Wiktionary:Votes/2022-01/New phrasebook regulations

Voting on:

 * 1) Replacing policy-TT with policy-VOTE in WT:PHRASE.
 * 2) Adding the following L2 section to WT:PHRASE between   and  :

Translations
All English phrasebook entries shall contain a  section. As usual, this section may optionally consist of only a soft-redirect using trans-see. The following applies to the entries where this is not the case:

If, for a given phrase, there exists at least one language in which the phrase can be expressed in a polite as well as a / normal register, two translation boxes, glossed as familiar / normal and polite respectively, shall be provided (in this order). Translations into languages that don't feature this distinction for this phrase shall be supplemented to the first box only. If, for a given phrase, no language makes this distinction, only one translation box without a gloss, i.e. trans-top, shall be provided. For phrases that are inherently nonpolite or for ones that are almost exclusively uttered in polite contexts, only one translation box, familiar / normal or polite / normal respectively shall be provided.

Plural forms must not be provided as translations, unless this is inherently necessary due to the nature of the phrase (for example, ), or editors have collaboratively decided to include plural translations for an individual phrase. If, for a certain phrase, both singular as well as plural forms are to be documented, the translation boxes as prescribed by the above paragraph shall be duplicated with singular and plural prepended to the gloss respectively.

For entries with multiple senses, the above rules apply on a per-sense basis with a short gloss of the sense prepended to the prescribed translation box title.

If a translation of a phrase depends on the natural gender of one person, the corresponding gender must be denoted using the designated argument of t. If a translation depends on the genders of multiple persons (as may be the case for certain languages with polypersonal agreement), the corresponding language's editor community should decide how to best document this. For languages that feature an orthogonal formal / informal distinction in addition to the familiar / polite distinction, the treatment of those forms is also to be decided upon by those languages' editor communities.

Rationale:

 * 1) The quality of translations provided in our phrasebook entries varies greatly. Entries like 'are you allergic to any medications' feature many polite/formal translations with no corresponding familiar form. Implementing this hard split not only allows interested users to add missing t-needed for the registers, such requests could also be machine-generated. This is currently not possible due to the inconsistent register qualifiers.
 * 2) Plural translations are but bloat in many phrasebook entries. In some, they may even be worse than bloat by being actively confusing, e.g. in do you have a menu in English: The context where this phrase is uttered is clear, yet the Swedish translation leaves open the choice between singular and plural. So which one would I have to use if I sat in a Swedish restaurant?
 * 3) In cases where both a plural as well as a singular translation would make sense, the proposed system is again superior by allowing t-needed if either form is missing.
 * 4) It also enforces a correct classification: The translations are currently by far not all qualified as polite/familiar singular/plural; this qualification would happen implicitly if we gloss the translation boxes as suggested.
 * 5) For phrases where, say, two wordings are common in the target language, the proposal would again make recognizing which combination (e.g. "wording B & polite") is missing a lot easier.
 * 6) Having the same qualifiers dozens of times in a translation box (as in are you married but also many others) is inherently bloaty and better solved by instead categorizing a priori according to that characteristic.
 * 7) Using policy-VOTE is proposed because non-superficial changes to the phrasebook policy should require community consensus anyway as we have to update hundreds of pages in accordance; hence also this vote.

Schedule:

 * Vote starts: 00:00, 17 January 2022 (UTC)
 * Vote ends: 23:59, 14 February 2022 (UTC)
 * Vote created: Fytcha (talk) 21:23, 3 January 2022 (UTC)

Discussion:

 * [[Image:Wikt rei-artur3.svg|20px]]
 * [[Image:Wikt rei-artur3.svg|20px]]

Support

 * 1)  &mdash; Fytcha〈 T | L | C 〉 23:59, 16 January 2022 (UTC)
 * 2)  --Numberguy6 (talk) 03:39, 17 January 2022 (UTC)
 * , looks sound enough, although I expected this vote to contain some criterias for non-idiomatic phrases, commonality basis (i.e. number of usages), which kind of non-idiomatic set phrases are entry-worthy, etc. rather than it being based only on translations -- not complaining, just this was kind of unexpected (to me). —Svārtava [t•c•u•r] 04:18, 17 January 2022 (UTC)
 * 1)  I'd ideally like a provision for other possible splits, some of which would be peculiar to individual phrases, but this is a good starting point. Hazarasp (parlement · werkis) 11:46, 17 January 2022 (UTC)
 * 2)  This, that and the other (talk) 03:35, 18 January 2022 (UTC)
 * 3) . Makes sense, and none of the objections I've seen convince me. Imetsia (talk) 17:38, 20 January 2022 (UTC)
 * 4) . AG202 (talk) 02:58, 26 January 2022 (UTC)
 * 5)  Vininn126 (talk) 16:17, 26 January 2022 (UTC)
 * 6)  ←₰-→  Lingo Bingo Dingo (talk)  09:55, 10 February 2022 (UTC)
 * 7)  &mdash; seems reasonable - John Cross (talk) 09:45, 12 February 2022 (UTC)

Decision

 * Passed 10-0-0 &mdash; S URJECTION / T / C / L / 09:58, 15 February 2022 (UTC)