Wiktionary:Votes/2008-01/Homophones section

• Voting on: Establishment of a new L4 Header ====Homophones==== for the Pronunciation section.

  Currently, information about homophones is intermingled with information about the pronunciation of the word itself (phonetic transcriptions, audio files, rhymes, hyphenation). But all homophone information is about other words, not about the current entry itself, and so its content parallels the sections Synonyms, Related terms, Derived terms, etc. in that respect. Since these other groups have separate subsections, it seems reasonable to do the same with Homophones. This would necessitate a change to WT:ELE and Wiktionary:Pronunciation, but we can hold off on writing the new text for those pages until we have agreed to have the section and then hammered out the desired style (see archived discussion below for several previous suggestions).

• Vote ends: 26 February 2008 23:59 UTC
• Vote started: 27 January 2008 23:59 UTC
• Vote created: EncycloPetey 18:48, 27 January 2008 (UTC)
• Discussion:

  Wiktionary:Beer_parlour_archive/2007/April#Homophones_as_a_L4_header (archived discussion)

  Wiktionary_talk:Pronunciation#Homophones_section (discussion of possible format, should this proposal be approved)

1. Support EncycloPetey 18:49, 27 January 2008 (UTC)
2. Support —Ruakh_TALK 19:08, 27 January 2008 (UTC)
3. Weak Support. In many cases this would probably be preferable, and I wouldn't mind seeing it at all. But we can't rule out other ideas this early, especially since there is tons of information that could go in the pronunciation section that to date I've never seen in even a single entry. DAVilla 20:32, 29 January 2008 (UTC)

   I have changed weak support to support because this vote does not say that the homophone header must be used exclusively, and I'm in favor of having flexibility where we haven't pinned down standards. DAVilla 07:03, 11 February 2008 (UTC)

4. Support Connel MacKenzie 04:32, 2 February 2008 (UTC) I said elsewhere, that I thought this already had general approval. I am convinced (or reminded?) by EncycloPetey's examples below, that the result is a Good Thing. Most bot work I've done has had to account for the heading existing inconsistently. Having only one place to look for these, would be kinder to bots and humans alike. One concern I didn't see asked nor addressed, is about blank ===Pronunciation=== sections (i.e. that have only a subordinate homophones heading.) To humans and bots, it doesn't make much sense to allow a blank pronunciation section, if the word can be confused for another (by it's pronunciation.) (I apologize if it is addressed somewhere, but I simply missed it.) --Connel MacKenzie 04:32, 2 February 2008 (UTC)

   If you generate a list of such pages, I will work to add pronunciations to those pages. You are correct that words having homophones really need to have a transcription of their pronunciation given and probably audio as well. --EncycloPetey 04:40, 2 February 2008 (UTC)

   Well, that's not quite what I meant. Does this proposal say anywhere that the blank section is no good, reminding people to use {{rfp}} instead? It should, if it doesn't already. As far as the list: I generated one a long time ago for Dvortygirl to add audio to. We should have a new dump tomorrow - I'll try to do something after the routine stuff is done. --Connel MacKenzie 14:19, 8 February 2008 (UTC)

   OK, I understand what you mean now. That would be a matter for explanation in ELE or Wiktionary:Pronunciation. This proposal is merely about formalizing the Homophones header, not about specific updates to the style guides. --EncycloPetey 19:13, 8 February 2008 (UTC)

5. Weak support. I think the pronunciation header is where this information should be. But as most words have only one homophone (if any), I feel a sub-header is like a sledgehammer to crack a nut. Only useful for the limited number of words that have several examples. -- Algrif 17:07, 4 February 2008 (UTC)
6. Support Nbarth (email) (talk) 18:27, 17 February 2008 (UTC)

   While agreeing with Oppose votes that homophones depend on pronunciation, I think the analogy with Synonyms and Antonyms (in EncycloPetey's response to Robert Ullmann) is persuasive: just as we list the individual definitions, uncluttered with synonyms and antonyms (which merit their own section), so too should pronunciations be uncluttered with homophones.

   I think this is ultimately a question about abstract correctness versus usability. For instance, to be very correct across multiple languages, perhaps we should index words by phones, not by spelling—after all, if you hear a word, how should you look it up? Indeed, perhaps for a given pronunciation, we should list words in other languages that have the same pronunciation! As abstractly correct as this is (and as useful to Macaronic writers), the usability problems it causes are clear.

7. Support H. (talk) 19:35, 29 February 2008 (UTC) Seems reasonable, though I’ll have to delete or edit {{homophones}} then. H. (talk) 19:35, 29 February 2008 (UTC)

1. Oppose Widsith 19:03, 28 January 2008 (UTC) I prefer to have it simply as a line in the Pronunciation section, which I find less cluttered. Widsith 19:03, 28 January 2008 (UTC)

   Please see my Rationale below; it will probably interfere with format if I place it here. --EncycloPetey 02:30, 30 January 2008 (UTC)

2. Oppose Wytukaze 19:36, 29 January 2008 (UTC) I'd prefer to have homophones subordinate to each pronunciation variant they correspond to. That is, under each different pronunciation (or next to, or whatever), not even after the pronunciations as a whole, and certainly not as a separate header. It seems needlessly redundant otherwise; after all, homophones by their very nature are dependent on pronouncing a word in a certain way. --Wytukaze 19:36, 29 January 2008 (UTC)

   That would make the Pronunciation section even longer than it is now. Right now, all the pronunciations possible for a single region are listed on a single line. To do what are are proposing, we would have to list each possible pronunciation on a separate line, which could triple the length of some Pronunciation sections and make it very difficult for users to see that variation exists within a region. --EncycloPetey 02:24, 30 January 2008 (UTC)

3. Oppose. What Wisith and Wytukaze said. Note that the current suggestion for how to format the Homophones section, if one is approved, has it listing pronunciations (in IPA, say) followed by the homophones. That is, each pronunciation will be transcribed under Pronunciation and then (those with homophones) again under Homophones. I do not think that that is better than what we have now. (But my "oppose" vote is not merely against that specific suggestion: it's against this proposal.)—msh210℠ 20:54, 29 January 2008 (UTC)

   But very, very few homoophones will require any such formatting. And note that under the Pronunciation section, a given pronunciation already can appear multiple times, if it occurs in more than one speech region. To do what you ask will greatly lengthen the Pronunciation section, not keep it concise. Also, you have not given any reason for opposing the Homophones header, you have merely given reason for disagreeing with one possible format that was suggested for the section. --EncycloPetey 02:27, 30 January 2008 (UTC)

   First, a given pronunciation shouldn't appear many times, once for each region: we can combine them, like (Wales, Australia, and Pitcairn Islands) /fu/ (unless Pitcairn Islands also has another pronunciation not shared by Wales). Second, since this is a formal vote, I don't need to give a reason, but note that I did so: I wrote "What Widsith and Wytukaze said".—msh210℠ 17:12, 30 January 2008 (UTC)

4. Oppose Conrad.Irwin 22:45, 29 January 2008 (UTC) I am against formalizing the use of yet another heading, in the few cases where there are five or more then it is worth having a seperate heading, for the large majority of cases though it is needlessly big and ugly. Conrad.Irwin 22:45, 29 January 2008 (UTC)

   I don't believe the proposal states under which cases it should be used. If you think it's worthwhile in a few cases, then it would be better to not have either formalized, or to accept either as alternatives. At present the header could be considered entirely incorrect. DAVilla 16:18, 30 January 2008 (UTC)

5. Oppose Robert Ullmann 11:23, 6 February 2008 (UTC)

   It must be "intermingled" with the pronunciations because it is specific to pronunciation and accents. Should just be a *: bullet line under the specific accent, as with rhymes (and for very similar reasons). I.e. concur with several previous comments. Robert Ullmann 11:23, 6 February 2008 (UTC)

   But we don't do other similar sections that way. Synonyms and antonyms are specific to particular definitions, and we don't do Synonyms and Antonyms that way. We list them in a separate section because they are separate words. Likewise, homophones are separate words, not an aspect of the Pronunciation of a word. --EncycloPetey 07:13, 7 February 2008 (UTC)

While I wholeheartedly support separating homophones off from the pronunciation section, I disagree with nesting it under the pronunciation section. The reason being is that this means it is yet another piece of information coming before the definition. While I can see the logic of having it a subsection of pronunciation, the simple fact is that a lot of our good entries suffer from forcing the average user to sort through a bunch of crap to find what they're looking for. Perhaps if/when we get the new format which allows users to easily hide the stuff they don't want to see, it'll be more practical, but for now I would only support it going after the POS line. Atelaes 02:27, 28 January 2008 (UTC)

Consider the anonymous comment posted earlier today at Talk:discrete. The user wondered why discrete wasn't defined as "lay low or unnoticable". That is the definition of (deprecated template usage) discreet, which is listed on the page as a Homophone, but in the discreet fashion currently advocated by ELE. If there had been a separate Homophones section, the other word would have been more obvious and the user would not have been confused as he was. In this case, the homophone what exactly what the user was looking for. We cannot assume which information a user will be most interested in.

I cannot understand wanting to put this after the POS header. Homophones have nothing to do with the entry on which they appear, except in terms of pronunciation. If (deprecated template usage) right has a homophone (it has several) would you advocate listing them under the noun, the adjective, the verb, the adverb, the interjection, or all five sections? Putting them under the Pronunciation (near the pronunciation of the word and rhymes link) makes far more sense. --EncycloPetey 02:32, 28 January 2008 (UTC)

Well I am all for having everything (including Pronunciation and Etymology, Homonyms by extension) below the definitions - however that is not the consensus here. What would be good is if our {{see}} templates also contained basic homophones, similar words, and other things that people might actually be looking for. Conrad.Irwin 22:55, 29 January 2008 (UTC)

That would overburden the template because homophones are dependant upon language. That is, we could end up with long lists that confuse people because the link is for a homophone in just one language listed on the page, even if the page contains multiple languages. Homophones must be listed within a specific language section to be of any use. The {{see}} template is only for diacritical variants that users may have trouble typing and for capitalization variants. As such, it precedes all language headers. The two ideas are mutually incompatible. --EncycloPetey 02:22, 30 January 2008 (UTC)

1. Abstain - I'll go with what's decided since I do not have a strong opinion on this. Williamsayers79 23:12, 8 February 2008 (UTC)

Below are examples of how the entry for the word (deprecated template usage) for would look under current ELE (using the Oppose comments above) and under my proposal. The formats below assume additional audio files will accumulate over time. My proposal works well and simply, because (as with most cases of homophones) the homophone is not dependent upon regional dialect. The opposition votes above have voiced a desire to see homophones listed after the specific pronunciation with which they compare, and listing each pronunciation separately. In this case, doing so would greatly lengthen the entry, and would make it much harder to read and parse the content.

Here is how the relatively simple case of (deprecated template usage) farther would look under each proposal:

In both examples above, it should be clear that my proposal does not produce clutter (despite what some opposition comments claim), nor is it needlessly big and ugly. --EncycloPetey 03:13, 30 January 2008 (UTC)

The more I think about it, the more I wonder if hiding the whole pronunciation section might be a good idea. Atelaes 03:20, 30 January 2008 (UTC)

Not if it's hidden as a default. The Pronunciation section is one I use a lot (and not just here). The same will be true of English learners, who often are more interested in a translation and pronunciation than in a definition. --EncycloPetey 03:37, 30 January 2008 (UTC)

In the for example above, why not add Homophones as an extra bullet point, at the same level as "stressed" and "unstressed", rather than as a separate heading altogether? Widsith 16:23, 30 January 2008 (UTC)

1. Why do you list "RP" (Received Pronunciation) at the same level as "US"? There is no single U.S. dialect: it would be General American or something else. But then none of those pronunciations would get slashes, they would get brackets.
2. Why under RP do you list "father (in non-rhotic accents)"? RP is a non-rhotic accent, so farther is simply a homophone. At the same time, the other pronunciation is rhotic, so it shouldn't list any homophones.
3. Why do we have pages like Rhymes:English:-ɔː(r) and Rhymes:English:-ɑː(r)ðə(r)? Are we suggesting that words that are not at least sometimes rhoticized could not rhyme with words that sometimes are?

It seems fairly straightforward to me to say that several problems could be solved by simply accepting that there are two major varieties of English and listing broad pronunciations phonemically under each variety, instead of having to hack the rhotic and non-rhotic accents into the same pronunciation using parenthesis. Parenthesis should be reserved for sounds that are sometimes made with the same accent, such as dropped syllables or liasons at the end of a word. Otherwise what's to stop us from writing /pik'ante(ɪ)/ in the translingual section? DAVilla 16:52, 30 January 2008 (UTC)

The reason "Homophones" is not listed alongside "stressed" and "unstressed" as a bullet point is that: (1) on the left example, I was following the opinions of the oppose votes above, namely that all items pertaining to a specific pronunciation should be listed under that pronunciation. (2) Placing it as a third bullet point implies that it is somhow a third option in pronunciation; which it is not. In a bulleted list, all items should be parallel to make logical sense. The list "stressed, unstressed, and homophones" is not a parllel list and so should not be set up to resemble such.

The choices in displays above were chosen to most closely match current practice. Much of the Pronunciation section formatting is not standardized at all right now. This Homophones vote was one step towards raising awareness to this and getting people to talk about how we want the section formatted. Since Homophones (as a separate section) would not impact the remaining format, this seemed like the best place to start discussion. The Rhymes pages were initially set up long ago, and not by me. The pagenames use (r) wherever there is a rhotic/non-rhotic potential. I assume this choice was made because some major dictionaries that use IPA do the same thing, and the intention was to combine UK/US potential homophones in such cases on a single page. --EncycloPetey 04:14, 2 February 2008 (UTC)

• No consensus. Note: This topic may merit further discussion and a future related vote. —Ruakh_TALK 23:01, 13 March 2008 (UTC)

• Passes (just) with a vote of 6-5-2 (or 7-5-2 with one late vote). This could have ended up "no consensus", however one of the abstentions is explicitly "I'll go with what's decided." --EncycloPetey 17:33, 1 March 2008 (UTC)

I don't think 7-5-2 is a passing consensus at all. As well, respectfully, I don't think it is appropriate for a vote's proposer to be the one to close it if it's not unanimous. Dmcdevit·t 20:12, 1 March 2008 (UTC)

I agree, except that "unanimous" seems like too high a standard. I'd say that as long as there's very clear consensus, it doesn't matter who closes it. (I mean, assuming they close it in the same direction as the clear consensus. :-P) —Ruakh_TALK 21:04, 1 March 2008 (UTC)

Incidentally, I also don't accept the assumption that "I'll go with what's decided" means "I cast my vote with the majority." —Ruakh_TALK 21:15, 1 March 2008 (UTC)

Respectfully, no one else closed this even though we're now four days past the close date. We have effectively 7-5-1 (see wording of the abstentions), which is greater than 50%. If you disagree with the count, you are welcome to raise discussion about interpretation. If you're concerned about the closing procedure itself, then please write a policy document to guide the rest of us. No such policy has ever been put forward to my knowledge. If you are wanting to see votes closed in a timely fashion, then thanks for volunteering to help out with this and please add your name to WT:DW. --EncycloPetey 21:10, 1 March 2008 (UTC)

Well, I guess my thinking is that there is nothing urgent about closing these votes. If there is a consensus (or not), there's no reason someone with an opinion can't just wait for an outsider to close it. Probably it wouldn't be a big deal most of the time anyway, but it just seems like poor form, and we can avoid the entire issue by exercising this common sense. Dmcdevit·t 21:20, 1 March 2008 (UTC)

I agree with DmcDevit. The vote was clearly divided, and I think that "No decision" would have been a more accurate decision for this vote. However, I also agree with EP that a policy on how exactly to interpret votes might be useful. 10-1 is a clear majority, and 5-5 is a clear deadlock, but we don't have a lot of guidelines on what in between those count as. -Atelaes λάλει ἐμοί 21:38, 1 March 2008 (UTC)

Four days is a pittance. I waited for objections for an entire month before closing one of my own votes, the second brand names vote which had objections and staunch opposition and all that despite a supermajority. Thank you for welcoming us to "raise discussion about interpretation", but I'd sooner just push for reinterpretation. Not having a written policy on what determines consensus is not sufficient reason to call the vote as you wish. Votes like this, such as the first brand names vote, have failed in the past. Even before that, we've had a precedent of using a supermajority, and if the question was even raised, I don't remember anyone challenging it. Anyways I feel strongly enough about this that I am willing to change my vote just to make sure it does not have a simple majority, if that would clarify the decision. I can't believe how easily you can approve your own proposal considering how quickly you call foul. What then, you expect us to cave in to you but not the other way around? And this is not the first vote that you have closed controversially. Sorry, but this proposal did not pass because it does not have the clear consensus of the community. DAVilla 05:41, 13 March 2008 (UTC)

Yes, we're all aware that you see voting as a strictly numerical exercise, but that's not what consensus is about. Did you read the reasons for oppose votes? Conrad says "in the few cases where there are five or more then it is worth having a seperate heading"? That is, we have an "oppose" vote that actually thinks the header is worth having in some cases, which is what a "support" vote is for. And please don't nurse sour grapes in future; that's unbecoming. Controversy will happen here, because that's part of commmunity discussion. It's better to move past disagreements rather than harbor resentment. --EncycloPetey 14:44, 13 March 2008 (UTC)

Okay, I'll agree to leave the sour grapes out of this. Yes, I did read Conrad's comment, and I responded to it because his vote seemed miscast. So yes, it seems like this should have passed, but I don't think that's our decision to make. I'm more than confident in Conrad's ability to place his vote correctly if fully informed. Even if some guidance may be necessary, it's still Conrad's decision, not yours or mine, and neither of us could any sooner strike his vote or move it to a different section than count it differently. Forgive the numeric bent, but the tally is 7-5-2, not 8-4-2-ish. What I might have suggested doing is extending the vote instead, and the late vote and even the one prior would have been a good excuse to do that. An alternative at this point would be to just drop this vote, and to reword it saying that a Homophones header is permissible, but not replacing the inline version in every case. I'm certain that at least that much has support, but if it was the intention here, then it wasn't clear. DAVilla 04:17, 15 March 2008 (UTC)

There is no conceivable way 6-5-2 can be considered "passing". It is not a consensus by any stretch, and the "support" votes are short of a majority. (And trying to "reinterpret" Connel's the other votes is ridiculous; if they wanted to vote "support" he would have.) EP, you are trying to push this without consensus, this is not acceptable. Only possible result here is "no decision". Robert Ullmann 14:50, 13 March 2008 (UTC)

Um.... Connel did vote to support. If you don't see consensus, that's fine, but I do see consensus. Apparently, we are working with different meanings of consensus. --EncycloPetey 15:03, 13 March 2008 (UTC)

Um, maybe you should look up consensus? (Is a dictionary ;-) 6-5-2 is not "general agreement" and certainly not "lack of active opposition" (your attempts to dismiss the serious opposition notwithstanding). And note you are getting very serious pushback on your attempt to treat it as decided. The result above shows that the issue was not sorted out properly before the vote. There's a message here for you. I have a suggestion: start a vote on the "opposite" idea: putting homophones on * line(s) like we do with rhymes (and sometimes hyphenations), and see how that vote goes. You may find a real consensus very rapidly. Robert Ullmann 15:17, 13 March 2008 (UTC)

I won't do that because it would be hypocritical. People get yelled at for starting votes on proposals they themselves do not support. --EncycloPetey 15:37, 13 March 2008 (UTC)

Not to be argumentative, but although it's the current practice, I'm not sure the idea to use that exclusively would have consensus either, and since it's already in practice anyways, the vote isn't really worthwhile. DAVilla 04:20, 15 March 2008 (UTC)

• No consensus. Note: This topic may merit further discussion and a future related vote. —Ruakh_TALK 23:01, 13 March 2008 (UTC)