Wiktionary talk:Votes/pl-2014-06/Excluding romanizations by default

This is something supporters of the proposal would know. One claim made is that romanizations are not words. In diff, an editor says that " insofar as we hold that Sanskrit is not written in the Latin script, mahā is not a Sanskrit word." --Dan Polansky (talk) 06:57, 14 June 2014 (UTC)Reply

Ouch. The proposal is vague and the rationale is nonexistent. I can’t make sense of it. To accept it verbatim would ban entries for half of the loanwords in English. —Michael Z. 2014-06-16 22:31 z

@Mzajac If you think the current wording still affects loanwords, please participate in the thread below which seeks to address that very issue. As for the rest, see my comment in the BP. - -sche (discuss) 23:19, 16 June 2014 (UTC)Reply

WT:CFI does not forbid romanizations. There is no other policy I know of that forbids romanizations; policy drafts are countless, created on a whim without a discussion, and are not policies.

As for common practice, romanizations have a long tradition of being included. Examples are as follows:

• a3 was included as Chinese Pinyin since 22 August 2006
• bān was included as Chinese Pinyin since 11 September 2006
• bēi was included as Chinese Pinyin since 9 December 2006, created by a bot
• abaku was included as Japanese romaji since 5 June 2007
• amagu was included as Japanese romaji since 7 November 2005
• arigatō was included as Japanese romaji since 26 October 2005

--Dan Polansky (talk) 06:57, 14 June 2014 (UTC)Reply

IMO this vote should specify in its wording whether it means to override pl-2011-10/Romanization of Gothic and pl-2011-09/Romanization of languages in ancient scripts 2 or, on the contrary, it means for the languages listed there to be excluded from its purview.​—msh210℠ (talk) 20:46, 15 June 2014 (UTC)Reply

@msh210 It says "Romanizations can be allowed on a per-language basis when there is a consensus to do so." Is this insufficient? If you think it insufficient, I will make the vote more explicit. --Dan Polansky (talk) 17:23, 16 June 2014 (UTC)Reply

I've made the vote more explicit. Let me know if this is insufficient. --Dan Polansky (talk) 17:52, 16 June 2014 (UTC)Reply

Much better, thanks.​—msh210℠ (talk) 18:50, 16 June 2014 (UTC)Reply

I have proposed a merger between this vote and pl-2014-06/Allowing attested romanizations at ]; please comment there.​—msh210℠ (talk) 17:19, 16 June 2014 (UTC)Reply

I think the votes should not be merged for two reasons. One reason is that they are not logically complementary: one can oppose excluding romanizations by default and at the same time oppose the criteria specified at the other vote. Another reason is that having these nearly complementary votes increases the chance that one of them will pass, and is less biased toward implying what the status quo is as for policy and common practice. --Dan Polansky (talk) 17:28, 16 June 2014 (UTC)Reply

I can see a good way to do it as a set of alternatives. One would be this, which is basically exclude everything, the second would be to allow attested romanizations in accordance with the conditions of the other vote, a third would be to allow all romanizations (whether the romanized form was attested or not), which would be the outcome of the Sanskrit vote with respect to that language. In theory, there could be a fourth option, to impose tighter restrictions on romanizations in a way that justifies the votes on Chinese and Japanese, but does not allow others, but I have yet to see a principled distinction made for a blanket rule to be applicable. bd2412 T 17:34, 16 June 2014 (UTC)Reply

What I can imagine is having one vote titled "Default treatment of romanizations", "Treatment of romanizations", "Treatment of romanization entries" or the like. The vote would then make two proposals using the wordings currently available on on separate vote pages, and each of these proposals would have its own support and oppose section. Thus, there would still be two votes, albeit on one page. A third proposal could be added if someone comes up with one. However, I emphasize the need for simplicity and clarity, for the reason of which I would like to avoid preference voting on more than two proposals, and similar voting structures; each proposal should have its own support and oppose section; if more than one proposal passes (unlikely), another discussion or vote can be created to find the most preferred proposal. --Dan Polansky (talk) 18:40, 16 June 2014 (UTC)Reply

I don't understand what you speak of as a fourth option (what do you mean "justifies the votes on Chinese and Japanese, but does not allow others"?), but since it seems you're not suggesting including such an option in a 'merged vote', I suppose it doesn't matter. @Msh210, BD2412 If you could draft a merged vote for everyone to look at and critique and tweak, that would be helpful, I think. I've bumped the 'allow' vote back four days, so that it matches the 'exclude' vote's start date, and gives time for drafting of a merged vote. - -sche (discuss) 18:42, 16 June 2014 (UTC) pings added - -sche (discuss) 22:51, 17 June 2014 (UTC)Reply

Clearly, judo is a romanization. This word is clearly attested in English, French and other languages, but as the proposal is excluding romanizations even when attested, I understand that this vote proposes to delete the English, French... sections of the page judo. I cannot disagree more. Lmaltier (talk) 18:51, 16 June 2014 (UTC)Reply

• I think that's a different aspect of the discussion. We might call judo a loanword, in which case it can presumably be treated as if it were a native word in whatever language. It's really no different in that regard from ravioli. There is an open question about how we label transliterations if they are to be kept. Do we call them all English (or whatever the language is into which they are transliterated)? Do we call them their native language (as we now do with Chinese and Japanese)? Do we craft a divide based on how they are used in print? bd2412 T 19:52, 16 June 2014 (UTC)Reply

  This is not a different aspect. Clearly, judo is an English word, it's also a French word, etc. This is not disputable. But this vote proposes to exclude them nonetheless (ravioli is very different, and would not be excluded, as this loanword is not a romanization, but vodka or judo would be excluded because they are romanizations, despite the fact that they are very widely attested in English). My own opinion is that the only criterion should be that the word, written this way, is used (not only mentioned) in the language under consideration. Its etymology (romanization or not) is irrelevant. Lmaltier (talk) 20:15, 16 June 2014 (UTC)Reply

  I would describe a romanization as a presentation of a foreign word in the roman alphabet, while remaining a foreign (i.e. not English) word. bd2412 T 20:20, 16 June 2014 (UTC)Reply

  Of course, this is the idea of the proposal. But this is not explicit, this is why it should be clarified. And many English words, such as izba or Kremlin, are romanizations of foreign words. I think that the requirement of attestations (as uses) is sufficient. Lmaltier (talk) 20:40, 16 June 2014 (UTC)Reply

  I have edited the vote by adding "for a language usually not written in Latin" and adding your judo example; if there are more issues, please let us know. I fear this might have introduced another issue, but I do not know which one yet. --Dan Polansky (talk) 21:03, 16 June 2014 (UTC)Reply