Wiktionary:Votes/2007-06/Level of basic headings

• Voting on: With the language at level 2, such as ==English==, what levels should be assigned to Etymology, Pronunciation, and the part of speech headings (Noun, Prefix, Initialism, etc.) for definitions? Note that there are special cases with several etymologies and pronunciations.

• Notice: Despite the attempt to completely represent all possible options, the outcome of this vote will not be considered to make any formal statement on policy due to oversight on options 5 and 6. However, it will be continued as a straw poll until the end date so that other issues can be worked out, possibly eliminating some of the options that clearly do not have community support. A new vote, based mainly on the results of this one, including comments, will be submitted at that time. DAVilla 05:47, 18 August 2007 (UTC)

• Vote ends: 15 16 September 2007 23:59 UTC Withdrawn
• Vote starts: 16 August 2007 23:59 UTC

• Vote created: DAVilla 17:22, 29 June 2007 (UTC)
• Discussion:

  Wiktionary:Beer parlour#ELE level 4 header sequence

  also some extended comments on the talk page

• Previous vote:

  Wiktionary:Votes/2007-06/Order of Etymology and Pronunciation

• Instructions:

  This vote is structured as a variation of approval voting. For each option, please vote in support or opposition. You may vote in favor of as many options as you wish. Abstention is implicit, but comments are always welcome.

• Always as ===Etymology=== (3) and ===Pronunciation=== (3), or ===Pronunciation=== (3) and ===Etymology=== (3), followed by ===POS=== (3).

1. Support DAVilla 06:20, 16 August 2007 (UTC)

1. Oppose Thryduulf 09:30, 16 August 2007 (UTC). Does not allow for clear separation of words with different etymologies or different pronunciations.

   Sure it does. The relevant etymology is the etymology section under which the POS heading falls, and the relevant pronunciation is the pronunciation section under which the POS heading falls. See example 1 and the very similar example 1.1. DAVilla 12:49, 16 August 2007 (UTC)

   No, it doesn't, because the POS section doesn't "fall" under any pron or ety; it is using the sequence and not the structure; if any pron or ety is missing or added, the POS may (probably will) end up with the wrong implied attributes Robert Ullmann 13:22, 16 August 2007 (UTC)

   Yes, that's true if there are at least two possible pronunciations and two possible etymologies, where one of each is not provided. That's possible with as few as two definitions. Give one definition an etymology that doesn't apply to the other, and the second a pronunciation but no etymology. You can't switch the order because the pronunciation doesn't apply to the first. So you'd be stuck, apart from creating a stub section. That's what I would do if I couldn't enter the pronunciation, and frankly if I the definition I was adding had a different etymology then I wouldn't even know that in the first place.

   More practically, in cases where there is more than one pronunciation, usually all pronunciations are given anyway, if any are. In cases where there is more than one etymology, it takes someone who knows something about etymology to even recognize that fact. Treating Pronunciation as subordinate to Etymology, it's easy to place definitions with unknown etymology above any Etymology section, which is what you would do if you were careful. A user careless with this is probably one who would be careless anyway, regardless of which option we choose. DAVilla 13:52, 16 August 2007 (UTC)

2. Oppose Rod (A. Smith) 18:46, 16 August 2007 (UTC) This structure would eliminate the valuable hierarchy that heading levels provides. Sequence would be a confusing substitute for hierarchy. Rod (A. Smith) 18:46, 16 August 2007 (UTC)
3. Oppose —Stephen 14:13, 17 August 2007 (UTC)
4. Oppose EncycloPetey 20:49, 17 August 2007 (UTC)
5. Oppose ArielGlenn 02:46, 19 August 2007 (UTC)

• Always as ===Etymology=== (3) and ===Pronunciation=== (3), or ===Pronunciation=== (3) and ===Etymology=== (3), followed by ====POS==== (4).

1. Support DAVilla 06:20, 16 August 2007 (UTC)
2. Support Rod (A. Smith) 18:46, 16 August 2007 (UTC) This structure makes good use of headling level hierarchy. Rod (A. Smith) 18:46, 16 August 2007 (UTC)
3. Support —Stephen 14:12, 17 August 2007 (UTC)

1. Oppose Thryduulf 09:31, 16 August 2007 (UTC) "Always as". There is no need to make the POS a lower level in cases that don't need it.

   One good reason is consistency. DAVilla 04:36, 19 August 2007 (UTC)

   How about consistency with the 90% of entries in which pron and ety don't appear, and the POS must be L3? Robert Ullmann 15:30, 19 August 2007 (UTC)

2. Oppose EncycloPetey 20:49, 17 August 2007 (UTC)
3. Oppose ArielGlenn 02:46, 19 August 2007 (UTC)

Comment: This might make more sense if everything were pushed down, with language at level 1. DAVilla 17:22, 29 June 2007 (UTC)

Comment: It would also be possible to keep subsequent headers at level 4, instead of pushing them up to level 5. In that case the only change would be POS going from 3 to 4, and in complex cases Etymology and Pronunciation are kept at 3. DAVilla 04:38, 19 August 2007 (UTC)

• Always as ===Etymology=== (3) and ====Pronunciation==== (4), or ===Pronunciation=== (3) and ====Etymology==== (4), followed by ====POS==== (4).

1. Support DAVilla 06:20, 16 August 2007 (UTC)
2. Support —Stephen 14:04, 17 August 2007 (UTC)

1. Oppose Thryduulf 09:30, 16 August 2007 (UTC) Etymology and pronunciation should be at the same level when there is no need to distinguish between multiples of either. Thryduulf 09:30, 16 August 2007 (UTC)

   That's equivalent to saying that you should usually have Etymology or Pronunciation at 3 but sometimes at 4, which is to say that your premise is identical to your conclusion. There's nothing wrong with that if you view them equally, but I rather dislike having them shift around like that. DAVilla 13:01, 16 August 2007 (UTC)

2. Oppose Rod (A. Smith) 18:46, 16 August 2007 (UTC) Pronunciation varies over time and distance, so it seems convoluted and misleading to make etymology subordinate to it. Rod (A. Smith) 18:46, 16 August 2007 (UTC)

   A previous vote weakly supported that arrangement, so all options here were written to allow it. DAVilla 20:04, 20 August 2007 (UTC)

   That vote said, “...Etymology should be first followed by Pronuncation, but Pronunciation can be first in certain complex cases if it simplifies the presentation.” Option 3 above, however, appears ambivalent about the order. Rod (A. Smith) 20:56, 20 August 2007 (UTC)

3. Oppose EncycloPetey 20:49, 17 August 2007 (UTC)
4. Oppose ArielGlenn 02:46, 19 August 2007 (UTC)

Comment: This might make more sense if everything were pushed down, with language at level 1. DAVilla 01:24, 8 July 2007 (UTC)

• Always as ===Etymology=== (3) and ====Pronunciation==== (4), or ===Pronunciation=== (3) and ====Etymology==== (4), followed by =====POS===== (5).

1. Qualified support provided all heading are pushed down: =Language= (1), ==Etymology== (2) and ===Pronunciation=== (3) or vice versa, ====POS==== (4) DAVilla 06:22, 16 August 2007 (UTC)

1. Oppose Thryduulf 09:32, 16 August 2007 (UTC) Strong oppose "always as", this is too hierarchical in most cases; this should be an option though there there are multiple pronunciaions and multiple etymologies for the same entry. I also strongly oppose moving language to level 1.

   I would like to know what is wrong with having anything at level 1. Do we anticipate levels higher than language?

   FYI this option could not pass on qualified support alone until such a resolution, if necessary, were supported independently of it. DAVilla 13:04, 16 August 2007 (UTC)

   The level 1 heading is the page heading, i.e. the word being defined. The only level one heading on the "compost" entry should be "compost". Thryduulf 13:57, 16 August 2007 (UTC)

   Ah, I see. Well I'm sure we could find a work-around to that, if it were really desired, maybe not for this option but really regardless of whatever passes. I mean, the page title is hard-coded anyways, and there is nothing more consistent or easier to spot-fix, except that it requires developer intervetion. But why reserve heading 1 for that? It's a complete waste of a heading, and an annoyance knowing that not only is =This= rarely if ever a possibility, but that there is an extra pair of equal signs wrapping each and every heading. DAVilla 14:44, 16 August 2007 (UTC)

2. Oppose Rod (A. Smith) 18:46, 16 August 2007 (UTC) Pronunciation varies over time and distance, so it seems convoluted, misleading, and error-prone to make etymology subordinate to it. (When a new pronunciation is found for one etymology but not another, we'd have to restructure the entry.) Rod (A. Smith) 18:46, 16 August 2007 (UTC)
3. Oppose EncycloPetey 20:49, 17 August 2007 (UTC)
4. Oppose ArielGlenn 02:46, 19 August 2007 (UTC)

Comment: This might make more sense if everything were pushed down, with language at level 1. DAVilla 06:23, 16 August 2007 (UTC)

• Simple cases as ===Etymology=== (3), ===Pronunciation=== (3), and ===POS=== (3), but allow ===Etymology=== (3) and ====Pronunciation==== (4) or ===Pronunciation=== (3) and ====Etymology==== (4) in complex cases, pushing up subsequent headings to ====POS==== (4).

1. Support Thryduulf 09:30, 16 August 2007 (UTC) strong support, as this is flexible enough to cope with just about everything. I'd also allow option 4's structure in cases that need it (i.e. multiple pronuciations and etymologies).
2. Support —Stephen 14:06, 17 August 2007 (UTC)
3. Support EncycloPetey 20:49, 17 August 2007 (UTC) - This option (and option 6) come closest to what I would like to see used, but neither one excatly matches my expectations. Given the number of options, I expect no single option to win a clear majority.

   It's really early, but right now this option and option 2 have a simple majority. To get a supermajority, meaning twice as many supporting votes to opposing, people might want to change their boarderline votes later on to support (or less weakly oppose) what they see as a compromise option. DAVilla 05:20, 18 August 2007 (UTC)

   Um.. options 2 and 5 are the only options right now with a net positive vote (+1). That's hardly ringing endorsement. --EncycloPetey 05:25, 18 August 2007 (UTC)

4. Support This and 6 are closest match even if neither of them are quite to my liking. ArielGlenn 02:41, 19 August 2007 (UTC)
5. Support Williamsayers79 17:25, 20 August 2007 (UTC)

   (Moved this up from the bottom of the section. DAVilla 21:47, 20 August 2007 (UTC))

1. Oppose DAVilla 06:20, 16 August 2007 (UTC)

   Oppose Rod (A. Smith) 18:46, 16 August 2007 (UTC) Pronunciation varies over time and distance, so it seems convoluted, misleading, and error-prone to make etymology subordinate to it. (When a new pronunciation is found for one etymology but not another, we'd have to restructure the entry.) Rod (A. Smith) 18:46, 16 August 2007 (UTC)

   Striking my opposition in light of this previous vote. Rod (A. Smith) 20:56, 20 August 2007 (UTC)

Comment: I believe this is status quo. DAVilla 18:01, 11 July 2007 (UTC)

This is the status quo except that pron-3 and ety-4 isn't documented or described anywhere, and the entries that do it push POS to level 5. (fixed typo) None of these options address the difficult case, where the sets of prons and etys overlap, but don't nest. (POS-a has pron 1/ety 1, POS-b has pron 1/ety 2, pos-c has pron 2/ety 2) Trying to vote without a coherent single proposal that defines all cases and covers all cases is just going to leave the problem unresolved. Robert Ullmann 13:03, 16 August 2007 (UTC)

Added new option below per this comment, and restarted vote.

See example 5 and example 5.1. DAVilla 13:10, 16 August 2007 (UTC)

Then it should say explicitly that when etys and prons overlap, but don't nest, that the prons are repeated at L4, while the ety is at L3. The other organization (pron at 3, ety at 4) is not the status quo; the status quo is using pron once at L3 at the top when it applies to all etys, and ety always at L3.

I think you should come up with ONE proposal that is coherent and well defined, and have a vote. Robert Ullmann 12:29, 17 August 2007 (UTC)

Crap, I didn't say that ===Pronunciation=== (3), ===Etymology=== (3) was also possible for this and the next option? Just shoot me already. DAVilla 05:37, 18 August 2007 (UTC)

• Simple cases as ===Etymology=== (3), ===Pronunciation=== (3), and ===POS=== (3), but allow ===Etymology=== (3) and ====Pronunciation==== (4) or ===Pronunciation=== (3) and ====Etymology==== (4) in complex cases, pushing up subsequent headings to =====POS===== (5).

# — This unsigned comment was added by Thryduulf (talk • contribs).

I was debating whether my comment below was a support or oppose, decided it was neither in the end so vote above struck. Thryduulf 16:55, 16 August 2007 (UTC)

1. Support EncycloPetey 20:49, 17 August 2007 (UTC) - This option (and option 5) come closest to what I would like to see used, but neither one excatly matches my expectations. Given the number of options, I expect no single option to win a clear majority.
2. Support This and 5 are closest match, even if neither are quite to my liking. ArielGlenn 02:43, 19 August 2007 (UTC)

1. Oppose DAVilla 13:35, 16 August 2007 (UTC)

   Oppose Rod (A. Smith) 18:46, 16 August 2007 (UTC) Pronunciation varies over time and distance, so it seems convoluted, misleading, and error-prone to make etymology subordinate to it. (When a new pronunciation is found for one etymology but not another, we'd have to restructure the entry.) Rod (A. Smith) 18:46, 16 August 2007 (UTC)

   Striking this oppose vote in light of previous vote. Rod (A. Smith) 20:56, 20 August 2007 (UTC)

Added per the comments above. Vote is restarted. DAVilla 13:10, 16 August 2007 (UTC)

This should be the standard used only for entries that have multiple pronunciation and multiple etymologies as part of a flexible system such as Option 5. Thryduulf 14:05, 16 August 2007 (UTC)

That might be possible to take into consideration if there is a close race between the two. Or would you like me to add it as another option? (We already have so many, though!) DAVilla 15:01, 16 August 2007 (UTC)

I think another option would make this even more confusing than this already is, and as I do not forsee this being a common scenario I think we just leave it for now. Thryduulf 16:55, 16 August 2007 (UTC)

Removing weird comment generated by.. ?? Not by my typing, in any case. Sorry! ArielGlenn 05:00, 19 August 2007 (UTC)

• Simple cases as ===Etymology=== (3), ===Pronunciation=== (3), and ====POS==== (4), but allow ===Etymology=== (3) and ====Pronunciation==== (4) or ===Pronunciation=== (3) and ====Etymology==== (4) in complex cases, followed by ====POS==== (4).

1. Support DAVilla 06:20, 16 August 2007 (UTC)
2. Support —Stephen 14:09, 17 August 2007 (UTC)

1. Oppose Thryduulf 09:30, 16 August 2007 (UTC) there is no need to make the POS a lower level heading than Etymology and/or Pronunciation unless there is a reason to do so. I also strongly oppose moving language to level 1, etc.

   Language doesn't have to be at 1 in this example. Without it we would, essentially, be adding a level to Parts of speech. Note that this vote does not address headings that are dependent on Part of speech, such as Synonyms and Translations, which do not necessarily need to be pushed down!

   With language at 2, Etymology at 3, and POS at 4, then headings like synonyms and translations must be at 5. In complex cases this would mean language 2, etymology 3, pronunciation 4, POS 5, Translations 6 and ttbc 7. This is too many levels in most cases, only when there are multiple etymologies and multiple pronunciations that need distinguishing should we have this. I've commented elsewhere about Language at L1. Thryduulf 14:10, 16 August 2007 (UTC)

   Yes, that would be too many levels, but that many are not necessary (and not just because of the strange extra TTBC heading, which isn't required by the way). As I said above, Synonyms and Translations do not have to be pushed down to level 5. If we wanted, they could remain at level 4, since there would be no ambiguity as to which POS they fell under. DAVilla 14:50, 16 August 2007 (UTC)

   Wait, did you mean your statement to apply to Option 6? Complex cases here have POS at 4. DAVilla 14:53, 16 August 2007 (UTC)

   Oppose Rod (A. Smith) 18:46, 16 August 2007 (UTC) Pronunciation varies over time and distance, so it seems convoluted, misleading, and error-prone to make etymology subordinate to it. (When a new pronunciation is found for one etymology but not another, we'd have to restructure the entry.) Rod (A. Smith) 18:46, 16 August 2007 (UTC)

   Per above, a previous vote weakly supported that arrangement, so all options here were written to allow it. DAVilla 20:10, 20 August 2007 (UTC)

   Fair enough. Opposition withdrawn. Rod (A. Smith) 20:56, 20 August 2007 (UTC)

2. Oppose EncycloPetey 20:49, 17 August 2007 (UTC)
3. Oppose ArielGlenn 02:55, 19 August 2007 (UTC)

Comment: This is a hybrid of options 2 and 3. It might make more sense if everything were pushed down, with language at level 1. DAVilla 18:01, 11 July 2007 (UTC)

1. Abstain —Ruakh_TALK 03:07, 19 August 2007 (UTC)
2. Abstain Widsith 15:57, 19 August 2007 (UTC)
3. Abstain Jeffqyzt 13:58, 20 August 2007 (UTC) Or, consider this an oppose to all of the above. See the talk page for my understanding of the status quo, which is what I would like to see as at least one of the options (none of the above seem to match it, IMO.) --Jeffqyzt 13:58, 20 August 2007 (UTC)
4. Abstain Connel MacKenzie 15:41, 22 August 2007 (UTC) In the numerous rounds of attempting to dictate heading order unambiguously, it has become clearer and clearer (to me) that it cannot be done, without radical layout simplification. Without an overhaul to group all definitions together (broken down by POS) then disambiguated in each subsequent section (like translations currently are) any attempts at specifying orders of the non-definition headings are rather pointless. My personal views aside, this straw poll doesn't even say what it hopes to accomplish. --Connel MacKenzie 15:41, 22 August 2007 (UTC)

• Withdrawn DAVilla 00:13, 23 August 2007 (UTC)