Wiktionary talk:Votes/2024-07/Remove "Quotations" sections
Ambiguity between senses
[edit]What to do with the quotation at how's your father? It ambiguously uses both senses 1 and 2, which is why I placed it in the separate Quotations section.
If more examples like this can be identified, I think we should restrict the use of Quotations to precisely this situation - cases where the quotation is illustrative and relevant (and does not deserve to be relegated to the Citations page) but does not neatly fit under any particular sense. This, that and the other (talk) 01:19, 17 July 2024 (UTC)
- There are plenty of languages where names for plants, fruits and animals differ by dialect, and the same name may point to (slightly) different ones (cf. mesimarja). I'm sure that for those terms, an ambiguous quotation would not be rare. Thadh (talk) 08:04, 17 July 2024 (UTC)
- I do wonder how common ambiguous senses would be in these cases. A good majority of the instances of words in these cases, I believe, would have clues (either from context or from the place the work was made) that would point to a certain meaning. (There could also be a template like the rfquote one for quotes that one isn't sure how to place.) CitationsFreak (talk) 10:13, 21 July 2024 (UTC)
- AFAICT/IMO it'd be fine to move the quotation at how's your father to the Citations: page; I can't say I share the view that it is particularly illustrative or important. More generally, I doubt any citation which is impossible to assign to a definition is so much more illustrative than definition-specific quotations as to merit the extra prominence of its own section while the others are collapsed under definitions. I don't want to make policy based on hypotheticals, but if there were a truly exceptional quotation that exemplified something about the usage of a term but wasn't assignable to a definition, my inclination would be to present it with an explanation of what about the usage of the term it exemplified (e.g., that usage was typically ambiguous, or that authors relied/played on ambiguity), in ====Usage notes==== (like some entries like -' currently use quotes to illustrate points made in ====Usage notes====). In cases like mussel where most quotes don't give enough detail to tell which of several similar organisms is meant, the question is (AFAICT) often moot as there is a general sense they can go under (as at mussel), but otherwise sense-unclear quotations can go on the Citations: page, IMO. - -sche (discuss) 12:29, 17 July 2024 (UTC)
- @-sche: With LDLs, the issue is not how important a quotation is, but that one actually exists. If all you have is one or two quotes of which one is ambiguous, and a dictionary entry saying that both senses exist, I would think it's preferable to include the quote even though it is ambiguous. Thadh (talk) 13:28, 17 July 2024 (UTC)
- Sure, but (if it can't go under a definition,) that quotation can be included on the citations page. - -sche (discuss) 02:00, 22 July 2024 (UTC)
- @-sche: With LDLs, the issue is not how important a quotation is, but that one actually exists. If all you have is one or two quotes of which one is ambiguous, and a dictionary entry saying that both senses exist, I would think it's preferable to include the quote even though it is ambiguous. Thadh (talk) 13:28, 17 July 2024 (UTC)
- @-sche: While the reasoning about outdated text in WT:EL is correct, due to the reason circumscribed in this section it does not follow that we should remove the section type, as it does not appear superfluous.
- I like to use it as a fallback an dump for rather obscure entries when the Citations namespace would be too likely not opened, due to the additional effort (reader inertia) or due to the citations page not being noticed, though I know about
{{seeCites}}
; currently, since a few weeks ago, in Vector 2022 I even have many false blue links to the Citations and Talk namespaces. Examples are دِيس (dīs), because for grasses in medieval you can never be sure with reasonable effort which species was designated in those instances, and أَنْبَج (ʔanbaj) containing early attestations of various spellings or alternative forms so they aren’t spread around. In كَاغِد (kāḡid) and شَيْرَة (šayra) you also see a smooth transition of quotation sections to references sections. - Removing the section altogether is to see a problem that is not there; this section header is so rare that employing it reflects advanced knowledge. It is also inconsistent with the possibility of adding usage examples (apparently as remarks in usage notes, where sometimes even quotes are used) and synonyms in separate sections. Fay Freak (talk) 19:46, 17 July 2024 (UTC)
- @Fay Freak Sections for synonyms (and other -nyms) for specific senses are suppose to have
{{sense}}
for each separate definition. DCDuring (talk) 20:56, 17 July 2024 (UTC) - For things like dīs, I would consider adding a "supersense" along the lines of "Any of several similar grasses having [describe their common attributes]" with the specific senses as subsenses; quotes that only support "grass" but not a specific grass could go under the "supersense". (Failing that, I still don't see a reason not to put ambiguous cites on the citations page, and stick
{{seeCites}}
under whichever sense(s) it might be. Indeed, I feel like our tendency to make links to the Citations: page sense-specific might be something we should re-examine; maybe, to TTO's point at the bottom of this page, we should stick "see quotations" after the headword line or something, along with T:tlb.) - -sche (discuss) 21:13, 17 July 2024 (UTC)
- @Fay Freak Sections for synonyms (and other -nyms) for specific senses are suppose to have
Pinging people from the BP discussion: @RichardW57m, Benwing2, Sgconlaw, DCDuring, Andrew Sheedy, AG202, Urszag, Ioaxxere. What do y'all think of this idea? (My inclination is still to remove the sections; IMO anything that can't go under a sense can go on the citations page, and if we have Quotations sections at all, there's no chance people will only use them for ambiguous quotes: we already see people use it for single [Bible] quotes they consider 'important', which no version of the policy has authorized, and we see people intentionally misformat [as usexes] random quotes [below senses] that they consider 'important' and don't want javascript to collapse. But if there's significant support for this, it'd make sense to split this into e.g. "Option 1: remove...", "Option 2: repurpose...") - -sche (discuss) 20:00, 17 July 2024 (UTC)
- @-sche: I also favour removing the sections. — Sgconlaw (talk) 20:05, 17 July 2024 (UTC)
- Yeah I also think we should remove the sections. Benwing2 (talk) 20:42, 17 July 2024 (UTC)
- Remove. We will be creating a somewhat misleading impression that all uses belong solely to a single definition, but I don't see how we can avoid that common problem for polysemous terms. DCDuring (talk) 20:56, 17 July 2024 (UTC)
- @DCDuring: How are we creating this impression is there is no
{{sense}}
in the quotation section and multiple senses under the part-of-speech header? Adding quotes without assigning them to numbered senses looks intentionally neutral to me, towards any statement about what sense or possibly multiple senses the quotation belongs to. - Indeed when a speaker uses a term he has had multiple impressions to which he connects the eventual use and he might not realize that the organism name he uses referred to different things each time in his life; at one time دِيس (dīs) was “rush”, another time it was “club-rush”, at a different place it was “diss”, and the medieval Arab did not have the taxonomic education to know that these things have no relation to each other, so later on it is left open what the mat is made of he has been informed to be from دِيس (dīs). I might also define حَمْض (ḥamḍ) with many definition lines rather than “a term used by Bedouins for saline, mostly chenopodiaceous and zygophyllaceous plants”, and only from this circumstance, arbitrary splitting or combining of senses, it depends then if we can include a quotation in the mainspace or only the citation-space. The removal of this section seems also like a nudge towards unreliable specificity. But perhaps it wasn’t even a sense that is illustrated but form and environment.
- The four examples above I can only imagine made worse by the removal of the generic quotation space in entries. Fay Freak (talk) 21:34, 17 July 2024 (UTC)
- @DCDuring: How are we creating this impression is there is no
- I'm in agreement with the above. Ambiguous cites can be interesting, but I'd rather see the Citations namespace get more use than see unnecessary citations clutter up the entry.
{{seeCites}}
can be used below specific definitions that apply rather than being the only thing in Quotations sections (in the case of ambiguous cites, the template could go under both definitions. I suspect it's very rare for a usage to be ambiguous between more than two senses). Andrew Sheedy (talk) 21:28, 17 July 2024 (UTC)- How do they clutter if they do not expand unless selected? Fay Freak (talk) 21:34, 17 July 2024 (UTC)
- Also, how do we refer to the citation namespace under specific senses if for a given specific sense we don’t know whether the citation namespace has quotations specific to this sense? We would need to put
{{seeCites}}
under some section – which happens to be Quotations. Even when the senses of quotations are clear it is often still better to only link the citation namespace once and not under every sense, seen on yawk. Fay Freak (talk) 21:39, 17 July 2024 (UTC)
I believe that the way the proposal is written right now is exactly the way I have been using Citations pages for (1) citations for potential senses that I'm not sure have or will ever reach WT:ATTEST levels (Citations:Tusalla, Citations:Hasantu), (2) citations with ambiguous or undetermined meaning, but apparent importance or relevance (Citations:hobbit- the earliest quote, Citations:Taiwan- the 1600 quote, Citations:Badong- I believe that "Badong, Hubei" is what this NYT quote means, but I technically don't know if there's some other Badong), (3) citations that are weaker or less important than others that are on entry but still illustrative under Wiktionary:Quotations in some way (Citations:transgender, Citations:Beijing), (4) citations that are good, but include fringe content that is extremely distracting and hence not good for the dictionary mainspace, and (5) citations that are for spellings of words that are either close alternative forms or synonyms of the entry term (which do not yet have their own independent entry) that would be interesting to someone learning about the entry term. I have never even once used a Quotations header on an entry; I was never told to use them and I just kind of left them alone. --Geographyinitiative (talk) 10:26, 20 July 2024 (UTC)
What text exactly is being changed?
[edit]Would it be possible to add the policy text changes to the vote proposal? I know it's not required, but I find it helpful when it's made abundantly clear what text is being added/changed/deleted in the vote itself, rather than finding out afterwards (even if it's fairly obvious as in this case). Something similar to Wiktionary:Votes/pl-2022-07/Stubifying alternative forms or Wiktionary:Votes/2024-06/CFI for mainspace constructed languages. AG202 (talk) 17:49, 17 July 2024 (UTC)
- How's this? (If the text of the wording acknowledging Citations: pages exist proves contentious I am happy for it to be a separate discussion.) - -sche (discuss) 19:48, 17 July 2024 (UTC)
- Looks good, thank you! AG202 (talk) 23:01, 27 July 2024 (UTC)
It might be helpful to specifically mention this template, since I believe it is predominantly found under a Quotations header. Perhaps there could be an additional sentence along the lines of "If there are citations found on the Citations page that are not in the main entry, the {{seeCites}}
template may be placed in the entry, following the citations for each of the relevant senses." Andrew Sheedy (talk) 21:32, 17 July 2024 (UTC)
Statistics on usage of Quotations header
[edit]The Quotations header is really not used all that often. It appeared 6200 times in the most recent database dump. Most instances of the header (89%) are immediately followed by {{seeCites}}
or {{seeMoreCites}}
.
Here's a language-by-language breakdown of the entries that would need to change were this vote to pass. Many are ill-formatted or contain only one sense under which the quote could easily be placed.
This, that and the other (talk) 02:59, 19 July 2024 (UTC)
- I should add that, having reviewed a random sample of this data and not found a single entry where the Quotations header is really the best approach for displaying quotations, I currently plan to support this vote as it is written. I'd be happy to be proven wrong - if someone can identify entries from this list that really can't work without the Quotations header, I (and others) might change my (their) mind. This, that and the other (talk) 03:02, 19 July 2024 (UTC)
- It's also worth noting that several of the terms with other content actually just point to the Citations page using a manual link rather than a template (e.g. /b/tard, Gingging, or Barney-style--which actually links to the discussion page!), or even completely misuse the section (like arsewards, which uses it to link to the OED entry or bʼakʼtun, which uses it to house encyclopedic/etymological information). Latin achaemenis and antepagmentum just contain requests for quotations. And those are just from a small selection that I looked at. Andrew Sheedy (talk) 04:53, 19 July 2024 (UTC)
- Yeah, the use of this just to provide a "seeCites" link is something I find particularly unnecessary; there's a Citations tab on every page. In the past, someone argued it was necessary because on some mobile devices there's not a link to the Citations tab, but if that's true that's a problem we need to fix independent of whether we have a Quotations header or not! (I moved the content of bʼakʼtun's quotes section to the talk page.) - -sche (discuss) 21:32, 19 July 2024 (UTC)
- Yes, I agree. The fact that nearly a full quarter of our Old Japanese entries have a completely useless header is indicative of it being used as filler more than anything else, really. Perhaps @Poketalker can explain their reasoning, but to me, it just looks like padding out the entry. Theknightwho (talk) 13:16, 20 July 2024 (UTC)
- I think we could do better at making the Citations page more visible - after all, all the other tabs along the top tab bar are editor-focused, not reader-focused, so it's easily missed or ignored. A bespoke header just for this purpose probably isn't the way to go though. Perhaps greater use of
#*
? Or something different. This, that and the other (talk) 04:57, 21 July 2024 (UTC){{seeCites}}
- I wonder if, instead of (or in addition to) having sense-specific
{{seeCites}}
s, we should have a "see quotations" link after the headword or something. - -sche (discuss) 02:02, 22 July 2024 (UTC)
- I wonder if, instead of (or in addition to) having sense-specific
- I think we could do better at making the Citations page more visible - after all, all the other tabs along the top tab bar are editor-focused, not reader-focused, so it's easily missed or ignored. A bespoke header just for this purpose probably isn't the way to go though. Perhaps greater use of
- Yes, I agree. The fact that nearly a full quarter of our Old Japanese entries have a completely useless header is indicative of it being used as filler more than anything else, really. Perhaps @Poketalker can explain their reasoning, but to me, it just looks like padding out the entry. Theknightwho (talk) 13:16, 20 July 2024 (UTC)
- As an academic, I can make conclusions from quantitative studies. With this vote, I expect to pass, we will streamline our complexity. I have removed the Arabic occurrences already. Fay Freak (talk) 05:57, 22 July 2024 (UTC)
More to add
[edit]@-sche would you consider modifying the vote so that it removes (a) the sentence "Quotations here are formatted normally but without definition numbers", which would be redundant if the vote passes, and (b) the "hrunk" example, which is a confusing example because it conflates noun and verb senses of this putative word?
I'd advocate for replacing the "hrunk" example with a new paragraph that simply says
For details of how to format quotations, see Wiktionary:Quotations.
This, that and the other (talk) 03:25, 22 July 2024 (UTC)
- I had left "Quotations here are formatted normally but without definition numbers" alone because, if the rest of the paragraph is changed as proposed (from being about ====Quotations==== sections to being about Citations: pages), that line seems to remain valid in the new context. But I also have no objection to removing it. (I think we could/should replace the fake "hrunketh" cite with a real cite even without a vote.) - -sche (discuss) 04:30, 22 July 2024 (UTC)