Wikipedia talk:Manual of Style/Disambiguation pages

From Wikipedia, the free encyclopedia
Jump to: navigation, search
WikiProject Disambiguation
WikiProject icon This page is within the scope of WikiProject Disambiguation, an attempt to structure and organize all disambiguation pages on Wikipedia. If you wish to help, you can edit the page attached to this talk page, or visit the project page, where you can join the project or contribute to the discussion.
WikiProject Manual of Style    (Inactive)
WikiProject icon This page was within the scope of WikiProject Manual of Style, a project which is currently considered to be inactive.

Rationale for introductory line[edit]

What is the rationale in favor of the introductory line? I’ve yet to find one disambiguation page where it is not a hindrance. It provides no information, does not help reading, and becomes cumbersome when variants are included. Palpalpalpal (talk) 23:31, 12 March 2017 (UTC)

It tells the reader succinctly and in plain language what sort of page they have landed on. Essentially, it says, "Not sure yet what you are looking for, but here are some options."--Trystan (talk) 23:43, 12 March 2017 (UTC)
The page’s title and layout already make that pretty clear. The user is presented with various choices, and understands at once that he’s not at the end of his research. Sounds like this rule was made by people not well versed in design. Here, more is less; users are not dumb. Are there any other encyclopedias, or dictionaries, that chose the same approach? Palpalpalpal (talk) 10:54, 13 March 2017 (UTC)
Doesn’t anybody else have anything to say about this? This policy does affect a lot of articles. Palpalpalpal (talk) 11:55, 23 March 2017 (UTC)
The introductory line explains what we've chosen to be the Primary Topic, which won't be what the reader landing on the dab page is looking for. We then offer them the range of options they may be looking for. Seems all very reasonable to me. 13:05, 23 March 2017 (UTC)
I believe it useful: It does provide information (what the primary topic is - they may not have navigated away from the primary topic), it distinguishes the page from other types of pages in a consistent way, and once you have seen a few of them, you will quickly learn to skip over it if you are not interested and head to the section you ARE interested in. --John (User:Jwy/talk) 15:12, 23 March 2017 (UTC)

Clarification of policy on partial title matches[edit]

There are two clauses about PTM best practices that should probably be clarified. As it stands, they seem somewhat contradictory.

The first was added to MOS:DABSEEALSO on 19 August 2011 by JHunterJ:

Articles with the item as part of the name, but that are unlikely to be called by the ambiguous title: e.g., Spanish moss as part of a Moss dab page.

The second was added to WP:PTM on 26 October 2016 by JFG; Swpb promptly revised it to read as follows:

Instead of listing partial title matches, consider adding the {{look from}} or {{intitle}} templates in the "See also" section, which link to all articles starting with or containing a particular term, respectively.

The first clause suggests that it's okay to put unambiguous PTMs in the see also section of a dab page, but offers no hints as to when this is and is not a good idea. Meanwhile, the second clause seems to discourage editors from following the first clause.

I haven't found a discussion of either guideline in the respective talk page archives (although JHunterJ mentions his edit here). Have either of these, or their relationship, been discussed?

In order to reconcile these two guidelines, we should be clear about the rationale behind behind them, and articulate a clearer, more cohesive policy.

It seems to me that Spanish moss was chosen as an example because this is a special case for MOS:DABSEEALSO, but that is not explicit in the policy. I interpret it as a special case for two reasons:

  1. There are plenty of unambiguous moss PTMs that are not included in the see also list. (Moss Side and Moss Man are two examples.)
  2. All of the unambiguous moss PTMs in Moss (disambiguation) have something in common: They are not mosses, and as such they are not mentioned in Moss (the primary topic). As plants that have the word moss in their name, yet are not mosses, they are a kind of pseudo-class. The see also section of Moss (disambiguation) lists the members of this pseudo-class.

I think a clearer policy would take these questions into account:

  1. What are we trying to achieve when we include an unambiguous PTM in the see also section?
    • Or, How does an editor determine whether an unambiguous PTM is appropriate for the see also section of a dab page?
    • Or, What kinds of unambiguous PTMs should not be included in the see also section?
  2. Are there cases where listing PTMs in see also is preferable to using {{look from}} or {{intitle}}? (Such as when there are, say, fewer than five PTMs?)

Ringbang (talk) 18:57, 2 May 2017 (UTC)

IMO, with the availability of {{look from}} and {{intitle}}, individual unambiguous PTMs should not be listed unless:
1. They can be individually justified as reasonable search targets, and
2. They are not merely specific instances of a listed entry.
The user looking for fern moss will find it under moss, so it should not be listed separately; the user looking for Spanish moss will not, as so it should. There might be rare reasons to list PTMs that don't meet (1) and (2), but I don't think PTMs should ever be listed merely because there are less than five of them. There's another whole discussion to be had on when to list {{look from}} and {{intitle}}. —swpbT 20:16, 2 May 2017 (UTC)
I agree that each one must be justified as a reasonable search target. Maybe we can cite and explain some examples to make it easier for editors to make that justification objectively.
With enough samples of useful PTMs (like your Spanish moss observation), we might be able to generalise them into use cases and guidelines. The Spanish moss case could translate as something like: If a user might reasonably expect to find the subject at the primary topic, but doesn't, and the name of the subject is a PTM, it's a candidate for 'See also'.Ringbang (talk) 22:44, 2 May 2017 (UTC)
I think it could be difficult and WP:CREEPy to generate a bunch of separate use cases; I was trying to generalise to a simple set of rules that will take editors to the point where common sense can take over. I think, e.g., that my two rules cover the moss case well, and I think most others. Why don't we implement those two rules to start, and then add specific use cases as needs arise? —swpbT 12:51, 3 May 2017 (UTC)
Agreed. I didn't mean to imply that the guidelines are for the policy page, but to try to address what I mentioned earlier in my post about understanding the rationale (in order to generalise something from the cases). But yes, I do think we could immediately clarify the policy based on your suggestions.
For line 6 of MOS:DABSEEALSO, I propose: "Terms that are unlikely to be called by the ambiguous title, but which are realistic search targets."
For the final paragraph of WP:PTM, I propose: "Some partial title matches may be listed in the "See also" section if they are realistic search targets. Otherwise, consider using the templates {{look from}} or {{intitle}}. These link to all articles starting with or containing the term, respectively." —Ringbang (talk) 23:48, 3 May 2017 (UTC)
We're close, but I think MOS:DABSEEALSO #6 should direct the reader to WP:PTM, instead of trying to partially repeat it, since that's what leads to contradictions, so: "6. Certain partial title matches."
WP:PTM should then include both criteria, as well as the illustrative moss example:
Individual partial title should not be listed unless:
1. They can be individually justified as reasonable search targets, and
2. They are not merely specific instances of a listed entry (For example, Spanish moss is listed on Moss (disambiguation), because it is not actually a type of moss.)
Otherwise, consider using the templates {{look from}} or {{intitle}}. These link to all articles starting with or containing the term, respectively.
swpbT 13:17, 4 May 2017 (UTC)
I've done the above. —swpbT 13:26, 8 May 2017 (UTC)

I've undone the addition because I don't see consensus for its inclusion. It seems to directly contradict the WP:PTM guideline by advocating the inclusion of PTMs. I also don't understand what these provisions are even supposed to mean. ("individually justified", as opposed to what, "generally justified"? Justified how? What does that mean?)  Sandstein  20:40, 29 May 2017 (UTC)

I think swpb's two-rule approach actual encapsulates current and best practice.  — SMcCandlish ¢ ≽ʌⱷ҅ʌ≼  00:00, 19 September 2017 (UTC)

New template {{canned search}} – use encouraged![edit]

A discussion at WikiProject Anthroponymy/Standards raised a common question that's shared by anthroponymy pages and dab pages: when are automated search results more helpful than curated pages? So I thought, why not provide both? I've made a template, {{canned search}}, that links to the automated search results for up to five terms. I envision this being used selectively, in the "See also" section of dab (and anthro) pages where the search results may offer value beyond what the curated page has. Hopefully, this will resolve a lot of disputes about whether to create a certain dab or anthro page or to leave a term to the search function. If it's well received, I'd like to insert a recommendation for its use on this MOS page. —swpbT 18:15, 4 August 2017 (UTC)


Peter Hall has recently changed to be transcluded into Hall (surname)#P [1] (by User:Michael Bednarek). I undid this once the dab had section headings added, which broke the transclusion. I've never seen transclusion used on a dab, and a quick search of the archive [2] looks like it's not commonly discussed, but has been negatively.

That dab advice at WP:Transclusion doesn't have a dab/list example despite referring to the third example (which is actually an article/embedded list), and seems to contradict our usual MOSDAB minimalism at MOS:DABICON Including images and transcluding templates are discouraged unless they aid in selecting between articles on the particular search term in question. , but I could understand the use with dabs and SIAs (more so than two dabs), so I've marked WP:Transclusion as disputed (pointing back to here). Thoughts? Widefox; talk 00:53, 13 September 2017 (UTC)

I've recently started using labelled-section transclusion to the same effect. Obviously, if the two pages are ordered differently, as with Peter Hall, then the result isn't going to look good. Otherwise, I don't see why not to, and there're clear benefits. If a reader is browsing the surname index for a certain Hall that they only know the occupation of, but not the given name (which in probably nine cases out of ten is precisely what these indexes are used for), then all the people with the surname should be accessible on a single page. Our current customary arrangement is to force such readers to click through to several dab pages before they're able to find what they're looking for. That's just bad. The way to avoid that without duplicating content is either via redirects (all the Peter Halls are listed at Hall (surname) and Peter Hall redirects there), or via transclusion. The latter is slightly better for our readers, and the only downside I can think of is making editing a little bit more difficult for newbies: I normally add an html comment explaining what's going on, but I have no idea how that works for people using the visual editor. – Uanfala 06:04, 13 September 2017 (UTC)
The upside is clear for the SIA at the expense of tags in the dab, but what are the downsides? When I added sections to the dab, the Peter Halls disappeared from the SIA. Would anyone notice? Is the dependency worth it? This Wikipedia:Transclusion costs and benefits doesn't say much. The LST adds a dependency of the section names. Result = fragile? Widefox; talk 10:54, 13 September 2017 (UTC)

Repeating section header in entry description?[edit]

If it's clear from the section header what the entries in a section are (e.g. "Songs" or "Ships"), I don't think it's valuable to repeat that information in every entry description: see my edit here, and to this MoS page here. I was reverted on both edits by JHunterJ, who says that descriptions should (always?) be a noun phrase. Not only is this guidance nowhere in the MoS, but even if it were generally accepted, I think it is harmful in such cases to insist on such a rule. We know how JHunter feels, so we're at 1 to 1 — how do the rest of you feel about the question? —swpbT 12:52, 14 September 2017 (UTC)

Meh, I can see it both ways. Outside of the fine point of when redundancy is useful, user:swpb if your edit is contested [3], it's reasonable to have that contested issue undone in MOSDAB [4]. Widefox; talk 14:12, 14 September 2017 (UTC)
I never suggested it was unreasonable: this is the D in a textbook WP:BRD. AGF Widefox: I don't dispute the propriety of the reverts, I dispute the value – and I resent any suggestion otherwise. —swpbT 14:24, 14 September 2017 (UTC)
Meh, my point is convince us, as neither side is in the MOS (which is the neutral description here). My recollection is that we encourage removing the redundant section name (generally), but as it's not in MOSDAB I guess that's in talk somewhere. I always remove the section name if used at the end of the description (countries etc), but optionally in the body. My take is that some links in this dab contain the redundant section name so I'd 1. make them consistent (similarly for the placement of dates), else if there wasn't any links repeating the section heading 2. remove them depending on how the dab looks/descriptions read. Generally remove, but not all redundancy is bad per se. I'd be in favour of adding to MOSDAB for the term at the end of the entry, but put value on scan speed at the start (consistency) or just call that personal style level choice, WP:CREEP. Widefox; talk 15:25, 14 September 2017 (UTC)
This is too abstract. Someone provide some examples where you think the repetition is vital. Others provide examples illustrating that it's terrible. We'll find the truth between the two extremes.  — SMcCandlish ¢ ≽ʌⱷ҅ʌ≼  23:54, 18 September 2017 (UTC)
To SMcCandlish and Widefox: I'm fine with it being clear that there is no general consensus for guidance either way. Let's get narrow: what about the specific page, Ecstasy, that started this? Do you think the repetition serves a useful purpose there or not? —swpbT 13:38, 20 September 2017 (UTC)
This edit strikes me as sensible and desirable, because we don't need to brow-beat readers with the obvious as if they have brain damage. In a section like "People", it is still of value to identify people more specifically, e.g. "biologist", "pop singer", "country singer", etc. At a shorter DAB page than Ecstasy a different approach would be to merge the "Music" subsections and just identify each entry as an album, song, whatever. On a long one, the subsectioning helps people find what they're looking for. But I'm still waiting for someone to make a case that, e.g. a section titled "Aircraft" should have entries like "* ABC-123, a British aircraft", etc., etc. If someone does want to make that case, I'll pre-counter it with the suggestion that it's much more sensible and helpful to have entries like "*ABC-123, a British biplane", and so on.

As for the guideline edit, I would support a version that said "it usually does not" rather than "it does not". There are going to be exceptions, e.g. when the heading in rather generic, and some entries are subtopically more specific. Take for example a "Songs" heading, under which an entry is '* "Ecstasy", a rap by DJ Funky Monkey', another is '* "Ecstasy", a hard-techno track by Mutual Mastication", and another is "* "Ecstasy", an a cappella song by Yojimbo X. Doodah'; then it would make sense for all the entries to have some kind of labeling like this, even if it's just "song", for consistency (though being genre-specific would be more helpful to the reader, e.g. "an R&B song"). This sort of situation is probably pretty common with "songs" in particular because the word is popularly used in a very broad way that means "individual pieces of popular music", even if for a piece that doesn't have a vocal performance in it.  — SMcCandlish ¢ ≽ʌⱷ҅ʌ≼  14:03, 20 September 2017 (UTC)

(after ec) To be clear, here is a diff illustrating the disagreement (and I can dig up many others similar to this). In this case I agree with swpb that including the words "a song" in the description of each entry in a section labeled as "Songs" provides absolutely no benefit whatsoever for readers. The justification that "descriptions are noun phrases" seems dubious to me. First, not all descriptions are noun phrases, and second, the context here is a list in which the description is a prepositional phrase. I don't see how this sort of list construct requires a noun phrase. olderwiser 14:09, 20 September 2017 (UTC)
To Widefox and Bkonrad: I like SMcCandlish's "usually does not" suggestion above for the guide; would you guys support that? —swpbT 14:17, 20 September 2017 (UTC)
Fine with me. It is usually impractical to expect style guidelines to provide certainty. olderwiser 14:19, 20 September 2017 (UTC)
I think I'll also change "repeated" to "repeated verbatim" to make it clear that nouns/noun phrases with greater specificity are fine. —swpbT 14:22, 20 September 2017 (UTC)
I actually disagree with this and prefer repeating it verbatim. #1, each entry should stand on its own, so I'm not a huge fan of doing something differently if there's 1 song vs. 10. #2, disambig fragments should generally be sentences if the comma is replaced by "is". So "Foo (song) is a song by Band Name" is valid; "Foo (song) is by band name" is much more abrupt, and is even weirder if the parens are ignored. SnowFire (talk) 15:37, 20 September 2017 (UTC)
disambig fragments should generally be sentences if the comma is replaced by "is". Why? How does the prolix repetition this can result in help readers? There's no reason to expect each entry to stand on it's own. For geographical entities, the description is often just a larger administrative area. Replacing the comma with "is" in such cases would result in comical effects, eg, * Green Lake (Cariboo), British Columbia would become * Green Lake (Cariboo) is British Columbia. And similarly, for geographic entities, where they are grouped by country (or state) we do not repeat the country or state in the description, although that would be expected if the entries were treated as stand-alone. olderwiser 15:53, 20 September 2017 (UTC)
(ec) Specifically, at Ecstasy I'm not sure it looks better without the "song"s or helps the reader much and we can't remove the redundancy with the first two visible links anyhow. I'm not convinced there's a clear-cut winner needing a rule. Meh. We do already have a redundant guide for those links Ecstasy (ATB song) and Ecstasy (Jody Watley song) at MOS:DABENTRIES - avoid "Dark Star" (song), a song by the Grateful Dead". The wording [5] is OK (cut "verbatum"), but it may be simpler to just add a follow-on example for when the redundancy comes from the section name. Widefox; talk 15:59, 20 September 2017 (UTC)
Obviously entries with parenthetical phrases are often going to repeat the header. That's not any kind of reason to introduce that repetition into other entries where it doesn't need to be. I'm not satisfied that people read the existing guidance against entries like "Dark Star (song), a song ..." as covering section headers too; if they did, we wouldn't be here. —swpbT 17:04, 20 September 2017 (UTC)
Concur with Bkonrad; DAB page entries are MOS:LISTS. Most list entries are not full sentences, so there's not any reason to expect them to be on DAB pages. The vast majority of cases are not, but are of the form "* Foobar, a baz from quux", and the like. Not sentences.  — SMcCandlish ¢ ≽ʌⱷ҅ʌ≼  16:53, 20 September 2017 (UTC)
Exactly. No one can give any justification for this "should be able to start with 'is'" or "has to be a noun phrase" rule besides ILIKEIT. Well, thats's a minority position. —swpbT 17:04, 20 September 2017 (UTC)

(de-indent) I'm not sure if people misunderstood my point above, or what, but for the record, I was not arguing that commas actually be replaced by the word "is"; I said that disambig entries should be readable as sentences if you replaced the comma with "is." (And that isn't from me, that's been around forever.) And more generally, you should be able to completely ignore the phrase in parentheses and read the sentence with just the main word, so don't rely on the parenthetical disambiguator in the description. Users don't always read section headers, either.

swpb, this is a policy page, not an article. This is where people advocate for guidelines, not argue whether an article is in compliance with existing ones, so "WP:ILIKEIT" is an irrelevant attack; someone could reasonably come here and suggest all disambig pages use a 30 point purple font if they wanted. The whole POINT of such pages is to advocate for what you think is better. That attack is just as easily turned on your opinions; since that would be stupid, can we agree that everyone's opinion has merit?

older != wiser, you call it "prolix repetition", but I see it as helpful and clear. Green Lake is a terrible disambig page IMO that is far too over-concise and doesn't describe its entries at all. What you describe is clearly silly, but an entry that said "Green Lake (Cariboo), a lake in the South Cariboo region of British Columbia" would be more to my preference. SnowFire (talk) 19:06, 20 September 2017 (UTC)

First, I'm sure everyone got that you don't want to literally replace commas with "is"; we understood exactly what you meant, but we don't agree, and we don't think that's ever been the guidance, explicit or otherwise. Secondly, no, I don't agree that these are just opinions with equal merit. There is standing guidance against unnecessary repetition, because it makes it slower for the reader to get to what they need. It's always going to be allowable to repeat things when that aids clarity; I don't think anyone here disagrees with your "Green Lake" entry, so that's really a strawman. This is about one specific kind of repetition, exemplified by Ecstacy, that is both unhelpful and already implicitly discouraged. —swpbT go beyond 19:17, 20 September 2017 (UTC)
I consider a gain for simplicity is to remove all the possessives in the song descriptions...e.g.

We don't need to maintain the gender (which involves checking the article body or even another article, so an extra chore), and just replacing the possessive is reason enough.

Green Lake could benefit from more description to disambiguate the two British Columbia entries. I've cleaned it up. I didn't add the redundant location info that's already in the parentheses, but no objection to them being added. Widefox; talk 20:21, 20 September 2017 (UTC)
The only major shortcoming I see at Green Lake is that it isn't always clear whether the entry refers to a body of water or a settlement of some sort. olderwiser 21:35, 20 September 2017 (UTC)
Re possessive's in song entries, the artist is redundant in the example just given. I'd frame that particular entry as
Or in fact, given more typical practice, since there is an article and it includes the artist in the title, no additional description is really necessary. The year might be included as some reader seem to find that useful. I think I'd need to see the suggestion about possessives in some actual lists to give more general opinion. For me, when I scan the list, I'd want to see the entries in roughly the importance for purposes of figuring out which I want. So if an song did not have an article, I think the next most useful bit of information is the artist more so than the album or the year. olderwiser 21:41, 20 September 2017 (UTC)