Template talk:cu-noun

Looks good, but it should have support for Glagolitic spellings as well! Perhaps in the manner of {{sh-noun}} for Latin/Cyrillic. --Ivan Štambuk (talk) 18:21, 14 July 2013 (UTC)Reply

I'm not finished yet... I'm also not sure if we need to duplicate all the definitions for both scripts like we do with SC. —CodeCat 18:25, 14 July 2013 (UTC)Reply

Yes we do. I'm sorry but this is unacceptable. Both scripts are equally valid and none deserves to be a mere "alternative form" of another. If anything, Glagolitic is older and most OCS Cyrillic manuscripts were copied from Glagolitic originals. Furthermore, there is no exact 1-1 correspondence between them. Each spelling with its quirks caries important linguistic and paleographic evidence. --Ivan Štambuk (talk) 18:32, 14 July 2013 (UTC)Reply

Why is that unacceptable? Cyrillic is used far more widely than Glagolitic especially in the study of OCS. In modern times Cyrillic is clearly the primary script for OCS. This is not a matter of alternative forms being any less than the primary form. It's just a matter of efficiency as the Glagolitic forms will not be used as much. With SC that is very different because both of its scripts are still widely used. Even for SC though I think that Cyrillic should redirect to Latin script, for efficiency's sake so that duplication and inconsistencies are avoided. —CodeCat 18:41, 14 July 2013 (UTC)Reply

Orthodox Slavic countries provide Cyrillic transcriptions of Glagolitic OCS manuscripts because Cyrillic is their native script, and Glagolitic is some obscure nonsense that died out centuries ago. That is not necessarily the case in non-Orthodox countries (e.g. Croatia where Glagolitic is even taught in high schools). Modern-day grammars usually use Latin transcriptions for practicality, but that is immaterial to this discussion - what matters is how these forms are attested. OCS represents an extinct language, nobody writes in it today. Redirecting variant spellings within the same script (with more or less irrelevant typographical distinctions, like the different symbols for /i/, vocalized jers and similar) to a normalized spelling is justified, but redirecting to a different scripts is not. We haven't done so for any other language, and I don't see why we should start with OCS. More than half of OCS canon is attested in Glagolitic and is unacceptable to downgrade Glagolitic script it to a secondary status. OCS vocabulary is only a few thousand words even it's not much to duplicate anyway.

Your comment on Serbo-Croatians demonstrates that you're not weighing in enough cultural considerations in these matters. Sometimes it is necessary to spend extra effort not to hurt somebody's feelings. Specifically regarding Serbo-Croatian, most editors are using a custom tool that automatically mirrors Latin entries to Cyrillic so all the effort it takes is simply copy/pasting. Serbian Wikipedia has a special PHP plugin that generates Cyrillic versions from Latin-script articles - perhaps Wiktionary one day could get the a similar plugin that would simplify maintenance in cases such as this. Note that I have no problem at all with you not duplicating Cyrillic entries to Glagolitic, but I'm against you removing such expanded Glagolitic entries.

Anyway, if you're unhappy with my suggestion we can involve more editors to see how the community "feels" about this. --Ivan Štambuk (talk) 19:21, 14 July 2013 (UTC)Reply

I would like to know what others think of this, yes. If we really placed people's feelings first, we wouldn't have merged SC in the first place. On a dictionary, practical and linguistic matters should come first. —CodeCat 19:24, 14 July 2013 (UTC)Reply

Merging B/C/S/M entries as one is one thing, but prioritizing a specific script and e.g. Ijekavian/Ekavian variant (these also cause duplication of content) is an entirely different thing. By doing the former we "hurt" everyone the same, but by doing the latter we hurt a specific subset of speakers. I can assure you that if Wiktionary normalized Serbo-Croatian entries on e.g. Ijekavian Latin (which is spoken/written by most of the speakers) Serbian nationalists would be all up in arms on how somebody has declared a war on their identity. If we can have both colour and color in English with their massive duplication of content, we can have both Cyrillic and Latin Serbo-Croatian. --Ivan Štambuk (talk) 19:37, 14 July 2013 (UTC)Reply

And as you well know, having both spellings in English with full duplication is not without controversy as well. Furthermore, we already have many entries where one English spelling does point to the other. —CodeCat 19:39, 14 July 2013 (UTC)Reply

Duplication for different spellings/script is justifiable in cases where each has widespread use, and each is culturally marked, so giving prominence to some specific form would be a value judgment. However, duplication within the same page, of what are otherwise 98% identical entries, like we had for B/C/S/M before unification, is simply retarded. My point is that the situation that we have for Glagolitic/Cyrillic OCS is less similar to Serbo-Croatian unification, and more to Serbo-Croatian Latin/Cyrillic treatment, and English colour/color treatment. The only reason why we don't have duplication of English entries everywhere is due to the fact that we're not dealing with such high-profile words as colour/color, and nobody bothers to enforce syncing (whatever gets first created gets to be the "default" spelling, and variants get redirected). That the issue of variant English spellings is an unsolved issue only supports the duplication as the only neutral solution. --Ivan Štambuk (talk) 20:20, 14 July 2013 (UTC)Reply

Well, I still invite you to discuss this more widely, because I'm sure there will be different views on this. —CodeCat 20:41, 14 July 2013 (UTC)Reply

(This comment pertains only to the mention of colour and color, not to whether or not Glagolitic and Cyrillic OCS should be merged.) CodeCat understates things when she says "having both spellings in English with full duplication is not without controversy as well". It is de facto not done anymore. Colour and color are the only entries which remain synchronized and unmerged (if you find any other entries that are synchronized, let someone know, so they can be merged), and the only reason they remain unmerged is so that they can serve as a Mahnmal to the otherwise unbelievable fact that Wiktionary was once so naïve that it did try to duplicate the contents of dozens of entries, and they did all spend most of their existence out of sync. - -sche (discuss) 21:38, 14 July 2013 (UTC)Reply

I don't really buy that. A bunch of entries were forcibly merged by editors under the excuse that other spelling allegedly constituted a copyvio of the preferred one, or that the preferred one was earlier created. That we have both color and colour only shows that there is no consensus on what spelling is the "proper" one (the only true answer: none is). All of these redirects will have to be undone in the future in order to provide a proper NPOV treatment of all English varieties, or solved through a technically more feasible solution. This is not a dictionary of neither American English nor British English, what it it should be is a dictionary of both without giving a special preference to either, but thanks to you guys it's at the same time a dictionary of both and neither. My opinion is that you are vastly overestimating the effort of keeping such entries in sync, and the process can be to a large degree automated. --Ivan Štambuk (talk) 22:05, 14 July 2013 (UTC)Reply