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

Excluding romanizations by default

 * Voting on: By default, a romanization for a language usually not written in Latin shall be excluded as a separate entry, no matter whether the romanization is attested as per WT:CFI or not. Romanizations for such a language can be allowed on a per-language basis when there is a consensus to do so, as has happened for example via Votes/pl-2011-09/Romanization of languages in ancient scripts 2, which pertains to Etruscan, Gothic, Lydian, Oscan, Phoenician and Punic. For example, the exclusion does not apply to English judo, since English is usually written in Latin.
 * Rationale: For a rationale, see Wiktionary talk:Votes/pl-2014-06/Excluding romanizations by default. The voters only vote on the proposed action, not on the rationale.


 * Vote starts: 00:01, 21 June 2014 (UTC)
 * Vote ends: 23:59, 20 July 2014 (UTC)


 * Vote created: Dan Polansky (talk) 06:52, 14 June 2014 (UTC)
 * Discussion:
 * [[Image:Wikt rei-artur3.svg|20px]] Votes/pl-2009-12/Treatment_of_toneless_pinyin_syllables
 * [[Image:Wikt rei-artur3.svg|20px]] Beer parlour/2014/June
 * [[Image:Wikt rei-artur3.svg|20px]] Votes/pl-2014-06/Allowing attested romanizations
 * [[Image:Wikt rei-artur3.svg|20px]] Wiktionary talk:Votes/pl-2014-06/Excluding romanizations by default

Support

 * 1) . — Ungoliant (falai) 16:21, 23 June 2014 (UTC)
 * 2) . The policy on allowing pinyin, Jyutping and Japanese romaji should also be reviewed. This dictionary should not have any romanisation entries. Search by transliteration is the function that should be developed instead before this chaos of writing everything foreign in the Latin script gets out of hand. Wyang (talk) 23:47, 23 June 2014 (UTC)
 * 3) * I agree with Wyang that we should build this into search. We don't, for example, include transliterated versions of English words into other scripts (AFAIK) and this should be no different. Equinox ◑ 19:31, 24 June 2014 (UTC)
 * 4) * It's a nice sentiment, but shouldn't we first see if such a function can be enabled, before voting based on it potentially existing? Also, what do we do where the romanization of one word is the same as an existing definition for another word in a Latin script (like wang)? bd2412 T 19:45, 24 June 2014 (UTC)
 * 5) * Requirements should drive development, not vice versa. The search has definitely been improved since I first arrived here. Equinox ◑ 19:51, 24 June 2014 (UTC)
 * 6)  --Anatoli (обсудить/вклад) 06:14, 24 June 2014 (UTC)
 * 7)  --Dijan (talk) 06:39, 24 June 2014 (UTC)
 * 8) Hell yes! Allowing romanizations was a big mistake from the very beginning. -- Liliana • 19:20, 24 June 2014 (UTC)
 * Why? I've gotta say: I'm thoroughly unconvinced by the arguments put forth by the supporters of romanization exclusion. Purplebackpack89  (Notes Taken) (Locker) 19:25, 24 June 2014 (UTC)

Oppose

 * 1)  Dan Polansky (talk) 19:54, 23 June 2014 (UTC)  I see no good reason for excluding attested romanizations, that is, attested per WT:CFI, which includes the requirement that they convey meaning; for instance, romanization listed next to a non-romanized form in a dictionary does not convey meaning. We are not running out of digital storage; and these romanizations are not difficult to maintain. If we accept the argument that searching for romanizations finds the target native-script page and thus the romanization pages are unneeded (attested or not), we would exclude dedicated pages for rare alternative forms, place them only to Alternative forms section of the most common form, and let the reader use search to find that. A further consequence of acceptance of the search-argument would be the exclusion of inflected forms: the search function could find them as long as they are present in the lemma entry, although this would not work when the lemma entry uses an auto-inflection template. Generally, the search-argument would lead to exclusion of Chinese and Japanese romanizations as well, which we have decided to include; the distinction between difficult scripts and easy scripts for the purpose of deciding which romanizations to include would no longer apply, since the search function would find the target entry regardless of whether the script is easy or difficult. The claim that "romanizations are not words" made by some of the supporters is IHMO obviously incorrect. --Dan Polansky (talk) 19:54, 23 June 2014 (UTC)
 * 2)  - I didn't realize the vote was starting now. I would have combined the two proposals at issue. We are liable to end up with inconsistent results this way. In any case, there is no good reason to have a blanket exclusion of any class of attested words. Wyang makes an interesting point about having a search by transliteration function, but since we have no such function, it's a moot point. Even a vote to have such a function won't make it magically appear, absent developer intervention, and even that will not help for transliterated terms that coincide with terms that are actual words in other Latin-script languages. bd2412 T 23:58, 23 June 2014 (UTC)
 * The results will not really be inconsistent. My plan right now is not to touch Votes/pl-2014-06/Allowing attested romanizations, since I think the vote should have never been created. The present vote (Votes/pl-2014-06/Excluding romanizations by default) should be perfectly sufficient on its own, hopefully (but let us see) producing enough evidence that there is no consensual support for what it is proposing. --Dan Polansky (talk) 18:47, 24 June 2014 (UTC)
 * Meanwhile, Votes/pl-2014-06/Allowing attested romanizations is so far showing that there's no consensus that attested romanizations are to be allowed. And although we may agree that in the absence of consensus to change the status quo, the status quo remains, we disagree on what the status quo is, which means this whole affair is looking unfortunately similar to a caucus race. I hope the situation of one or the other vote changes and one of them passes; otherwise, it may just end up that a month from now, you or BD creates some romanization entries according to your interpretation of things, and Anatoli or Ungoliant or Wyang or Renard or I tags them // according to existing practice and our interpretation of things, and then RFD threads ensue. - -sche (discuss) 19:10, 24 June 2014 (UTC)
 * I think the evidence as for status quo before the votes is overwhelming; you have posted close to no evidence supporting your claims as to it. The consensus (or its lack) will be easily extracted by looking at the posts at both votes. Once the consensus or its lack will be clear, your actions not supported by consensus and contradicting existing written policies (WT:CFI) will be likely undone by admins who will then know what the consensus or its lack are. It is not a matter of "interpretation". I do know from experience that when Wiktionary policies are not going to be enough, you will invent them on the spot, or imply that Wikipedia policies apply to Wiktionary, as you did at User_talk:Dan_Polansky/2013. I certainly do not trust you to report existence of policies and your administration of these policies. --Dan Polansky (talk) 19:18, 24 June 2014 (UTC)
 * 1)  - This will only make it more difficult for readers to look up words in languages that do not use Latin script. For example, someone trying to look up a Japanese term who cannot read any of the Japanese scripts, or who does not have a keyboard capable of typing in the Japanese scripts. -Cloudcuckoolander (talk) 01:00, 24 June 2014 (UTC)
 * 2)  Seems to unfairly limit the scope of entries that could be created.  If it can be attested and people might look for it, we should have it.  Purplebackpack89  (Notes Taken) (Locker) 04:53, 24 June 2014 (UTC)
 * 3)   &mdash; Saltmarsh απάντηση 04:55, 24 June 2014 (UTC)
 * 4)  I can't accept "usually not written in Latin script" as too high a standard for when we should start recording actual usage, and it's clear that even when a language is written virtually exclusively in the Latin script, like Gothic, some people will choose to interpret that as romanization.--Prosfilaes (talk) 20:03, 24 June 2014 (UTC)
 * To clarify, do you oppose excluding attested romanizations for Russian, meaning entries in Latin script having Russian L2-heading? Or do you actually support such an exclusion of Russian romanizations? Furthermore, Gothic is already regulated by another vote, and this vote says IMHO quite clearly that it only states the default, and expressly references Votes/pl-2011-09/Romanization of languages in ancient scripts 2 for Gothic. --Dan Polansky (talk) 20:46, 24 June 2014 (UTC)
 * I'm not sure I understand the question exactly. Given a string of letters with an established meaning, a word I would say if that didn't seem so controversial in this discussion, I think we should have an entry for it, or at least an entry trivially findable for our users for it (i.e. dropping Latin macrons and Esperanto diacritic normalization and all that). Gothic is just an example; as "Don't Proliferate; Transliterate!" says (sections 1 and 5, unless you're really interested in historic Greek) the Latin script is used for many languages, some dead before Rome was founded, by the people who actually use them in the real world, and encoding their scripts in Unicode is (mostly was) a hobbyist game instead of something interesting to scholars. I don't think the list of languages for which that is true is particularly bounded.--Prosfilaes (talk) 21:03, 24 June 2014 (UTC)

Abstain

 * 1) [[Image:Symbol abstain vote.svg|20px]] Abstain. Romanizations are already excluded by default, as anyone who's seen one created learns — the entry is tagged  and/or moved to the native script form by someone knowledgeable, unless a vote has allowed the language in question to have romanizations. (In particular, if a user were to create entries of the sort proposed in the BP — for ad-hoc romanizations, "cited" with "citations" like [//en.wiktionary.org/w/index.php?title=User:-sche/svobodnyx&oldid=27198972 these] — I think even some of the people who favor including romanizations would laugh the user out of RFD.) An effort to change this practice and allow romanizations is underway here: Votes/pl-2014-06/Allowing attested romanizations. If that vote does not pass, is it a good idea to codify the exclusion of romanizations in the way this vote proposes, or would the wording of this vote have some unintended side effects? It's hard to say. I don't think the concern raised on the talk page (that this vote would ban ) holds any water. (Judo is a loanword, handled under ==English== ===Noun=== headers. Perhaps the users were thinking of, the actual romanization entry, where one can see the ==Japanese== ===Romanization=== headers, and note that they were approved by vote?) Other concerns may be more valid, however. - -sche (discuss) 20:04, 23 June 2014 (UTC)
 * This link shows where template:wrongscript is used. Evidence that the template is being used to delete entries without the proper RFV venue is absent. "as anyone who's seen one created learns ..." is not evidence, since it is not something I can inspect; it is merely something I can try to do, thereby leading to a conflict with the editors who support the proposal of the vote. So again, the claim that "romanizations are already excluded by default" lacks supporting evidence, anything that I can look at right now without engaging in edit war with editors supporting this vote. By contrast, I have posted evidence that romanizations were being entered without being immediately deleted and that without a vote granting them an exception at Wiktionary_talk:Votes/pl-2014-06/Excluding_romanizations_by_default--I have posted links to items that you can inspect such as amagu; many more such items can be found by inspecting Category:Japanese romaji. --Dan Polansky (talk) 20:16, 23 June 2014 (UTC)
 * 1) Seems unnecessary as since romanizations aren't words they don't meet WT:CFI anyway. Renard Migrant (talk) 16:55, 13 July 2014 (UTC)
 * 2) Arguing "since a is true then b" makes a vote irrelevant is a bit obnoxious when the truth or falsehood of a is a key issue under argument. I continue to find the assertion that collections of letters that convey meaning from one person to a general audience aren't words because they're in the Latin script (which matters in certain cases but not others) to be bizarre. I think that's been argued out enough; let it stand that what you claim as "since" is in fact a matter of disagreement that was part of what these two polls were hoping to settle.--Prosfilaes (talk) 22:38, 13 July 2014 (UTC)

Decision
Fails 5–6 (45%). Thus, the status quo is maintained with no change in policy effected; see the text above for discussion about what the status quo is. See also Votes/pl-2014-06/Allowing attested romanizations. &#x200b;—msh210℠ (talk) 18:28, 23 July 2014 (UTC)