opine at Talk:John H. Collins. — Gorthian (talk) 01:05, 24 August 2018 (UTC)
Aegyptus (mythology)
An editor has recently created Aegyptus (mythology), which looks like a disambig page but it includes its own footnotes. I wonder if it should be combined with the older Aegyptus (disambiguation) or if the references would be better used in any of the connected mainspace articles. Thanks. ---DOOMSDAYER520 (Talk|Contribs) 18:30, 27 August 2018 (UTC)
- It's a set index article, so it should be moved to a set index title. -- JHunterJ (talk) 20:14, 27 August 2018 (UTC)
- It's not an optimal situation to have a handful of entries confusingly spread across two dab pages (nevermind that one of them can (or has been) reformatted into something that some people here would say is not one). It's best to merge that into Aegyptus (disambiguation) and not worry too much about the footnotes. – Uanfala (talk) 20:43, 27 August 2018 (UTC)
- Aegyptus (mythology) is the usual way of titling set index articles (they're not dab pages) about names of characters in Greek mythology. Sometimes there are more than a dozen distinct figures sharing one name. See Category:Set indices on Greek mythology or, just as one example, Clymene (mythology). There are probably hundreds of such articles. - Station1 (talk) 21:03, 27 August 2018 (UTC)
- @Station1: ...and very often those WP:SIAs on Greek mythology and prehistory are a confusing-to-readers unhelpful mess. Yes, I do know about and have consulted DGRBM. Narky Blert (talk) 05:04, 30 August 2018 (UTC)
- I don't disagree, but wouldn't combining them with dab pages just make things worse? Station1 (talk) 05:10, 30 August 2018 (UTC)
- @Station1: ...and very often those WP:SIAs on Greek mythology and prehistory are a confusing-to-readers unhelpful mess. Yes, I do know about and have consulted DGRBM. Narky Blert (talk) 05:04, 30 August 2018 (UTC)
- Aegyptus (mythology) is the usual way of titling set index articles (they're not dab pages) about names of characters in Greek mythology. Sometimes there are more than a dozen distinct figures sharing one name. See Category:Set indices on Greek mythology or, just as one example, Clymene (mythology). There are probably hundreds of such articles. - Station1 (talk) 21:03, 27 August 2018 (UTC)
Bot-creation of INTDABLINK redirects
Based on Widefox in the section above complaining about the lack of many "(disambiguation)" redirects to disambiguation pages, I've initiated a request for this at Wikipedia:Bot requests#Create WP:INTDABLINK redirects. Thryduulf (talk) 15:17, 30 August 2018 (UTC)
- Great Thryduulf! I don't think you believed me at the start, but glad we're making progress on agreed things now. Regards Widefox; talk 15:23, 30 August 2018 (UTC)
- @Widefox: It's not a question of not believing you - it's just that (disambiguation) redirects to disambiguation pages are not relevant to the discussion above. I initiated this only because you were expending complaining that nobody was fixing this problem than it would take for you to fix it. Thryduulf (talk) 10:01, 1 September 2018 (UTC)
- I will still want to delete the redirects created for name articles that are currently mistagged as dab pages. There aren't many of those left but the "useful for readers" argument will not wash if the redirect is brand new. —Xezbeth (talk) 06:50, 31 August 2018 (UTC)
- How many, handful or more than 100? Widefox; talk 13:34, 31 August 2018 (UTC)
- No idea, they're almost impossible to find besides just randomly stumbling across them. I'm just assuming there's not many left since I've fixed so many of them over the last ~6 years. —Xezbeth (talk) 16:17, 31 August 2018 (UTC)
- This query finds 4561 pages X in Category:Surnames or Category:Given names which are the target of a redirect called X (disambiguation). 90% are surnames. If "name articles" meant something else, we can probably amend the query to find them.name Certes (talk) 16:49, 31 August 2018 (UTC)
- If I just answered the wrong question, then PetScan for surnames or given names may be more useful. These lists includes many false positives, mainly real disambiguation pages having a People section amongst other meanings. Certes (talk) 16:58, 31 August 2018 (UTC)
- Those aren't the ones that are hard to find. There have been many name lists with just a dab template and nothing else. —Xezbeth (talk) 19:26, 31 August 2018 (UTC)
- Thanks for clarifying. If you can give any examples of such name lists, someone might be able to think of a way of searching for them. Certes (talk) 19:44, 31 August 2018 (UTC)
- Xezbeth, what does "fixing" these redirects amount to? – Uanfala (talk) 20:50, 31 August 2018 (UTC)
- One thing to watch out for is pages which are incorrectly tagged as {{disambig}} {{surname}} instead of {{disambig|surname}}. I've fixed dozens of those (sometimes by splitting a surname page out because the DAB page was getting unwieldy). (In my experience, {{given name}} problems are less common, by a factor of at least ten or twenty; perhaps because surnames are serious, but any fool can dream up a new given name which has never been used before and will never be used again.)
- I've lost count of the number of disambig pages which I've recategorised as name pages (and vice versa, often name -> {{hndis}}). I gave up routinely WP:G6-speedying (disambiguation) redirects to pure name pages after a couple of requests were declined on the grounds that they "weren't doing any harm". I have better things to do than to try to educate WP:ADMINs (except for those few who I've reverted for creating a WP:INTDABLINK error by mis-editing a hatnote or see-also link by removing the {disambiguation) qualifier from a link).
- Perhaps there is a case for a minor maintenance category? rather wordily entitled Category:Pages with (disambiguation) qualifiers which redirect to pages which are not disambiguation pages. I wouldn't search for candidate articles, but would cheerfully populate it whenever I fell across one. (I cycle through Disambiguation pages with links - 7th or 8th time through? - so I'm more likely to fall across such errors than most.) Narky Blert (talk) 02:01, 1 September 2018 (UTC)
- Here is an example that I deliberately haven't fixed yet: Danesi. If a bot creates (disambiguation) redirects to every dab page, it would obviously include ones like this that shouldn't be dab pages in the first place. —Xezbeth (talk) 05:57, 1 September 2018 (UTC)
- Why shouldn't that be a disambiguation page? It lists articles associated with "Danesi" - if we only had one article about someone with that name Danesi would redirect to it as {{R from surname}}, but we have multiple so disambiguation is needed. I can see the argument for replacing {{Disambig}} with {{hndisambig}} but a human name disambiguation is still a disambiguation page. Thryduulf (talk) 10:01, 1 September 2018 (UTC)
- This is indication of just how fundamental the misunderstanding is. Danesi is a list of people who happen to have the same surname. A disambiguation is most emphatically NOT list of
articles associated
with a term. Surnames are a gray area in that in some contexts individuals are often referenced by the surname alone, hence the guidance at WP:NAMELIST. {{hndisambig}} would never be the right template to use for a list of people with the same surname (or for that matter a list of people with the same given name). Yes, to a limited extent, such lists share some similarities with disambiguation pages, but expanding the confusion between disambiguation pages and other types of pages doesn't really help anyone. older ≠ wiser 10:21, 1 September 2018 (UTC)- The problem with that is that if I wanted to find an article about someone referred to in a source as "Danesi" with no knowledge of their first name, I would search for either "Danesi" or "Danesi (disambiguation)" if I thought it likely that the person I wanted was not going to be the primary topic - a list of people with the same surname functions as a disambiguation page. It is my experience as a reader that where there are multiple articles associated with the same term that there will be a page that disambiguates them, without noticing (or caring) whether that is technically a disambiguation page or not. Thryduulf (talk) 10:34, 1 September 2018 (UTC)
- Searching for "Danesi" is sensible.Searching for "Danesi (disambiguation)" presumes some familiarity with the arcanum of Wikipedia editing. No one (except perhaps some Wikipedia gnomes) would do a Google or Bing search for "Danesi (disambiguation)". Why should we deliberately reinforce the confusion between disambiguation pages and non-disambiguation pages merely for the convenience of some experienced (but misinformed) editors? older ≠ wiser 10:46, 1 September 2018 (UTC)
- No, nobody (or at least very few people) would enter "Danesi (disambiguation)" into google, but very many people who have read Wikipedia have encountered pages ending in (disambiguation) - they're linked right at the top of many of the most popular articles for example and they do enter such search terms in the internal search engine and/or navigate directly to is (e.g. last night I navigated directly to Steve (disambiguation) when looking for Steve (atmospheric phenomenon) as I knew it wasn't the primary topic but didn't know what the article was titled) (this comes up nearly every time one is nominated at RfD, and has been explained in detail in the RfC above). I've also yet to see any evidence that this supposed confusion between dabs and SIAs or lists of people with the same name actually impeding anybody finding the article they want. What exactly is the harm caused? If someone is searching for "Danesi (disambiguation)" they are clearly looking for a list of articles associated with that title - whether that is provided by a disambiguation page, a set index article, a name page or something else doesn't matter. Thryduulf (talk) 11:39, 1 September 2018 (UTC)
- Well, then the RFC is perhaps misleading. What it should be proposing is to eliminate the distinction between disambiguation pages and set indices (and perhaps other types of pages that some editors find convenient to use for purposes similar to disambiguation pages. That would entail adding the __DISAMBIG__ magic word into the other templates so that those other types of pages will show up as requiring disambiguation when links to them are inadvertently created. older ≠ wiser 11:58, 1 September 2018 (UTC)
- The RfC is explicitly not proposing to eliminate the distinction, because there is no need to do so and the distinction makes sense in at least some cases. All that is required to gain all the benefits for readers is to allow (not require) "(disambiguation)" redirects to pages that function as indexes to articles that are reasonable targets for a given search term, whether they are technically disambiguation pages or not. Thryduulf (talk) 12:28, 1 September 2018 (UTC)
- No, that is the practical effect of what many supporters are arguing--that there is no functional difference between the types of pages so far as readers are concerned. So rather than dance around this, better to just be out front with it. older ≠ wiser 12:39, 1 September 2018 (UTC)
- Thryduulf
Why shouldn't that be a disambiguation page?
- Bkonrad is right - that is a fundamental confusion about the consensus at WP:D WP:MOSDAB and WP:SIA. Surnames are by consensus a generally tolerated invalid item on a dab page relegated to the bottom directly before the "Other uses" section and for larger numbers always split out into a surname article, such a page that is exclusively surname is never a dab. I agree with your sentiment that the distinction between SIAs and dabs may hinder navigation and addressing that would be more useful than proposing support of niche redirect search hack (by design a minority flow which is against the majority navigation route assumed by WP:PRIMARYTOPIC, so a niche search by niche users). Widefox; talk 13:24, 1 September 2018 (UTC)- WP:PRIMARYTOPIC works exactly as it should do, doesn't need changing, and is not relevant to this search term. "(disambiguation)" searches are used (by many more people than you seem to acknowledge) when people know what they want is not (or is unlikely to be) the primary topic. With geographical place names someone searching for the city in Canada will search directly for London, Ontario rather than go via London. When other topics are consistently named, people will do the same thing - e.g. Esperanto language got over 1000 hits in August despite the article being at Esperanto. People use "(disambiguation)" though when don't know what the article is titled - I gave the example of Steve (atmospheric phenomenon) above, and see also everything in Category:Redirects from incomplete disambiguation. Yes, fewer people want to read about a book for children than a colour, but that does not mean that the non-primary topic is less valid or less encyclopaedic, that we should make it unnecessarily harder for people to find what they are looking for, force them to first view a page they know they don't want, or require them to know whether the list is a disambiguation page, an SIA or a surname page before having visited. If you want to remove the distinction between them, go ahead and get consensus for that, but be clear that this would be a completely separate discussion and is not required to resolve the navigation problems identified. Thryduulf (talk) 16:08, 1 September 2018 (UTC)
- That long reply doesn't answer why you don't know disambiguation fundamentals as directly quoted (and it's pertinent - about SIAs and dabs) per WP:CLUE
not in terms of rhetoric but in terms of his or her understanding of the established policies and guidelines of Wikipedia and of knowing what works and what doesn't.
. Widefox; talk 18:26, 1 September 2018 (UTC)- Danesi is correctly categorised as a {{surname}} page of interest to Wikipedia:WikiProject Anthroponymy but not to Wikipedia:WikiProject Disambiguation, along with thousands of other surname pages. I see no problem of any kind whatsoever here, except perhaps strawman in some of the preceding posts. Narky Blert (talk) 20:58, 1 September 2018 (UTC)
- That long reply doesn't answer why you don't know disambiguation fundamentals as directly quoted (and it's pertinent - about SIAs and dabs) per WP:CLUE
- WP:PRIMARYTOPIC works exactly as it should do, doesn't need changing, and is not relevant to this search term. "(disambiguation)" searches are used (by many more people than you seem to acknowledge) when people know what they want is not (or is unlikely to be) the primary topic. With geographical place names someone searching for the city in Canada will search directly for London, Ontario rather than go via London. When other topics are consistently named, people will do the same thing - e.g. Esperanto language got over 1000 hits in August despite the article being at Esperanto. People use "(disambiguation)" though when don't know what the article is titled - I gave the example of Steve (atmospheric phenomenon) above, and see also everything in Category:Redirects from incomplete disambiguation. Yes, fewer people want to read about a book for children than a colour, but that does not mean that the non-primary topic is less valid or less encyclopaedic, that we should make it unnecessarily harder for people to find what they are looking for, force them to first view a page they know they don't want, or require them to know whether the list is a disambiguation page, an SIA or a surname page before having visited. If you want to remove the distinction between them, go ahead and get consensus for that, but be clear that this would be a completely separate discussion and is not required to resolve the navigation problems identified. Thryduulf (talk) 16:08, 1 September 2018 (UTC)
- If set index articles have no practical difference from disambiguation pages, and if the MOS for disambiguation pages is there for the practical benefit of the readers, then set index disambiguation pages should follow the disambiguation style guidelines. The insistence on not following the style guidelines on SIAs was made because they were supposedly practically different from disambiguation pages. -- JHunterJ (talk) 12:46, 13 September 2018 (UTC)
- Exactly. The elephant in the room is if some (class of) SIAs are just second class dabs. It's better to address that cause than layering a redirect hack on top of the underlying issue. Widefox; talk 13:52, 16 September 2018 (UTC)
- One of my bugbears is WP:SIA pages like Vostok (inhabited locality) (and I've seen several dozen pages similar to that). That isn't an SIA (a set of items of a specific type that also share the same (or similar) name) – that's a group of places which randomly happen to be in the same country and to have the same name (which is this case happens to be the Russian word for 'east'). That's {{geodis}} material. Springfield is a good DAB page; Springfield (inhabited locality) would be an exact parallel to Vostok (inhabited locality), and would be just plain silly. Narky Blert (talk) 20:24, 27 September 2018 (UTC)
- Narky Blert Hmm, that's a longer list. There's a few of these (inhabited locality) in Russia SIAs, e.g. Pechenga (inhabited locality) is smaller, both are created by User:Ezhiki. I see no need for a short list like that to have, essentially a bad version of WP:DOUBLEDAB. It should be merged into Pechenga and thus flag up incoming links (Pechenga, Russia Konstantin Korovin). Is WP:INCDAB the way forward for these or just merge, merge started here Talk:Pechenga Widefox; talk 12:23, 5 October 2018 (UTC)
- One of my bugbears is WP:SIA pages like Vostok (inhabited locality) (and I've seen several dozen pages similar to that). That isn't an SIA (a set of items of a specific type that also share the same (or similar) name) – that's a group of places which randomly happen to be in the same country and to have the same name (which is this case happens to be the Russian word for 'east'). That's {{geodis}} material. Springfield is a good DAB page; Springfield (inhabited locality) would be an exact parallel to Vostok (inhabited locality), and would be just plain silly. Narky Blert (talk) 20:24, 27 September 2018 (UTC)
- Exactly. The elephant in the room is if some (class of) SIAs are just second class dabs. It's better to address that cause than layering a redirect hack on top of the underlying issue. Widefox; talk 13:52, 16 September 2018 (UTC)
- Thryduulf
- No, that is the practical effect of what many supporters are arguing--that there is no functional difference between the types of pages so far as readers are concerned. So rather than dance around this, better to just be out front with it. older ≠ wiser 12:39, 1 September 2018 (UTC)
- The RfC is explicitly not proposing to eliminate the distinction, because there is no need to do so and the distinction makes sense in at least some cases. All that is required to gain all the benefits for readers is to allow (not require) "(disambiguation)" redirects to pages that function as indexes to articles that are reasonable targets for a given search term, whether they are technically disambiguation pages or not. Thryduulf (talk) 12:28, 1 September 2018 (UTC)
- Well, then the RFC is perhaps misleading. What it should be proposing is to eliminate the distinction between disambiguation pages and set indices (and perhaps other types of pages that some editors find convenient to use for purposes similar to disambiguation pages. That would entail adding the __DISAMBIG__ magic word into the other templates so that those other types of pages will show up as requiring disambiguation when links to them are inadvertently created. older ≠ wiser 11:58, 1 September 2018 (UTC)
- No, nobody (or at least very few people) would enter "Danesi (disambiguation)" into google, but very many people who have read Wikipedia have encountered pages ending in (disambiguation) - they're linked right at the top of many of the most popular articles for example and they do enter such search terms in the internal search engine and/or navigate directly to is (e.g. last night I navigated directly to Steve (disambiguation) when looking for Steve (atmospheric phenomenon) as I knew it wasn't the primary topic but didn't know what the article was titled) (this comes up nearly every time one is nominated at RfD, and has been explained in detail in the RfC above). I've also yet to see any evidence that this supposed confusion between dabs and SIAs or lists of people with the same name actually impeding anybody finding the article they want. What exactly is the harm caused? If someone is searching for "Danesi (disambiguation)" they are clearly looking for a list of articles associated with that title - whether that is provided by a disambiguation page, a set index article, a name page or something else doesn't matter. Thryduulf (talk) 11:39, 1 September 2018 (UTC)
- Searching for "Danesi" is sensible.Searching for "Danesi (disambiguation)" presumes some familiarity with the arcanum of Wikipedia editing. No one (except perhaps some Wikipedia gnomes) would do a Google or Bing search for "Danesi (disambiguation)". Why should we deliberately reinforce the confusion between disambiguation pages and non-disambiguation pages merely for the convenience of some experienced (but misinformed) editors? older ≠ wiser 10:46, 1 September 2018 (UTC)
- The problem with that is that if I wanted to find an article about someone referred to in a source as "Danesi" with no knowledge of their first name, I would search for either "Danesi" or "Danesi (disambiguation)" if I thought it likely that the person I wanted was not going to be the primary topic - a list of people with the same surname functions as a disambiguation page. It is my experience as a reader that where there are multiple articles associated with the same term that there will be a page that disambiguates them, without noticing (or caring) whether that is technically a disambiguation page or not. Thryduulf (talk) 10:34, 1 September 2018 (UTC)
- This is indication of just how fundamental the misunderstanding is. Danesi is a list of people who happen to have the same surname. A disambiguation is most emphatically NOT list of
- Why shouldn't that be a disambiguation page? It lists articles associated with "Danesi" - if we only had one article about someone with that name Danesi would redirect to it as {{R from surname}}, but we have multiple so disambiguation is needed. I can see the argument for replacing {{Disambig}} with {{hndisambig}} but a human name disambiguation is still a disambiguation page. Thryduulf (talk) 10:01, 1 September 2018 (UTC)
- Xezbeth, what does "fixing" these redirects amount to? – Uanfala (talk) 20:50, 31 August 2018 (UTC)
- Thanks for clarifying. If you can give any examples of such name lists, someone might be able to think of a way of searching for them. Certes (talk) 19:44, 31 August 2018 (UTC)
- Those aren't the ones that are hard to find. There have been many name lists with just a dab template and nothing else. —Xezbeth (talk) 19:26, 31 August 2018 (UTC)
- No idea, they're almost impossible to find besides just randomly stumbling across them. I'm just assuming there's not many left since I've fixed so many of them over the last ~6 years. —Xezbeth (talk) 16:17, 31 August 2018 (UTC)
- How many, handful or more than 100? Widefox; talk 13:34, 31 August 2018 (UTC)
@Widefox: I agree with you on Pechenga, and have voted in support. If a DOUBLEDAB seems advisable, e.g. because a DAB page is getting too long for comfort, it should be done properly using a secondary DAB page like Pechenga, Russia (currently a redirect! with several bad incoming links) not an SIA. I've seen a lot of 'Placename, Iran' pages like that, and they work well.
I routinely add {{R from incomplete disambiguation}} and {{R from ambiguous term}} tags (among others) to redirects I fall across while fixing links to DAB pages. It only takes a second or two, and helps any future reader of those invisible pages.
If you want another horror, try Gold (compilation album), yet another member of Category:Disambiguation pages with (qualified) titles. Narky Blert (talk) 13:02, 5 October 2018 (UTC)
- Yup, exactly. As for these, e.g. Vostok (inhabited locality), I'm doubtful "(inhabited locality)" is useful generally for an SIA or dab, unless there's a good argument for it? A country qualifier seems more useful in this case, and a dab rather than SIA seems more useful.
- Well this is a bit perverse Gold (compilation album) (disambiguation). Suggestions? Widefox; talk 14:04, 5 October 2018 (UTC)
- That one would function better as a set index or dynamic list. It could also be merged (back?) into Gold (disambiguation) since it isn't that big. —Xezbeth (talk) 14:22, 5 October 2018 (UTC)
- Agree a list, say SIA, as it's quite large for the dab. Widefox; talk 14:44, 5 October 2018 (UTC)
Done. I requested G6 for the redirect Gold (compilation album) (disambiguation) but it was declined by User:Nyttend "The target page disambiguates several dozen compilation albums entitled "Gold")". It's an SIA. User talk:SoWhy I've just read the close #RfC: INTDAB links to non-dab pages, and I'm not sure if that close puts enough weight on the fact that the redirects are created solely as a technical dependent page per consensus at WP:DABNAME ...when a disambiguation page...This type of redirect is used to indicate any intentional links to the disambiguation page, ...
(which seems more than just creation, but actual definition of this type of redirect existing to prevent erroneous links on dab pages - which never applies to SIAs, as they are lists WP:SIANOTDAB). In that context, the argument for them to be treated as a dependent page should be strong in itself, and the burden to change that on proposers. The question being if/why the redirects are/aren't explicitly marked as a dependent page, but a technicality that can and should be fixed. The close seems to invert the burden, which may be in line with the defaults for redirects, but against those defaults for dependent pages. So, given that the numbers of !votes seems slightly more for No (10?) than Yes (8/9?), I'm puzzled why the burden seems inverted and it isn't closed no consensus given the difficult balance of those two competing defaults, combined with no consensus for a new, secondary use as a search mechanism? Widefox; talk 12:50, 8 October 2018 (UTC)
- Agree a list, say SIA, as it's quite large for the dab. Widefox; talk 14:44, 5 October 2018 (UTC)
- That one would function better as a set index or dynamic list. It could also be merged (back?) into Gold (disambiguation) since it isn't that big. —Xezbeth (talk) 14:22, 5 October 2018 (UTC)
Palaichori
What's the best resolution for new dab Palaichori? The article previously at that title has been cut and pasted to Palaichori Oreinis. The village website suggests that Palaichori is the term for the two adjacent villages (or perhaps the two parts of one village), so perhaps it should be a not-very-broad concept article rather than a dab. Suggestions or remedial action welcome. Certes (talk) 10:33, 7 September 2018 (UTC)
- This history makes interesting reading. But unattributed cut and paste move of a long-standing multi-editor-edited page is not the solution. I suggest we need to revert the move and then tweak appropriately and make a proper WP:RM. PamD 13:40, 7 September 2018 (UTC)
- Thanks, Pam. I've reverted and left Palaichori Oreinis as a redirect for now. The article can probably be split without too much controversy, as I've suggested to the editor. Certes (talk)
Disambiguation in draft
I have a disambiguation page waiting review at Draft:Vantablack (disambiguation). I'm not sure if I went through the right venue to request disambiguation pages, but is someone able to review it? 2601:589:8000:2ED0:34D4:CDC1:9863:612F (talk) 21:21, 12 September 2018 (UTC)
- I put it into the article space. Thanks for creating it! -- JHunterJ (talk) 12:49, 13 September 2018 (UTC)
How many more page views constitute "more likely" and "much more likely"?
The criteria at WP:PRIMARYTOPIC includes the phrase, "much more likely than any other single topic, and more likely than all the other topics combined"
. One of the common ways this is determined is by looking at page views. I have three questions:
- Given two topics, A and B, about how many more page views does A have to get than B to qualify as
"much more likely"
to be sought than B? I know there is resistance to specifying exact numbers (not sure why, exact though somewhat-arbitrary numbers are used successfully for much more serious issues in the real world, like speed limits, ages to drive, consent, drink, etc. - can you imagine how many more disputes there would be in the real world if real world rules were as wishy-washy as WP rules are?), but how about at least some ranges? Getting 1.1 time as many page views, for example, I think clearly does not qualify as "much more likely". But what about 1.5 times as many? Twice as many? Three times as many? At what point is it clearly "much more likely" to be sought? - Given a number of topics, how many more page views does one have to get than all the others combined to qualify as "more likely than all the other topics combined"? Isn't merely getting more page views (technically even just one more) than the total number of page views all the other uses get combined, sufficient to qualify as
"more likely than all the other topics combined"
? - If we can get consensus on these two questions, should we provide guidance at PT accordingly?
Thanks, --В²C ☎ 21:13, 13 September 2018 (UTC)
- You need to stop obsessing about page views and start considering the other criteria. Page views are a useful way to help determine the significance of a topic, but nothing more. Apple is so obviously an apple that it doesn't matter how many page views the corporation gets. The page views are irrelevant.--Ykraps (talk) 23:10, 13 September 2018 (UTC)
- In past move discussions I've used "ten times more visits" as the minimum threshold to start considering page views as a significant factor with respect to usage, and I was not the only one. The defining factor though is not the relative amount of page views with respect other pages, but the relative strength of usage with respect to the historical significance of the other articles. I.e. it's not enough that one page gets much more visits (see Madonna), it's also needed that the others don't have a competing claim of primacy by other parameters. Diego (talk) 08:45, 14 September 2018 (UTC)
- Possibly unhelpful, but this is my observation: Page views don't necessarily indicate the reader found what they were looking for, there could be a multitude of people visiting the wrong page in some scenarios. If is a dab, what they click next is more significant. —cygnis insignis 08:53, 14 September 2018 (UTC)
- I would usually assume "much more" would mean something like 90-95%. This came from Talk:Memphis (disambiguation)#Requested move 4 September 2018 where it has been argued that the TN city doesn't clearly meet either criteria. I would point out B2C's post here saying that where Birmingham, Alabama has been viewed 32601 times and Birmingham has been viewed 79197 times, that not being enough. But Birmingham in the West Midlands is the original and the usual "City, State" format probably deals with that and people interested in the Alabama city surely know that its named after the UK one. Thises views [[11]] show a larger margin but for Memphis being less relevant as the TN city is English speaking, unlike the city in Egypt. I would personally recommend that we change it to something like this:
- Possibly unhelpful, but this is my observation: Page views don't necessarily indicate the reader found what they were looking for, there could be a multitude of people visiting the wrong page in some scenarios. If is a dab, what they click next is more significant. —cygnis insignis 08:53, 14 September 2018 (UTC)
- In past move discussions I've used "ten times more visits" as the minimum threshold to start considering page views as a significant factor with respect to usage, and I was not the only one. The defining factor though is not the relative amount of page views with respect other pages, but the relative strength of usage with respect to the historical significance of the other articles. I.e. it's not enough that one page gets much more visits (see Madonna), it's also needed that the others don't have a competing claim of primacy by other parameters. Diego (talk) 08:45, 14 September 2018 (UTC)
- A topic is primary for a term with respect to usage if it is the subject sought 90% or more 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 all the other topics associated with that term.
However we should take into account in situations where hatnotes would be enough, such as when there are only 2 or 3 subjects or when there are only 2 "main" topics. Crouch, Swale (talk) 08:55, 14 September 2018 (UTC)
- A broad consensus is not possible. It has to be judged on a case-by-case basis, via WP:RM, WP:RFD or a talk page. Every ambiguous term has a unique set of articles with a unique set of characteristics that need to be considered. —Xezbeth (talk) 11:07, 14 September 2018 (UTC)
- Yes it might be assessed up to a point on a case by case basis but a clearer guideline would reduce disputes and make assessment easier and reduce outcomes being based on who happens to be making the assessment. Crouch, Swale (talk) 11:17, 14 September 2018 (UTC)
- If you think clearer titling guidelines are needed, read busywork and find something more productive to do. —SmokeyJoe (talk) 13:56, 14 September 2018 (UTC)
- Yes it might be assessed up to a point on a case by case basis but a clearer guideline would reduce disputes and make assessment easier and reduce outcomes being based on who happens to be making the assessment. Crouch, Swale (talk) 11:17, 14 September 2018 (UTC)
- Google search results are not WP:RS. Nor are page views.
- "One of the common ways this is determined is by looking at page views." That statement is WP:OR, and is disqualified.
- Never lose sight of why we are here! It's not about us, it's about the readers. Solutions do not need problems – especially solutions founded on over-literal over-rigid readings of the guidelines which overlook WP:IAR.
- Which is worse? (1) A reader clicks and lands on a disambiguation page, and has to click again to find what they were looking for. (2) A reader clicks and lands on an ill-chosen WP:PTOPIC which confuses or misleads them.
- I have an informal test for PTOPIC, which has nothing to do with ephemera like page views and search hits. How likely is it that a reader who knows of topic A also knows of topic B? If the answer is 'overwhelmingly', then B is PTOPIC; otherwise, it is not.
- If you would prefer to sort out a disaster area rather than to count the angels dancing on head of a pin, I suggest that you look at the incoming links to Esplanade. Narky Blert (talk) 20:13, 17 September 2018 (UTC)
- Very well put. Which is worse? (1) A reader clicks and lands on a disambiguation page, and has to click again to find what they were looking for. (2) A reader clicks and lands on an ill-chosen WP:PTOPIC which confuses or misleads them. I might cite that elsewhere.
- And I also like your informal test for PTOPIC, always have, I think it's far more useful than page views, see User:Andrewa/The Problem With Page Views. But shouldn't it be symmetrical? What if it's also overwhelmingly likely that a reader who knows of topic B also knows of topic A? But if it's true one way but not the other, then agree. Which would make New York City clear P T of New York for example. Andrewa (talk) 13:56, 15 October 2018 (UTC)
What is the purpose of WP:PRIMARYTOPIC?
What is the purpose of primary topic? We have (rather vague - see discussion just above this one) guidelines for determining primary topic, but we really don't say why any ambiguous name should ever be the title of any article. What problem does this solve? Why is this better than always have a dab page at every ambiguous base name as some seem to favor? Or is it even better?
The reason I ask is I think stating a purpose might help develop consensus about whether a primary topic exists in a given situation. So I'd like to hear what others think about the purpose of primary topic. Don't be shy about stating the obvious. Just because it's obvious to you doesn't mean it is to others. Thanks, --В²C ☎ 00:24, 2 October 2018 (UTC)
- Purpose? I think the main purpose was to make it easier on average for editors to wikilink. A thoroughly unworthy purpose in my opinion. I think nothing would be lost by any reader if Paris were located at Paris, France. --SmokeyJoe (talk) 04:15, 2 October 2018 (UTC)
- Well, that explains much. Do you believe making wikilinking easier is also the point of WP:CONCISION? —В²C ☎ 07:10, 2 October 2018 (UTC)
- No, concision is good advice for any message. It’s good for all writing. Minimise the words to convey the same information. Some go too far though, confusing concision with brevity. —SmokeyJoe (talk) 08:53, 2 October 2018 (UTC)
- I thought the main purpose was to get readers onto the article their looking for as smoothly as possible as well as avoiding surprise. Crouch, Swale (talk) 08:56, 2 October 2018 (UTC)
- Exactly. A major purpose of a primary topic is to take readers where they expect to go when they type that topic into the search box. If we moved Paris to Paris, France, what would we do with the title Paris? Redirecting to the city article that we just moved wouldn't achieve much other than changing the title displayed above the article: all wikilinks would continue to lead to their current destinations, so it's still PT by the back door. Moving Paris (disambiguation) to Paris might do more harm than good, because 90%[citation needed] of readers who type "Paris" would be ASTONISHed to land on a dab. We'd also get a plethora of links-to-dab to fix, because new editors (and those set in their ways) would continue to write "Durand was born in Paris". That's not just extra work for editors; it's inconveniencing readers until someone gets around to fixing it. I think we should use primary topics less often, and I often !vote for a dab to go at the base name when others see a PT, but they certainly have their place. Certes (talk) 09:11, 2 October 2018 (UTC)
- That’s criticism of MALPLACED, not support of PT.
- Exactly. A major purpose of a primary topic is to take readers where they expect to go when they type that topic into the search box. If we moved Paris to Paris, France, what would we do with the title Paris? Redirecting to the city article that we just moved wouldn't achieve much other than changing the title displayed above the article: all wikilinks would continue to lead to their current destinations, so it's still PT by the back door. Moving Paris (disambiguation) to Paris might do more harm than good, because 90%[citation needed] of readers who type "Paris" would be ASTONISHed to land on a dab. We'd also get a plethora of links-to-dab to fix, because new editors (and those set in their ways) would continue to write "Durand was born in Paris". That's not just extra work for editors; it's inconveniencing readers until someone gets around to fixing it. I think we should use primary topics less often, and I often !vote for a dab to go at the base name when others see a PT, but they certainly have their place. Certes (talk) 09:11, 2 October 2018 (UTC)
- I thought the main purpose was to get readers onto the article their looking for as smoothly as possible as well as avoiding surprise. Crouch, Swale (talk) 08:56, 2 October 2018 (UTC)
- No, concision is good advice for any message. It’s good for all writing. Minimise the words to convey the same information. Some go too far though, confusing concision with brevity. —SmokeyJoe (talk) 08:53, 2 October 2018 (UTC)
- Well, that explains much. Do you believe making wikilinking easier is also the point of WP:CONCISION? —В²C ☎ 07:10, 2 October 2018 (UTC)
- “wouldn't achieve much other than changing the title displayed above the article”? That something is significant, but it is not all. It would also achieve precision in urls, hovertext, category listings, and any number of downstream uses. —SmokeyJoe (talk) 09:29, 2 October 2018 (UTC)
- SmokeyJoe, well, if the article at Paris was moved to Paris, France but Paris remained a WP:PRIMARYREDIRECT to Paris, France, that city would still be treated as the primary topic and its purpose would be served identically. The question here (in terms of your example) is what is the purpose of having the city at Paris (directly or indirectly) rather than having the dab page there? --В²C ☎ 15:37, 2 October 2018 (UTC)
- There is logic to PrimaryTopic, but getting readers to the page they want is not it. Now you are moving to a different question “where should dab pages be located”. Paris is a good example of a valid PrimaryTopic, as everyone who knows any Paris knows that Paris, France is a bigger topic, and so it is logical that the dab page should not be at the base name. —SmokeyJoe (talk) 22:21, 2 October 2018 (UTC)
- You believe getting users to the page they want is NOT what PT is about? Then how do you explain,
...much more likely than any other single topic, and more likely than all the other topics combined—to be the topic sought when a reader searches for that term
? And what do you believe it IS about? Are you suggesting that for a given term X, only a topic with name X known by anyone who knows any X should qualify as the PT for X? Why does it matter whether everyone knows about it? What if everyone knows about it, but very few ever look it up? And how would you even know how many "know" about what? Using page view counts to determine how likely someone is to seek one topic relative to the others is far more useful and is completely objective. --В²C ☎ 23:30, 2 October 2018 (UTC)
- You believe getting users to the page they want is NOT what PT is about? Then how do you explain,
- There is logic to PrimaryTopic, but getting readers to the page they want is not it. Now you are moving to a different question “where should dab pages be located”. Paris is a good example of a valid PrimaryTopic, as everyone who knows any Paris knows that Paris, France is a bigger topic, and so it is logical that the dab page should not be at the base name. —SmokeyJoe (talk) 22:21, 2 October 2018 (UTC)
- "You believe getting users to the page they want is NOT what PT is about? Then how do you explain, ...much more likely than any other single topic, and more likely than all the other topics combined—to be the topic sought when a reader searches for that term?"
- I explain that as a very poor guideline. Who wrote it? Who is psychoanalysing what readers? Where is the issue of systematic bias, the fact that Americans have far easier access to fast devices and connectivity than Borneans?
- "Are you suggesting that for a given term X, only a topic with name X known by anyone who knows any X should qualify as the PT for X?"
Almost, but within reason. Not "anyone", but "any reasonable group of readers"
- "Why does it matter whether everyone knows about it?" Because it is not OK to be routinely ASTONISHing reasonable groups of readers with topics they have never heard of under titles that they expect to be a title for a topic they do know about.
- "What if everyone knows about it, but very few ever look it up?" I think: Give up on what you think people might look up. Not only is that not reliable, it is not even data. Titles should be decided based on sources. A topic is only a local topic if the topic's sources have only local distribution.
- "And how would you even know how many "know" about what?" Look at the the distribution of publications that are used as sources for the article. Give some consideration to missing sources, and add the missing sources before suggesting a new title.
- "Using page view counts to determine how likely someone is to seek one topic relative to the others is far more useful and is completely objective"
- No, it is not. --SmokeyJoe (talk) 00:49, 10 October 2018 (UTC)
- (edit conflict) Many topics are so clearly what 99% of readers are looking for that they will look for it at the name. 99% of people looking for information on Paris, France, will look for it at Paris because the other Parises are something people will not look for at Paris. That's why Paris has 202,628 pageviews in the last 30 days and Paris (disambiguation) has 3,303 (okay, that's 98.4% if you want to be picky). Making Paris the DAB will be a disservice to almost all readers, not to mention break thousands or millions of incoming links from external sites. Other examples needed? Okay:Munich 104,517 views vs. Munich (disambiguation) 1,102 viewsBerlin 146,025 views vs. Berlin (disambiguation) 1,405 viewsRome 150,422 views vs. Rome (disambiguation) 1796 viewsetc.As PRIMARYTOPIC makes clear, the purpose of the guideline is mainly service to the readers by providing the topic that is highly likely to be "the topic sought when a reader searches for that term". Regards SoWhy 09:24, 2 October 2018 (UTC)
- What about Unlikely? are readers not more Likely to search for Probability when searching for "Unlikely", see that RM especially that related terms such as Certainly and Definitely work. Crouch, Swale (talk) 09:30, 2 October 2018 (UTC)
- The number of hits to the DAB VS PTOPIC are interesting, see these views from this comment. Notice how the Lewis DAB page got only over 17x the views compared to 100x for some others. Crouch, Swale (talk) 09:36, 2 October 2018 (UTC)
- The question was why do we have PTOPIC, not is PTOPIC always applied correctly. Many terms we use in guidelines and policies are necessarily vague but that doesn't mean the guideline is problematic. In the end, it's up to consensus to determine in which cases a DAB is preferable to a PT. Regards SoWhy 11:51, 2 October 2018 (UTC)
- The number of hits to the DAB VS PTOPIC are interesting, see these views from this comment. Notice how the Lewis DAB page got only over 17x the views compared to 100x for some others. Crouch, Swale (talk) 09:36, 2 October 2018 (UTC)
- What about Unlikely? are readers not more Likely to search for Probability when searching for "Unlikely", see that RM especially that related terms such as Certainly and Definitely work. Crouch, Swale (talk) 09:30, 2 October 2018 (UTC)
- WP:PRIMARYTOPIC says:
- A topic is primary for a term with respect to usage if it is highly likely—much more likely than any other single topic, and more likely than all the other topics combined—to be the topic sought when a reader searches for that term.
- I think that is sufficient. Do we really need to spell out that the reason we care what readers are seeking is that we want them to find it? PrimeHunter (talk) 16:35, 2 October 2018 (UTC)
- PrimeHunter, based on the arguments I see in RM discussions almost every day that don't seem to recognize this purpose of PT, and some made above in this discussion, yes, I think we do have to spell out that the purpose of PT is to help readers get to the topics they are seeking. Not to mention how the "historical significance" criteria has muddied the waters of what you have quoted. --В²C ☎ 17:55, 2 October 2018 (UTC)
A currently active RM at Talk:Mario Beaulieu (politician)#Requested move 18 September 2018 may be relevant in that it posits that a comparison between two same-named regional (not national) politicians shows that because one has 17 times more views than the other, the one with more views should be the primary topic. Neither politician, however, has an international reputation, thus raising doubt as to whether a primary topic is called for. Roman Spinner (talk • contribs) 19:04, 2 October 2018 (UTC)
- "has international reputation" isn't a PT criterion. 17 times the views and indistinguishable historicality is a good indication of calling for a primary topic. -- JHunterJ (talk) 20:28, 2 October 2018 (UTC)
- It is in practice, and it should be explicitly. An American niche fan topic, unknown outside the American fan base, should not be PrimaryTopic if there is a globally known less fanatic topic known by the same name. The fans will find their topic even if it has a slightly longer title. Page views are a good an easy indicator, but there may be reason to overlook this indicator. —SmokeyJoe (talk) 22:28, 2 October 2018 (UTC)
- This gets to the heart of what I'm asking about. "An American niche fan topic, unknown outside the American fan base, should not be PrimaryTopic if there is a globally known less fanatic topic known by the same name." Why? This gets back to the purpose of PT, which is what I'm asking about. Yes, the "fans will find their topic even if it has a slightly longer title", but they will have to do so only after clicking through a dab page. Isn't that what we're trying to avoid? If not, then why have PTs at all? Why not just have a dab page there every time? See, PrimeHunter? This is why we have to spell it out. Let's not dilly dally around the apparently obvious. Let's be clear. --В²C ☎ 23:19, 2 October 2018 (UTC)
- “Why?” is ASTONISH and meta:Introduction to Structurism (the encyclopaedia should have a logical structure). The fans will not need extra clicks to get to their precise (nb minimally precise) title, unless someone does something silly like follows MALPLACED and puts the dab page at a catchy basename. Where different groups may separately and reasonably think their different topics are both PrimaryTopics, it is most important that the dab page does not go at the base name. I am still yet to hear a single reason why all DAB pages are not better suffixed “(disambiguation)”. The PT discussions are distorted due to the MALPLACED policy error. An imperfect PrimaryRedirect is far less bad than to put either article under the basename title that google (and all sorts of other ways in) will take readers to the wrong page. —SmokeyJoe (talk) 23:30, 2 October 2018 (UTC)
- You are also talking about a different case, at least. An American niche fan topic, unknown outside the American fan base, would still be PrimaryTopic if there is another less fanatic topic also not known globally by the same name. The case of "Mario Beaulieu" involves two topics with the same "fan base" (Canadian regional politics), one of whom vastly outstrips the other in usage. -- JHunterJ (talk) 15:49, 4 October 2018 (UTC)
- JHunterJ wrote: "An American niche fan topic, unknown outside the American fan base, would still be PrimaryTopic if there is another less fanatic topic also not known globally by the same name". I disagree. If the other, less fanatic (do you mean less ghits/pageviews?), for example a Bornean niche fan topic of coincidental same name, exists, then its fans can be expected to be ASTONISHED if taken to the unrelated American topic. Hatnotes, for an example a hatnote at the American topic just for the Borneans, is an ugly solution, impacting the article, and not necessarily understood for by Borneans. If one group may be ASTONISHED, there is not PrimaryTopic. Both should be titled per PRECISION.
- The Mario Beaulieus is a much less interesting case because the audience of one should be aware of the other, meaning not ASTONISH factor. However, one of them remains titled in violation of PRECISE. --SmokeyJoe (talk) 00:36, 10 October 2018 (UTC)
- The titling isn't up to WP:D, but to WP:AT and the appropriate project naming conventions. And I disagree that fans of any niche topic will be astonished to discover that their niche topic isn't the primary topic for any give title, even if the primary happens to be a topic in some other niche. -- JHunterJ (talk) 12:15, 10 October 2018 (UTC)
- This gets to the heart of what I'm asking about. "An American niche fan topic, unknown outside the American fan base, should not be PrimaryTopic if there is a globally known less fanatic topic known by the same name." Why? This gets back to the purpose of PT, which is what I'm asking about. Yes, the "fans will find their topic even if it has a slightly longer title", but they will have to do so only after clicking through a dab page. Isn't that what we're trying to avoid? If not, then why have PTs at all? Why not just have a dab page there every time? See, PrimeHunter? This is why we have to spell it out. Let's not dilly dally around the apparently obvious. Let's be clear. --В²C ☎ 23:19, 2 October 2018 (UTC)
- It is in practice, and it should be explicitly. An American niche fan topic, unknown outside the American fan base, should not be PrimaryTopic if there is a globally known less fanatic topic known by the same name. The fans will find their topic even if it has a slightly longer title. Page views are a good an easy indicator, but there may be reason to overlook this indicator. —SmokeyJoe (talk) 22:28, 2 October 2018 (UTC)
The purpose of WP:PRIMARYTOPIC as a guideline is driving readers to an article they're likely to be looking for, but also not driving them to a topic they're unlikely to be looking for (that's why WP:ASTONISH is explicitly mentioned). That's a reason why page views is a poor indicator of primacy on its own, and why historical significance is an essential criterion to take into account.
When there are several split target audiences that may be looking for different topics, that one article is slightly more popular for one group does not provide any evidence as to what other demographic groups may be looking for. Someone recently mentioned "even if I'm not looking for the primary topic, at least I've heard of it before", and I think that's a good criterion. Diego (talk) 16:13, 4 October 2018 (UTC)
- Paris? Wasn't he Helen's boyfriend? or a character in Romeo and Juliet?
- IMO a WP:PTOPIC needs to be at least 90% and preferably 99% of all search requests. I see far too many WP:RM proposals based on marginal differences or on WP:RECENTISM.
- Always remember why we are here. It is not to enforce rules. It is to help readers to find what they are looking for, with accuracy and certainty.
- Hindu deity Krishna shares his name with Krishna (Malayalam actor) and Krishna (Telugu actor). I wonder how many of the incoming links to Krishna suggest that the deity (who, by my understanding, is dead) has starred in movies? Narky Blert (talk) 06:15, 5 October 2018 (UTC)
- I'm working through those but there are probably only a few dozen bad links. Even in films, most links are to the original Krishna who is a popular character. I don't think I've yet found two with the same correct target; the first one was meant for Krishna (Kannada actor) who wasn't even on the dab. The most numerous fixes I've found in recent trawls are President → President (corporate title) (1644), Republic of China → Republic of China (1912–1949) (970), Adventure → Adventure film (443) and Family → Family (biology) (405). There are lots of bad links to Alcohol (meaning drinks rather than the type of chemical) but the best destination is controversial. Certes (talk) 10:33, 5 October 2018 (UTC)
Some very good points above. I have for some time been working on a proposal to deprecate Primary Topic, see User:Andrewa/Primary Topic RfC for my latest thoughts. In actual scenarios, it generally does more harm than good... perhaps always! Some of this is counter-intuitive, but when I actually work on examples, that's my conclusion.
I'm not ready to propose it yet, but if you'd like to help refine it, the place to go is User talk:Andrewa/P T examples and scenarios and of course User talk:Andrewa/Primary Topic RfC. Andrewa (talk) 01:00, 7 October 2018 (UTC)
- This is exactly what I'm getting at. A minority but nevertheless many of the arguments I see in RM discussions imply that PT really has no purpose. This, PrimeHunter, is why I think we need clarification about the purpose of PT. The bottom line is how are readers best served when one enters an ambiguous term in the Search box and presses return, and the topic of one article
is highly likely—much more likely than any other single topic, and more likely than all the other topics combined—to be the topic sought when a reader searches for that term
? --В²C ☎ 20:53, 7 October 2018 (UTC)- Disagree that your bottom line is the bottom line. Also, your bottom line is far from objective, and suffers from biases.
- A failing of erring on granting PT is that an ambiguous term becomes available in the autofill Go box.
- A failing of MALPLACED is that it puts a DAB page on the ambiguous term. It is this MALPLACED old bad idea that I think has caused so many to want to be overzealous in grant PTs.
- PRIMARYREDIRECT errors are minor compared to bad, PRECISE-failing titles. —SmokeyJoe (talk) 21:55, 7 October 2018 (UTC)
- The fact that an experienced editor like you thinks the much older concept of PT is somehow supported by the much younger concept of MALPLACED is just another reason why the purpose of PT needs to be clarified. —В²C ☎ 20:21, 8 October 2018 (UTC)
- Actually, I think your beef is with WP:DABNAME, upon which MALPLACED is founded. —В²C ☎ 20:29, 8 October 2018 (UTC)
- Not quite, but close. There is a logic to PT, but the driver for many bad PT decisions seems to be MALPLACED / DABNAME. If the DAB page on the basename would cause people to unwanting load a DAB page, the solution is to title the DAB unambiguously. Why not? Getting people to the page they want is a reason to support more emphasis on PRECISE and RECOGNISABILITY. DAB pages on basenames fail both. —SmokeyJoe (talk) 21:29, 8 October 2018 (UTC)
- There's logic to P T, but it's faulty. It's based on the assumption that having the article that the reader wants at the base name is going to make it easier for them to find it. It seems intuitively obvious that this is true, but unfortunately it's not. That's just one of the problems with P T as it is now, but if we can muster the evidence and get (I predict very grudging) consensus on that, then possibly the whole house of cards comes down.
- Alternatively, there are a couple of related problems with disambiguation that could be fixed independently and the process might stir people's mental juices. Watch this space. Andrewa (talk) 22:38, 8 October 2018 (UTC)
- Watching. Interesting for sure. I tend to agree that the logic for PT is flawed. Paris is a very strong contender for PT, yet no reader would be harmed if it were at Paris, France, with nothing at Paris. Google, wikilinks, autocomplete, never take readers to redlinks. If it must be blue (why?), it could redirect to Paris, France. PRIMARYREDIRECTs are fine, they don't force imprecise titles. Paris is strong for a PT because everyone multiple criteria support it. Page views. Long term signficance (which does not mean "oldest"). Does not derive from another important use. Everyone who has heard of another Paris should know of the French city. WP:PT would be much safer from astonishing anyone if ALL of these criteria were required. One, eg pagviews, others failing, is not good enough. --SmokeyJoe (talk) 00:21, 10 October 2018 (UTC)
- Exactly. That would be a great improvement. But there seems no chance of it happening, we have tried many times. So my at first reluctant conclusion became, nothing lost by going for the more radical and far simpler option of deprecating the concept. But the more I developed this idea, the more attractive it became. It has surprised me how much it has going for it. And (at the risk of being boring) the most remarkable thing is, it's not to the readers' advantage to have an article at an ambiguous base name. It seems obvious that it is. But it's not. Andrewa (talk) 10:50, 13 October 2018 (UTC)
- Watching. Interesting for sure. I tend to agree that the logic for PT is flawed. Paris is a very strong contender for PT, yet no reader would be harmed if it were at Paris, France, with nothing at Paris. Google, wikilinks, autocomplete, never take readers to redlinks. If it must be blue (why?), it could redirect to Paris, France. PRIMARYREDIRECTs are fine, they don't force imprecise titles. Paris is strong for a PT because everyone multiple criteria support it. Page views. Long term signficance (which does not mean "oldest"). Does not derive from another important use. Everyone who has heard of another Paris should know of the French city. WP:PT would be much safer from astonishing anyone if ALL of these criteria were required. One, eg pagviews, others failing, is not good enough. --SmokeyJoe (talk) 00:21, 10 October 2018 (UTC)
- Not quite, but close. There is a logic to PT, but the driver for many bad PT decisions seems to be MALPLACED / DABNAME. If the DAB page on the basename would cause people to unwanting load a DAB page, the solution is to title the DAB unambiguously. Why not? Getting people to the page they want is a reason to support more emphasis on PRECISE and RECOGNISABILITY. DAB pages on basenames fail both. —SmokeyJoe (talk) 21:29, 8 October 2018 (UTC)
- Simplicity we could eliminate all primary topics, and use primary redirects. PT (without a redirect) is the simplest way to get a reader to the article, and similarly for writers. It's not just about navigation, but complexity. There may be reasons to increase complexity, but if search needs improving it's easier to improve search than make the structure more complex than needed (that's my gut reaction). Widefox; talk 12:01, 9 October 2018 (UTC)
- Interestingly, could be a good idea. PRIMARYREDIRECTs do not conflict with PRECISE. PTs do. --SmokeyJoe (talk) 00:24, 10 October 2018 (UTC)
- Yes, it's a very good idea. It's a more complex solution than deprecating P T, and only solves half the problem IMO. But it does address an important issue, and would mean that every article was at an unambiguous name, so I'd support it.
- It's actually a very good idea. It just mustn't happen. [12] Andrewa (talk) 03:22, 10 October 2018 (UTC)
- Just to clarify, that wasn't a proposal, but highlighting how using primary redirects to eliminate all primary topics just makes things more complicated. It's just an unnecessary layer of indirection. I prefer simple, let alone stable namespace. Take one scenario - we shouldn't have to move Paris, just because there's a new film called Paris. Ambiguous titles are fine if we "WP:BALANCE" the probability. We trade off ambiguity for complexity. Any proposal needs to take any trade offs into account. Widefox; talk 13:27, 12 October 2018 (UTC)
- Interestingly, could be a good idea. PRIMARYREDIRECTs do not conflict with PRECISE. PTs do. --SmokeyJoe (talk) 00:24, 10 October 2018 (UTC)
Simplicity
Simplicity we could eliminate all primary topics, and use primary redirects. PT (without a redirect) is the simplest way to get a reader to the article, and similarly for writers. It's not just about navigation, but complexity. There may be reasons to increase complexity, but if search needs improving it's easier to improve search than make the structure more complex than needed (that's my gut reaction). [13]
As an RM regular, I can assure you that P T as it currently stands is anything but simple. See wp:NYRM for the specific example that eventually led to my wondering, is P T really worth it? Our eventual (and I hope permanent) solution was, DAB at base name. But this is arguably contrary to current policy, as New York City is probably the P T of New York. It took twelve years and some very hard work and bitter discussion just to move the state article away from the base name, despite consistent consensus that the state was not the P T. And it cost us several editors over those years. I would never have been involved myself had I known that NYRM2016 would fail as it did! It still astounds me, and was probably not in glorious hindsight worth it.
And that was the start of a journey for me. And the real breakthrough came when I finally questioned the common assumption that having an article at its base name advantages those looking for it. In fact it seems to me that the opposite is true. I'm gathering examples at User talk:Andrewa/P T examples and scenarios and it's early days yet, much analysis to be done before the RfC is ready. But input there very welcome!
In my current view, deprecating P T is a very desirable simplification! But as I said, still early days. Andrewa (talk) 19:29, 9 October 2018 (UTC)
- Just in case this isn't clear - the end result is simplicity. The process to get there is a different question entirely. Widefox; talk 21:43, 9 October 2018 (UTC)
- A very important distinction, and I've phrased my post badly I guess if you see the need for that clarification. It's the end result that most interests me.
- P T complicates things at all levels. But I suspect this is as counter-intuitive to others as it was to me. That's why I give the history above. Andrewa (talk) 23:45, 9 October 2018 (UTC)
- OK. I was looking at PT afresh. The premise that PT complicates things needs reasoning, as using an ambiguous title for a most likely topic reduces the complexity to that when there's no ambiguity. It's pragmatic. In comparison with EB we seem to me much better organised, specifically for navigation with dabs and concise titles. From what I can understand, this the motivation comes largely from a concern about search? Widefox; talk 12:53, 10 October 2018 (UTC)
- (I've taken to using the abbreviation P T, see WP:PT for why.)
- Agree that The premise that PT complicates things needs reasoning. And that's at User:Andrewa/Primary Topic RfC (and its three predecessors to which it links), and discussion at User talk:Andrewa/Primary Topic RfC would be welcome. Particularly, we might start a section there on the claim that P T simplifies things.
- ...using an ambiguous title for a most likely topic reduces the complexity to that when there's no ambiguity... No. It simplifies things for readers who are (1) looking for that article and (2) unaware that other meanings of that title may be considered primary by others. For all other readers (including of course any who don't agree with us on what the P T is), it complicates things. Most of the damage can be addressed by redirects, as discussed elsewhere. But this isn't simplicity. The simplest method is, have a DAB (or a redirect to a more general DAB) at every ambiguous article title. But that's not necessarily the best method. Andrewa (talk) 20:04, 10 October 2018 (UTC)
- The simple rule of treating all ambiguous titles equally does not lead to a simple outcome, no. We should put readers first, and PT discussions are irrelevant for readers. If the PT is wrong, then it can be fixed. The principle remains. Nope - it is right to give due weight to topics. It is that simple. Anything else is, by definition, undue weight. Equal weight is just a special case of undue weight - it guarantees that irrespective of how important the topic is, the title changes whatever new (ambiguous) article is created or retitled. That affects stability, expectation, and is the opposite of the simplicity of getting gold when inputting "gold". Widefox; talk 11:59, 15 October 2018 (UTC)
- Well put, and I'm sure others are thinking along similar lines without putting it nearly so well! It's a intuitively attractive approach, and the benefits are obvious. But what seems obvious is not always true.
- Yes, it seems simple to regard some topics as more important and give them extra weight by giving them the article base names even if those are ambiguous, and disambiguating only the less important topics. It seems intuitively obvious that this is a good idea. And it worked well for the first few years of Wikipedia, when most of the articles to be written were about these important topics. But as the years have gone by, it has become problematic and not simple at all. I think this is important enough to reset the indenting. Andrewa (talk) 12:38, 15 October 2018 (UTC)
- The simple rule of treating all ambiguous titles equally does not lead to a simple outcome, no. We should put readers first, and PT discussions are irrelevant for readers. If the PT is wrong, then it can be fixed. The principle remains. Nope - it is right to give due weight to topics. It is that simple. Anything else is, by definition, undue weight. Equal weight is just a special case of undue weight - it guarantees that irrespective of how important the topic is, the title changes whatever new (ambiguous) article is created or retitled. That affects stability, expectation, and is the opposite of the simplicity of getting gold when inputting "gold". Widefox; talk 11:59, 15 October 2018 (UTC)
- OK. I was looking at PT afresh. The premise that PT complicates things needs reasoning, as using an ambiguous title for a most likely topic reduces the complexity to that when there's no ambiguity. It's pragmatic. In comparison with EB we seem to me much better organised, specifically for navigation with dabs and concise titles. From what I can understand, this the motivation comes largely from a concern about search? Widefox; talk 12:53, 10 October 2018 (UTC)
- Just in case this isn't clear - the end result is simplicity. The process to get there is a different question entirely. Widefox; talk 21:43, 9 October 2018 (UTC)
Problem number one: Controversial cases in which the primary topic is not so easily determined. In the early days, these were few and far between, and a minor annoyance at worst, although we've always had some such as Macedonia. But as the years have gone by and the article count has grown, they have become more and more numerous. In a larger encyclopedia in which articles on most of the trivially primary topics (such as Mathematics and London) have already been written, it becomes more of an issue. It is no longer either the simple method or the simple outcome.
Problem number two: In glorious hindsight, it was never terribly logical. Intuition can be misleading! When we actually do some case studies, we find that having an article at an ambiguous name is not generally helpful. Intuition tells us that it's going to be easier for readers to find the article on Capitol Limited (Amtrak train) if we move it to Capitol Limited. In fact, the opposite is the case. See #A current case in point below. Andrewa (talk) 12:38, 15 October 2018 (UTC)
Related issues
As promised. Get the brain juices going on these relatively simple proposals. And they might even give you a new perspective on P T. Andrewa (talk) 00:38, 9 October 2018 (UTC)
Unambiguous redirects
Whenever an article is moved from an unambiguous, disambiguated name to the base name, a redirect is created, and that is good. I'll call these unambiguous redirects.
But articles that have never been moved, that is they've always been at these ambiguous base names, typically have no such redirects. This means that, in a search results list, there may be no result that unambiguously identifies the topic to the reader. This is an obvious (or is it?) hindrance to readers. They must guess that the article that they want is at the base name, and certainly many will do this. But not everyone, particularly if there's a long results list.
We should as a matter of policy have an unambiguous redirect pointing to every article that is at an ambiguous name. (I'd prefer that we didn't have any articles at these names. But this will prevent nearly all of the damage, and at no downside that I can see.) Andrewa (talk) 00:38, 9 October 2018 (UTC)
- Andrewa Titles, navigation and search are a bit of a trilemma. Redirects being a secondary issue. If the aim is to improve search, improving the software may help without having to touch titles and navigation. Widefox; talk 10:00, 9 October 2018 (UTC)
- Software modification may indeed be an alternative solution. But why would we even look for it? This proposal does not touch article titles, nor does it touch software.
- Agree that Titles, navigation and search are a bit of a trilemma. Redirects being a secondary issue.
- But it is a valid (but often ignored) principle of software maintenance that you fix the little problem that you understand, because then the big problem that you don't understand often magically goes away. This is a little problem. This trilemma is an enormous topic, and discussing it is unlikely to produce anything other than discussion.
- So let us just fix the little problem that some readers can't see the article they want in a search results list. The problem and solution are both easily understood. The downside is...? Andrewa (talk) 15:35, 9 October 2018 (UTC)
- You're free to create {{R from unnecessary disambiguation}} redirects should you see any that should be created. You don't need to update policies or guidelines to do so, noting there are already several thousand of them (that have been categorized as such). -- Tavix (talk) 15:49, 9 October 2018 (UTC)
- Would you consider removing the bullet point from that post? It seems unnecessary, and likely to just make the stringing (Generally colons and asterisks should not be mixed) harder to follow. Andrewa (talk) 16:24, 9 October 2018 (UTC)
You're free to create {{R from unnecessary disambiguation}} redirects should you see any that should be created. You don't need to update policies or guidelines to do so... True, but misses the point. The point I'm making is that these harmless, common and useful redirects should always exist. And ensuring that they do exist does require a change of policy. Andrewa (talk) 18:47, 9 October 2018 (UTC)
- No, ensuring they exist requires editors to take the time to create them. No policy or guideline is stopping them from being created. -- Tavix (talk) 18:50, 9 October 2018 (UTC)
- I'm afraid I think this still misses the point. Yes, wikignomes can already create them, and will if the policy is changed, but generally won't if it remains as is.
- So the issue here is, are they a good idea? And if they are, then their creation should be encouraged by the appropriate policy, and it will then happen. Not overnight, but in due course. Andrewa (talk) 19:03, 9 October 2018 (UTC)
- For redirects, WP:RCATs—not "policy"—are employed to explain the reason a redirect exists. One of these RCATs, {{R from unnecessary disambiguation}}, encourages the creation of the type of redirect you describe and explains why these are helpful redirects to anyone who views the redirect. Literally the only reason that they are not ubiquitous is due to the effort it takes to create them. If you think there should be more of them, please, be bold and start creating them. This is not a "policy" issue. -- Tavix (talk) 19:17, 9 October 2018 (UTC)
- Good point. If we were to update that template's documentation to explicitly encourage creation of these redirects, that would be quite sufficient.
- I don't see that it does so now, in fact the very phrase redirect from unnecessary disambiguation discourages them.
- The disambiguation is, in terms of current policy, unnecessary. But the point I'm making is that the redirects themselves are useful, and not just for preserving incoming external links. And that's why I called them unambiguous redirects not redirects from unnecessary disambiguation. Andrewa (talk) 00:22, 10 October 2018 (UTC)
- For redirects, WP:RCATs—not "policy"—are employed to explain the reason a redirect exists. One of these RCATs, {{R from unnecessary disambiguation}}, encourages the creation of the type of redirect you describe and explains why these are helpful redirects to anyone who views the redirect. Literally the only reason that they are not ubiquitous is due to the effort it takes to create them. If you think there should be more of them, please, be bold and start creating them. This is not a "policy" issue. -- Tavix (talk) 19:17, 9 October 2018 (UTC)
- No, ensuring they exist requires editors to take the time to create them. No policy or guideline is stopping them from being created. -- Tavix (talk) 18:50, 9 October 2018 (UTC)
- An alternative which I hadn't thought of is to have all articles at unambiguous names, and whenever a P T is agreed to exist, make that base name into a primary redirect. Good suggestion above (but there are other parts of this post with which I don't agree, see #Simplicity). Andrewa (talk) 03:14, 10 October 2018 (UTC)
What does "(disambiguation)" mean anyway
We should rename every disambiguation page to something that is normal, common, easily understood English.
The qualifier (disambiguation) is Wikipediaspeak, not English. A name such as Wave, list of articles would be far more helpful, particularly to newcomers looking at either a Google or Wikipedia search results list.
And there's nothing stopping us from creating redirects from such names, except that without a consensus decision it might be seen as disruptive. Andrewa (talk) 00:38, 9 October 2018 (UTC)
PS I'm not proposing that we deprecate the term disambiguation in Wikipedia documentation etc.. Just in page names in the main namespace. Updating the doco etc to eliminate the term would be arduous, unnecessary, pointless, stupid, and a few far less flattering terms. We all know what it means. Andrewa (talk) 00:48, 9 October 2018 (UTC)
- The question is "helpful" - are there any metrics or evidence for useability? Ignoring disambiguation and dabs not technically being "lists", we're free to choose a different arbitrary identifier. "(disambiguation)" is 16 characters ", list of articles" is 18. There may be shorter ones "(articles)", "(multiple)", "(ambiguous)", "(list)" / variants without brackets. Considering we're currently struggling to delete newly created invalid (disambiguation) redirects, we'd have to keep both in use, so wouldn't any alternative have to be much more than a marginal improvement for that cost? If we decoupled the titles from the MW search, they could be translated in the search into human readable, which comes back to the central question of what is the COMMONNAME of disambiguation, disambiguation would indicate it is. Widefox; talk 11:45, 9 October 2018 (UTC)
- Replying to Widefox 11:45, 9 October 2018 (UTC):
- Lots of good issues here, thank you!
- But if we focus on elegance as we, the contributors, see it, then we tie ourselves in knots (and we're good at that). What is elegant to some is ugly to others. We need to focus on the reader... but even that is controversial, aaargh! Another knot in which we can tie ourselves.
- Agree that The question is "helpful". But do we really need metrics or evidence for useability? Is there any doubt that (disambiguation) is our own local jargon, and would be better avoided?
- Agree that dabs not technically being "lists" can (and I guess will) be made an issue, does it have any downside for the reader? What in terms if WP:IAR does improve Wikipedia mean?
- Agree that "(disambiguation)" is 16 characters ", list of articles" is 18. There may be shorter ones "(articles)", "(multiple)", "(ambiguous)", "(list)" / variants without brackets. Do the extra two characters really matter, compared to avoiding Wikipediaspeak in article titles? But other suggestions welcome. I considered ,index of articles and just (index). Either would do. Some of your suggestions just seem to be changing to another local jargon term.
- But disagree that disambiguation would indicate it is (ie is the COMMONNAME). Disambiguation redirects to the article on Word-sense disambiguation, a topic of Computational linguistics. See also Disambiguation (disambiguation)! There's far more chance that readers will understand the term disambiguation in its Wikipediaspeak sense than as the technical term it is in computational linguistics. But even better, use a term that is common English! And disambiguation is not. Andrewa (talk) 16:35, 9 October 2018 (UTC)
- "list of articles" gives no indication that it's about different meanings. "X, list of articles" sounds like a list of articles about whatever topic the reader mainly associates with X. PrimeHunter (talk) 16:47, 9 October 2018 (UTC)
- But it's not about different meanings, it's just about getting the reader to the right article. If they think it's a list of articles about whatever topic the reader mainly associates with X then that's better than a term they don't understand at all... and if they're looking for another topic that they don't mainly associate with X and can't see the article they want in the search lists, won't they give it a go? Rather than something they don't understand at all? Perhaps it's still not ideal and I'm keen to hear better suggestions. Andrewa (talk) 18:00, 9 October 2018 (UTC)
- But disambiguation is not "a list of articles about whatever topic the reader mainly associates with X". And disambiguation is not "a term they don't understand at all...". Just because you think someone in theory doesn't understand it (despite all the evidence given to you otherwise), doesn't mean we should introduce all these issues where there isn't one in the first place. -- Tavix (talk) 18:08, 9 October 2018 (UTC)
- Yes, the discussion here has been rather waffly and opiniated. Not sure how to get it back on track. Andrewa (talk) 18:18, 9 October 2018 (UTC)
- The name of a disambiguation page is also seen by readers who are not looking for a specific meaning, e.g. in searches. If they click "X, list of articles" because they are looking for articles about the main meaning of X then they have been misled if they get a single link to the main meaning and many links to unrelated meanings. And like others, I question your premise that readers generally don't know what disambiguation means. PrimeHunter (talk) 18:23, 9 October 2018 (UTC)
- Thank you, that addresses the issues here well! But perhaps often would be a better description of my premise than generally. And even that's too strong. Even sometimes would be reason enough to question whether there's a better identifier for these important pages. (The question then of the process to fix it, and whether it's even worth the effort, only becomes relevant if we have consensus that there is. And I admit it is not looking good.) Andrewa (talk) 00:09, 10 October 2018 (UTC)
- The name of a disambiguation page is also seen by readers who are not looking for a specific meaning, e.g. in searches. If they click "X, list of articles" because they are looking for articles about the main meaning of X then they have been misled if they get a single link to the main meaning and many links to unrelated meanings. And like others, I question your premise that readers generally don't know what disambiguation means. PrimeHunter (talk) 18:23, 9 October 2018 (UTC)
- Yes, the discussion here has been rather waffly and opiniated. Not sure how to get it back on track. Andrewa (talk) 18:18, 9 October 2018 (UTC)
- But disambiguation is not "a list of articles about whatever topic the reader mainly associates with X". And disambiguation is not "a term they don't understand at all...". Just because you think someone in theory doesn't understand it (despite all the evidence given to you otherwise), doesn't mean we should introduce all these issues where there isn't one in the first place. -- Tavix (talk) 18:08, 9 October 2018 (UTC)
- But it's not about different meanings, it's just about getting the reader to the right article. If they think it's a list of articles about whatever topic the reader mainly associates with X then that's better than a term they don't understand at all... and if they're looking for another topic that they don't mainly associate with X and can't see the article they want in the search lists, won't they give it a go? Rather than something they don't understand at all? Perhaps it's still not ideal and I'm keen to hear better suggestions. Andrewa (talk) 18:00, 9 October 2018 (UTC)
- "list of articles" gives no indication that it's about different meanings. "X, list of articles" sounds like a list of articles about whatever topic the reader mainly associates with X. PrimeHunter (talk) 16:47, 9 October 2018 (UTC)
- Here you go. It's a proper English word with plenty of usage outside of Wikipedia. -- Tavix (talk) 15:52, 9 October 2018 (UTC)
- In esoteric circles, yes. The "disambiguation" article should educate you on the word [14]... There is actually no such article, see below. And is the bullet point really helpful?
See above.Andrewa (talk) 16:35, 9 October 2018 (UTC)
- In esoteric circles, yes. The "disambiguation" article should educate you on the word [14]... There is actually no such article, see below. And is the bullet point really helpful?
- It looks like you are creating a problem where there isn't one because you refuse to acknowledge that "disambiguation" is in fact a common English word. -- Tavix (talk) 17:11, 9 October 2018 (UTC)
- I respect your opinion that it isn't a problem, and that disambiguation is a common English word. But the evidence doesn't seem to me to support these opinions. Yes, it's in wiktionary. Exactly what does that prove? Andrewa (talk) 17:39, 9 October 2018 (UTC)
- Well... that it's an English word, not "Wikipediaspeak" as you claim. It's not even like it's a cryptic word! dis- "to remove or reverse" added to ambiguous "vague or unclear", with the -tion suffix to make it a noun, forms "a removal of vagueness or unclearness", which is precisely what disambiguation does. On the other hand, what is ambiguous is the proposal to replace "disambiguation" with "list of articles", which would signify any article related to the term, not simply articles with the same name—which is distinctly different -- Tavix (talk) 17:58, 9 October 2018 (UTC)
- Accepted that it's an English word, and apologies of my term Wikipediaspeak was understood to mean it's not English. It is, and even if it weren't previously our use of it would attest to it being English now, even if it were not in Wiktionary yet, and it is. And it was previously used in other very specialised areas. But not by the general readership, that's my point.
- The proposal is to replace (disambiguation), not specifically to replace it by ,list of articles. There may well be better alternatives. Andrewa (talk) 18:09, 9 October 2018 (UTC)
- Which simply isn't the case! Do you have evidence of anybody being confused by this word to the point where they couldn't get to what they were trying to find? -- Tavix (talk) 18:12, 9 October 2018 (UTC)
- What isn't the case? Yes, it would be good to gather evidence... either way... any suggestions as to how? Andrewa (talk) 18:20, 9 October 2018 (UTC)
- Well, there is no evidence that "disambiguation" is a confusing word for anyone, let alone enough people to the point where the cost-benefit ratio would be anywhere near appealing. If you posted links to people being confused by this word, that would be evidence. Until then, a proposal to alter hundreds-of-thousands of pages due to a theoretical problem is a non-starter. -- Tavix (talk) 18:33, 9 October 2018 (UTC)
- And if we had some links that did show people being confused by this word, then we'd have a case? I'm not quite sure how to gather them, but I'll have a go... provided that they would actually make a difference. Andrewa (talk) 00:13, 10 October 2018 (UTC)
- That should now read see above following this edit (which was mainly a good idea, but I thought if I did as recommended in that edit summary it would also cause other problems of stringing). I do not know why bullet points have been introduced here, they seem to serve no purpose other than to make the stringing hard to follow. Or perhaps that is it? Naw, AGF. But the wikilink I put in to wp:stringing doesn't seem to have helped. Andrewa (talk) 17:51, 9 October 2018 (UTC)
- Yes, disambiguation is a redirect, fair point, but still my question being is there a more common name, preferably more concise? "(word-sense disambiguation)" and "(WSD)" are worse. "(disambiguation)" may be ugly-useful. One could even reduce it to a prompt for the user to reinforce that selection is needed "(?)" (note we have (?)), "(*)", or just "()". I'd avoid "index" and maybe "list" completely due to already having those. Changing the identifier may bring some marginal benefit for users, especially if they type it (do they?), but it's such a disruptive change and MW would need to at least cover the redirect of all "(disambiguation)" urls to the new identifier, which could alternatively be done already in the other direction i.e. "(?)" -> "(disambiguation)". My gut reaction is that it's good enough, we're probably stuck with it like QWERTY keyboards i.e. high cost/highly established with marginal gain, and that cost means we'd want to identify the gain, yes. (SIAs on the other hand, but I digress). Widefox; talk 18:20, 9 October 2018 (UTC)
- Is it really such a disruptive change? That seems to be the unsaid assumption above. It would IMO be quite painless... except for messy discussions such as the above! Why do I bother... (;-> Andrewa (talk) 18:25, 9 October 2018 (UTC)
- The pain may be worth the gain, but until the gain is identified, it's an assumption that different is better. I see growing desire to have something better for searching/navigating, and our navigation assumption about PT -> dab -> article may not be the major navigation path or that useful nowadays? Stats anyone? What's coming with MW/WMF? Widefox; talk 18:41, 9 October 2018 (UTC)
- I think I provided a case that deserves an answer and hasn't had one... although I now see that I did say Wipediaspeak not English and stand corrected on that. What I was trying to say was, to the contributors that term is fine, but to the general reader it's not obvious what it means. Is that clearer now?
- And I still think that the pain is minimal. And it's not change for change's sake. There is an advantage in using page titles that the average reader understands easily, and particularly so with pages such as DABs which, by their very nature, are frequented by those who don't know what the page they really want is called. (Do I really need to look for evidence of that?) Andrewa (talk) 18:57, 9 October 2018 (UTC)
- Info cf. Encyclopedia Britannica 1. doesn't have dabs 2. has only one passing mention of "disambiguation" 3. has longer titles 4. relies more on search 5. seems worse Widefox; talk 19:33, 9 October 2018 (UTC)
- Interesting. I don't think anyone was suggesting that we do away with DABs. I'm actually suggesting that we should make them easier to find. Andrewa (talk) 00:02, 10 October 2018 (UTC)
- We should aim for the exact opposite - getting dabs out the way as much as possible, as they only a stepping stone to what the reader actually wants. If we had more information we wouldn't need them at all. Widefox; talk 13:03, 10 October 2018 (UTC)
- Interesting. I don't think anyone was suggesting that we do away with DABs. I'm actually suggesting that we should make them easier to find. Andrewa (talk) 00:02, 10 October 2018 (UTC)
- Info cf. Encyclopedia Britannica 1. doesn't have dabs 2. has only one passing mention of "disambiguation" 3. has longer titles 4. relies more on search 5. seems worse Widefox; talk 19:33, 9 October 2018 (UTC)
- The pain may be worth the gain, but until the gain is identified, it's an assumption that different is better. I see growing desire to have something better for searching/navigating, and our navigation assumption about PT -> dab -> article may not be the major navigation path or that useful nowadays? Stats anyone? What's coming with MW/WMF? Widefox; talk 18:41, 9 October 2018 (UTC)
- Is it really such a disruptive change? That seems to be the unsaid assumption above. It would IMO be quite painless... except for messy discussions such as the above! Why do I bother... (;-> Andrewa (talk) 18:25, 9 October 2018 (UTC)
- Yes, disambiguation is a redirect, fair point, but still my question being is there a more common name, preferably more concise? "(word-sense disambiguation)" and "(WSD)" are worse. "(disambiguation)" may be ugly-useful. One could even reduce it to a prompt for the user to reinforce that selection is needed "(?)" (note we have (?)), "(*)", or just "()". I'd avoid "index" and maybe "list" completely due to already having those. Changing the identifier may bring some marginal benefit for users, especially if they type it (do they?), but it's such a disruptive change and MW would need to at least cover the redirect of all "(disambiguation)" urls to the new identifier, which could alternatively be done already in the other direction i.e. "(?)" -> "(disambiguation)". My gut reaction is that it's good enough, we're probably stuck with it like QWERTY keyboards i.e. high cost/highly established with marginal gain, and that cost means we'd want to identify the gain, yes. (SIAs on the other hand, but I digress). Widefox; talk 18:20, 9 October 2018 (UTC)
Agree that we should get dabs out the way as much as possible. But disagree that this is the exact opposite. In fact my other relatively simple (famous last words) proposal aims to do exactly that... make it more probable that a reader will recognise a link to the exact article they want, and so avoid any need for them to go to the DAB.
But some readers will want to go to the DAB, for whatever reason... otherwise as you say, why have them at all? And for these readers, it's an advantage to have the DAB at a more recognisable name. Andrewa (talk) 19:33, 10 October 2018 (UTC)
- I'm not really following
make them easier to find
andget dabs out the way as much as possible
, but I presume this is a mix of search, navigation, and link recognition. In search, dabs could be used by MW to prompt users for the selection of article, thus avoiding the dab navigation step completely. Combined with auto-generation (in theory), it could be completely handled by MW, as well as adding the hatnote (for some but not all cases). Widefox; talk 18:31, 11 October 2018 (UTC)- Well, get dabs out the way as much as possible was your suggestion.. and I took it to mean, it's far better for a reader to be able to go straight to the article they want than to need to look at a DAB first, and I agree. Did I misunderstand you?
- But as I said, some people will (for the moment at least) still need the DABs (and otherwise why have them) and so (as with every other page) the more recognisable the page name is, the better... because a recognisable page name makes the page easier to find. Am I missing something?
- MW I assume is MediaWiki. Yes, it could probably be improved too. But in this case that seems to me to be a messy way of going about things. Andrewa (talk) 05:24, 12 October 2018 (UTC)
- I assumed that get dabs out the way as much as possible meant: Get dab pages off basenames. A DAB page on a basename is ASTONISHING to anyone using the Go box autocomplete if they think their topic is the primary topic. Even if they are mistaken. This is where some may say they need to be astonished out of their ignorance. I think the DAB page should be suffixed with "(disambiguation)", so that the autocomplete functionality tells the person that they are about to choose a disambiguation page, even if the basename redirects to the DAB page. --SmokeyJoe (talk) 05:41, 12 October 2018 (UTC)
- Yes Andrewa, that's what I meant - just status quo. I agree with your sentiment, I still don't understand the proposed change. Dabs are like Schrödinger's pages, we're dealing with uncertainty, but that collapses. Yes being more explicit about a dab not being an article may be helpful, but the namespace is a totally different question to the UI.
- SmokeyJoe - I did not mean "Get dab pages off basenames". I do mean status quo as in WP:MALPLACED is good. This is back to the trilemma... having all dabs as MALPLACED just to improve the Go box? Does the Go box behave differently at present with MALPLACED dabs? As I said, isn't it better to improve MW than workaround a less good interface using redirects (if that works)? I think it doesn't help to conflate/tradeoff namespace/navigation for search/Go box. I wouldn't call dabs ASTONISHING out of error (a bug in the data), but more out of design (a feature, and one that EB suffers from by not having, forcing users to search) and only when we probably don't have enough information (else they get out the way per PT). The Go box does have enough info to treat dabs differently. It literally doesn't matter what the reader assumes, only what info they give us to get them to the right place, and what we give them in an interactive way at the Go box. A dab feature is not an ASTONISHING bug. Yes we're free to have all dabs as MALPLACED, but what does that achieve? Widefox; talk 12:34, 12 October 2018 (UTC)
- I confess I'm at a loss as to what is being discussed here and why. There's no proposal to eliminate DABs, so can we stop talking as if there were? There's no case for modifying MediaWiki, that's just asking for trouble (as is using the abbreviation MW which means something else). There's a misunderstanding (not by me) of what was meant by get dabs out the way as much as possible, but that seems to be cleared up, and we seem to be unanimous that it's a good idea. Andrewa (talk) 21:44, 12 October 2018 (UTC)
- I'm not sure what this is if there's only one solution, MW is to be ignored when talking about search, and the problem isn't even well defined. Widefox; talk 12:12, 15 October 2018 (UTC)
- I confess I'm at a loss as to what is being discussed here and why. There's no proposal to eliminate DABs, so can we stop talking as if there were? There's no case for modifying MediaWiki, that's just asking for trouble (as is using the abbreviation MW which means something else). There's a misunderstanding (not by me) of what was meant by get dabs out the way as much as possible, but that seems to be cleared up, and we seem to be unanimous that it's a good idea. Andrewa (talk) 21:44, 12 October 2018 (UTC)
- I assumed that get dabs out the way as much as possible meant: Get dab pages off basenames. A DAB page on a basename is ASTONISHING to anyone using the Go box autocomplete if they think their topic is the primary topic. Even if they are mistaken. This is where some may say they need to be astonished out of their ignorance. I think the DAB page should be suffixed with "(disambiguation)", so that the autocomplete functionality tells the person that they are about to choose a disambiguation page, even if the basename redirects to the DAB page. --SmokeyJoe (talk) 05:41, 12 October 2018 (UTC)
Wikipedia:P T
User:Andrewa Sorry I've only just read Wikipedia:P T. Instead of me adding hatnotes (c.f. Wikipedia:PT), can you explain what it is, as I don't get it. Should it be marked an essay/possibly moved? Widefox; talk 13:40, 12 October 2018 (UTC)
- I suggest discuss at Wikipedia talk:P T, probably in the existing first section Wikipedia talk:P T#Why this page. It's not IMO an essay, as no opinions are expressed, and I'm not quite sure what moving it would achieve, or where you'd move it, or what would take its place... I guess making it a shortcut to either the guideline or the policy would be best, but that defeats part of the purpose, which was to help people to find both the policy and guideline without cluttering discussions.
- But a redirect to either would still fulfill the most important function of WP:P T, which was to avoid the messiness of WP:PT, which has at various times pointed to protection templates, WikiProject Portugal, Protected titles, and of course Primary topic most recently. It has been protected and unprotected, and see this edit for something quite worrying! Andrewa (talk) 14:56, 12 October 2018 (UTC)
- Discussion is probably best here Wikipedia:Miscellany_for_deletion/Wikipedia:P_T. Widefox; talk 15:50, 12 October 2018 (UTC)
- Agree, now that you have proposed it for deletion (which you have every right to do) that's the primary discussion until the MfD is closed. Andrewa (talk) 19:40, 12 October 2018 (UTC)
- Discussion is probably best here Wikipedia:Miscellany_for_deletion/Wikipedia:P_T. Widefox; talk 15:50, 12 October 2018 (UTC)
Scenarios and evidence
As I said before, I'm working on an RfC to deprecate primary topic (which I like to call P T for clarity but see Wikipedia:Miscellany for deletion/Wikipedia:P T). It seems to me that its purpose and benefits are not as obvious as I once assumed, and as many still do assume.
Let me try two scenarios, and ask everyone three questions:
- What would you do in this case?
- If what you'd do is different to what I'd do, why?
- If what you'd do is the same as what I'd do, do you think that others might do something different, and if so what and why?
This is again to get the mental juices going, and to provide me with input as to whether any RfC might have a chance, and if so exactly what that RfC would be. Andrewa (talk) 10:29, 13 October 2018 (UTC)
Scenario one
You want an article on wave as in physics, and you find links to two pages (or more, but two that look particularly relevant), one at wave (physics) and one at wave. I would click on the link to wave (physics). What would you do? Andrewa (talk) 10:31, 13 October 2018 (UTC)
- I would get annoyed. I think I should expect that wave is a very broad concept article probably covering some physics. I know there are different sorts of waves in physics. Wave (physics) is very likely overspecific, “physics” makes me think I'll get a page of diagrams and equations, as I have seen in some articles before. I’d probably open both, if working with the luxury of a multi-tab device with fast connection. If working with difficulty, I’d be wishing for a Wave (disambiguation) option, expecting a clear list of options on a short page with no heavy images. —SmokeyJoe (talk) 01:54, 15 October 2018 (UTC)
Scenario two
You want an article on wave as in physics, and you find links to two pages (or more, but two look possibly relevant), one at wave (disambiguation) and one at wave. There's no link to an article at a title such as wave (physics) which you recognise as the one you want, but you don't know what disambiguation means. I would click on wave. What would you do? Andrewa (talk) 10:34, 13 October 2018 (UTC)
- the first time I saw “disambiguation” I thought it odd, but quickly deconstructed the word to work it out. Whenever unsure what title I want, I go to the disambiguation page. I find disambiguation pages to be light and very helpful, often including things I had quite forgotten, if I ever knew, and always impressively articulated. I think the premise of “you don't know what disambiguation means” is true only very briefly. —SmokeyJoe (talk) 01:59, 15 October 2018 (UTC)
After two days
At the risk of arguing from silence, if nobody responds I'm tempted to conclude that everyone sees my answers as reasonable, and as reflecting what most readers will do. Or, is there a better way of measuring this? Andrewa (talk) 01:16, 15 October 2018 (UTC)
- I think you make a lot of sense. I do not follow everything perfectly, so I am not jumping in to sign of as in full agreement. WP:Silence exists, as a page, as a concept, for a good and necessary reason. Go to the next step. We are watching. Talk page silence means little. Edit the guideline. If anyone disagrees the onus is on them to say so, but only after you do something tangible. Talk page posts are not quite tangible. —SmokeyJoe (talk) 02:01, 15 October 2018 (UTC)
Replying to SmokeyJoe 02:01, 15 October 2018 (UTC) and not seeing any reason for the bullet...
Your replies above are very welcome. They raise a lot of issues that hadn't occurred to me.
You mention tabs, and mobile devices. The oldest version of WP:disambiguation currently in the database is this one from 2 February 2002, and it's not very informative. It doesn't speak of disambiguation at all, and might not have been called that either... we need to look further into the history to find that out. And its edit summary indicates that it's not the first version, but that earlier versions have been deleted in a database reorg at some stage.
But some of our younger editors were not even born in 2002. The Danger Hiptop, the first smartphone to have any impact outside of Japan, was released in 2002, so mobile computing as we now know it did not yet exist; The first release of Android was still six years in the future, in September 2008. Internet Explorer didn't adopt tabs until Internet Explorer 7 in 2006, the last major browser to do so admittedly.
The point is just, in many important ways, 2002 as seen by Wikipedians was a very different world to 2018. And it would be surprising if some of our principles weren't due for a rethink in view of this. Disambiguation in general, and Primary Topic in particular, look like good candidates for this. Andrewa (talk) 07:18, 15 October 2018 (UTC)
The point being
OK, I'm going to assume in the absence of any disagreement...
- Most readers, facing a choice between Wave and Wave (physics) and wanting an article on wave as in physics, will click first on Wave (physics). (They may then open another tab for Wave if their device facilitates this.)
- Most readers, facing a choice between Wave and Wave (disambiguation) and not having an article title that they identify as being the one they want, will click first on Wave.
If this is true, then it seems to me that clicking on the link to Wave (physics) should take a reader straight to that article, but that clicking on the link to Wave should take them to the DAB, and that this is the best setup regardless of any consideration of primary topic.
And that is the point of these scenarios. Comments? Andrewa (talk) 07:46, 15 October 2018 (UTC)
- If all page views came via WP's built-in search box then I might agree, but we need to consider other routes in which the reader never sees the option "Wave (physics)":
- via typing "Wave" and hitting Enter without reading through a list of possibilities
- via an external search engine, having searched for "Wave" without adding "(physics)"
- via a wikilink, possibly added by an editor who assumed that Wave means what we all know it means so that's what the article will be about
- via an external link to WP's "Wave" page, either one which already exists or one added later without checking where it leads (because it's obvious)
- For these reasons, I think we still need primary topics, though I agree that they should be used less often. Certes (talk) 12:20, 15 October 2018 (UTC)
- Very good point, and one I had already investigated at User:Andrewa/The third draft regarding avoiding primary topic#How readers find articles (and it might even have been you who suggested this to me). And it changed my approach somewhat. My conclusion was, they all benefited from moving the policy away from use of ambiguous names, if this were to be done carefully.
- And that means far more carefully than I first assumed, I admit that. There's no case for moving most articles currently at ambiguous names, and probably never will be for the vast majority of them. But avoiding that is surprisingly simple.
- So agree that it's not just WP's built-in search box that we need to consider. See #A current case in point for just one other route that is similarly affected.
- And agree that we still need primary topics, and my latest proposal is not to eliminate them. See User:Andrewa/Primary Topic RfC#Proposal and User:Andrewa/Primary Topic RfC#Scope.
- And much of the same benefit might be achieved simply by raising the bar on P T. Two problems. One, exactly how far do we raise it? Two, how do we get consensus to do so?
- (But in some ways I suppose my current proposal could be seen as raising the bar I guess. Which might answer question one! Is it too much to hope it might also answer question two?)
- My conclusion has been that replacing an illogical rule by a significantly less illogical one is still not the way to go! But I could be wrong. Andrewa (talk) 23:12, 15 October 2018 (UTC)
A current case in point
See User talk:Andrewa/P T examples and scenarios#Capitol Limited and Talk:Capitol Limited (Amtrak train)#Requested move 8 October 2018, which may have closed by the time you read this. The move seems inevitable, as it should be under current policy.
- Discussion is now moved to Talk:Capitol Limited#Requested move 8 October 2018 and closed as move, as expected. Don't know why the redirect creation was suppressed. Andrewa (talk) 22:16, 15 October 2018 (UTC)
But see also http://alderspace.pbworks.com/w/page/129510039/Capitol%20Limited which is the Google results list taken before the move... it looks perfect, does it not?
How exactly does this move benefit the reader? Isn't it at best a case of our following a rather pointless rule?
Which was of course the original question that sparked this whole section. Andrewa (talk) 09:35, 15 October 2018 (UTC)
This has now (correctly according to current policy) closed as move as expected. And also as expected, a Google of "Capitol Limited" now returns only the Amtrak train. All direct mention of our article on Capitol Limited (B&O train) has vanished from Google.
(And there's not even a hatnote from the Amtrak article to Capitol Limited (B&O train), just one to Capitol Corridor, operated by Amtrak in California, which is I guess very loosely related to both but doesn't mention the B&O train, run by the Baltimore and Ohio Railroad between New York City and Chicago via Washington D.C. and Pittsburgh, at all. But that's easily fixed.)
The challenge: This has obviously made Google less useful (understatement perhaps) for those looking for the B&O train (discuss if you disagree). But exactly how has it improved Wikipedia? Apart from compliance to a rule which, in this case at least, was quite possibly counterproductive?
It has unimproved Wikipedia. And there is no benefit to counter this unimprovement. Is there? What am I missing? Andrewa (talk) 23:02, 15 October 2018 (UTC)
- The move of Capitol Limited (Amtrak train) to the basename has further obscured the old Capitol Limited (B&O train). The new article doesn't discuss the history of the name, or inform readers that there was a previous train of the same name. For someone looking for the older train, confusingly there is early mention of "the other" in the lede sentence, but referring to something else entirely. So, I added a hatnote. So, now, instead of a PRECISE title, there is a short ambiguous title with a lot of title whitespace, and more hatnoting, hatnoting that is distracting to the majority who don't care, and not even shown for many depending on their device. So Google has noticed? The old page has less prominence. Fewer will find it. Google will notice. Pageviews will decline. This is a self-reinforcing bias. I think WP:PRECISE needs a thorough thrashing. Cut the lede word "usually". Cut the off-point blather after the first sentence. Cut the trailing backward off-topic clause. Cut the unhelpdul and odd"topical". Yields: PRECISE: "Titles should be precise enough to unambiguously define the scope of the article". --SmokeyJoe (talk) 23:52, 15 October 2018 (UTC)
- (Still not sure how best to respond to your mixed indents, but otherwise) Very well put! Lots of downside. Still interested to hear of any upside. But I'm expecting a deafening silence. And it raises some other good issues.
- Support "Titles should be precise enough to unambiguously define the scope of the article". But wouldn't this bring it into direct conflict with WP:P T, and perfectly into agreement with User:Andrewa/Primary Topic RfC? Because, that's exactly what I'm proposing there, with a plan of action to achieve this with a minimum of effort and inconvenience. Or that was my intent, anyway.
- Move was a good close (in theory... it was bungled in practice, I'm discussing that with the page closer). There's no grounds for MR. None of this was raised in the discussion, I considered whether it was pointy of me not to chime in but even if I had there was consensus to move IMO. The move was based on policy, I'd have needed to invoke IAR to oppose and IMO I can't validly do that until I've demonstrated some consensus for my concerns with WP:P T. And I don't think that has happened yet, but I'm hopeful it's not too far away. Andrewa (talk) 00:23, 16 October 2018 (UTC)
- I just searched Google and got both articles, as the third and fourth items. My understanding is article titles have little to no effect on Google results. Station1 (talk) 06:40, 16 October 2018 (UTC)
A second challenge: In terms of silence I think there's rough consensus here that this move was a very bad idea indeed. But I also think the move was completely justified in terms of policy.
Ergo, policy is at least a little but busted.
But is this an isolated case? I don't think so, obviously. SO, the challenge... Look around and comment at User talk:Andrewa/P T examples and scenarios, either commenting on an existing example or raising a new one in its own section.
I'm interested in two sorts of evidence of course:
- Examples (such as this one) where the policy I'm proposing would work a lot better than the existing policy.
- Examples where the policy I'm proposing would work a lot worse than the existing policy.
TIA Andrewa (talk) 01:31, 16 October 2018 (UTC)
- I agree "policy is at least a little but busted". I have long argued that there is a lack of logic in WP:PT, as written, and as practiced. It has more history and faithful adherents than underlying logic. I don't agree with repealing it entirely. I agree with User:Certes 12:20, 15 October 2018 "... we still need primary topics, though I agree that they should be used less often". --SmokeyJoe (talk) 01:53, 16 October 2018 (UTC)
Items only in other-language Wikipedias or Wikidata
I keep running across items that link to either a non-English Wikipedia (such as Alexander Nikolayevich Golitsyn on Alexander Golitsyn) or Wikidata. What do we do with these? Do the they require a blue link in English Wikipedia, like other red links? Leschnei (talk) 18:39, 8 October 2018 (UTC)
- We have links to other projects in articles so I don't generally see why not on DAB pages. However maybe we should restrict to times when the missing article is notable, but I don't see why WP:DABMENTION can't be used also. Crouch, Swale (talk) 20:31, 8 October 2018 (UTC)
- No per WP:DABSISTER
Disambiguation descriptions should not be created for subjects whose only articles are on pages of sister projects...
and WP:WRITEITFIRST. Comment out or create a stub is what I'd do. Widefox; talk 22:14, 8 October 2018 (UTC) - Curiously, MOS:WTLINK seems to allow sister project links for dictionary defs, which I don't think I've seen. I suggest we remove it from MOSDAB as wiktionary links seems to cover it. Widefox; talk 22:18, 8 October 2018 (UTC)
- No per WP:DABSISTER
- Thanks Crouch, Swale and Widefox, WP:DABSISTER is what I was looking for but, for some reason, couldn't find. Leschnei (talk) 22:55, 8 October 2018 (UTC)
- You mean, you haven't memorised WP:MOS??? I'm shocked, shocked!
- Anyway, WP:MOS is inconsistent. MOS:DABOTHERLANG says, "For foreign-language terms, be sure an article exists or could be written for the word or phrase in question" (emphasis added). That implies that a DAB page entry is OK even if there isn't a WP:DABMENTION. (That guideline wording is feeble. (1) There are any number of articles on non-English Wikipedias which wouldn't survive WP:AFD if translated. (2) There are very many more non-English topics which wouldn't survive AFD in the home language. Articles could be written about any of them; whether they'd survive more than 7 days is another matter.)
- IMO, WP:DABSISTER is incomplete, and WP:DABOTHERLANG is wrong. Both should be subject to WP:DABMENTION. (1) If a DAB page entry passes WP:DABMENTION, then an {{ill}} link adds information and helps other readers and editors, which is what we're here for. Among other things, it helps editors unify redlinks. (I cannot be alone in having turned redlinks into redirects to existing articles.) Alexander Nikolayevich Golitsyn is a good example. (2) If it doesn't, it points to no useful information in English Wikipedia, and should be commented out (better than deletion: it proposes a title). (I've probably broken that rule myself during multilingual searches, but plead WP:IAR.)
- Wiktionary links on DAB pages should always go through {{wikt}}: any sort of piped soft link on a DAB page is just what we don't need; they're bad enough in articles. Narky Blert (talk) 19:40, 11 October 2018 (UTC)
- Yes
could be written
isn't helpful and should be removed and replaced with WP:WRITEITFIRST, which is helpful. The full quote of MOS:DABOTHERLANG helps, as it qualifies...Usually this means that the term has been at least partially adopted into English or is used by specialists.
which implies it has to have relevance in English for en.WP. Practically, we should try to eliminate these grey areas so that editorial/notability judgement is minimised/eliminated from dabs and kept appropriately at the articles subject to WP:V etc by yes requiring a DABMENTION. - Regarding Alexander Golitsyn, I'd consider the Alexander Nikolayevich Golitsyn entry failing WP:DABRED, but easily solved by linking to House of Golitsyn which serves readers better than sending them to a sister project. Fundamentally, we must prioritise en.WP content over sisters to avoid WP:CFORK. Until a stub exists we just don't need the creep of judgement into dabs, together with the fundamental of WP:D
reasonably likely
. Widefox; talk 10:25, 12 October 2018 (UTC)- Thanks for the thoughtful replies. It usually comes down to some version of 'strict adherence to the "rules" (such as they are) taking a back seat to what is best for the reader'. I should write that on the back of my hand whenever I sit down for a bit of editing. Leschnei (talk) 12:58, 12 October 2018 (UTC)
- Generally MOSDAB (and WP:D) seem very handy for strict adherence which is aligned with readers. Widefox; talk 13:34, 12 October 2018 (UTC)
- I prefer the "{{ill}}; see [bluelink]" format to a redirect. Readers get the same information; and the article, being a redlink, is more likely to get written,
- Deliberately writing a stub article when there's more information in another language feels like a cop-out. I've done it a couple of times; once because there were two bad links to a DAB page and the three corresponding articles in other Wikipedias were in languages which Google Translate doesn't understand. I did what I could by approximating to Russian.
- Nevertheless, there is one Wikipedia which consists almost entirely of stub articles, and I applaud its editors. There are only two Wikipedias with over 5M articles, and Cebuano is one of them.
- The list of Wikipedias statistics makes for some startling reading if you look at it closely. I know of one language with 830k speakers which has no Wikipedia at all (Mizo).
- As for Wikidata, I consider it pestilential. Quality control is minimal, to put it mildly. DABlink problems in Wikipedia caused by blindly importing information from Wikidata are among the most difficult to solve. I know of only one editor who is able to tackle them, and it isn't me. In addition, I've had Interwiki links reverted because a Wikidata editor was unable to transliterate between the Roman and Cyrillic alphabets; and also on the ground that a WP:DAB in one language was not the same as a WP:SIA in another; even though the information overlap was huge, and despite the fact that every Wikipedia is entitled to set its own DAB and SIA rules. <sounding off /> Narky Blert (talk) 06:22, 13 October 2018 (UTC)
- I believe that DABSISTER should be updated to support using {{ill}} for topics that would meet notability requirements for en-wiki. The template was created in 2013, while DABSISTER dates from 2004, when the risk would have been an interwiki bluelink to a foreign language and no redlink. ~Hydronium~Hydroxide~(Talk)~ 09:20, 15 October 2018 (UTC)
- No - the crux is that notability question. For clearly notable a stub is easy, anything else editors should be able to WP:V in En.WP in English as that's easiest. En.WP is in a privileged position with sources, editors, readers hence articles. Those fundamentals aren't going anywhere to need DABSISTER changing, WP:WRITEITFIRST is the easiest for en.WP. This may be most clearcut with WP:BLPs and medical topics. Widefox; talk 12:31, 15 October 2018 (UTC)
- I believe that DABSISTER should be updated to support using {{ill}} for topics that would meet notability requirements for en-wiki. The template was created in 2013, while DABSISTER dates from 2004, when the risk would have been an interwiki bluelink to a foreign language and no redlink. ~Hydronium~Hydroxide~(Talk)~ 09:20, 15 October 2018 (UTC)
- Generally MOSDAB (and WP:D) seem very handy for strict adherence which is aligned with readers. Widefox; talk 13:34, 12 October 2018 (UTC)
- Thanks for the thoughtful replies. It usually comes down to some version of 'strict adherence to the "rules" (such as they are) taking a back seat to what is best for the reader'. I should write that on the back of my hand whenever I sit down for a bit of editing. Leschnei (talk) 12:58, 12 October 2018 (UTC)
- Yes