Wiktionary talk:About Serbo-Croatian

Definition from Wiktionary, the free dictionary
Jump to navigation Jump to search

Archived discussions[edit]

Four sister-project links[edit]

In a few entries (eg Transilvanija) there are four wikipedia boxes for links to the same subject in four languages (bs., hr., sr., sh.). If this is common, would it be better to make a {{sh-wikipedias}} to compactly show all four? --Bequw τ 20:51, 17 July 2011 (UTC)

Hm. I'm afraid you've touched a nerve there. >:-) Many on bs, sr and hr don't acknowledge sh as a language, not to say sh.wikipedia as a legitimate wikipedia. So creating such a template might be a contentious step. But I support it nevertheless. --BiblbroX дискашн 21:16, 17 July 2011 (UTC)

Another task to do[edit]

Category:Serbo-Croatian nouns lacking gender has 27 members, not too good that. Mglovesfun (talk) 21:40, 30 December 2011 (UTC)

Thank you for your selfless work of populating Serbo-Croatian cleanup categories. --Ivan Štambuk 21:50, 30 December 2011 (UTC)

Automatic Cyrillic Latin conversion tool[edit]

For your consideration a conversion tool Module:sh-translit. Se the talk page about how it can be used and the results. It needs some work on diacritic conversion and testing. --Anatoli (обсудить/вклад) 00:00, 28 March 2013 (UTC)

  1. Both directions (Latin->Cyrillic and Cyrillic->Latin) will be needed.
  2. There should be parameter for optional diacritic stripping.
If these conditions were met, then the module could be incorporated in headword templates (sh-PoS) removing the need for two parameters in most cases. The only exceptions would be words such as nadživjeti where -dž- at the morpheme boundary is not converted to џ since it represents two separate sounds. Rare exceptions such as this would be then handled by override parameters (defaulting to module if not present). --Ivan Štambuk (talk) 13:44, 9 July 2013 (UTC)

Additional referenece[edit]

Beside HJP as most editors are familiar with (which is for Ijekavian Croatian, though it contains many Bosnian/Serbian words marked as regionalisms), I recently stumbled upon another awesome resource: this one. It is almost as comprehensive as Rečnik srpskohrvatskoga književnog jezika (2nd Ed,1990, Matica hrvatska/srpska). After free registration, you get access to all of the dictionaries. It also lists accents and Ijekavian/Ekavian variant forms. By the virtue of being an online resource, it's much more convenient to use than RSKJ in PDF/DjVu format. You cannot copy/paste from it, but you need to dig up direct links to headwords in HTML source, which look like this. I wouldn't recommend adding it as a reference in SC entries quite yet, because the site is apparently "under construction". --Ivan Štambuk (talk) 13:55, 9 July 2013 (UTC)

Judging from the sample the resource seems quite impressive. Nice catch, Ivan. --biblbroksдискашн 14:09, 9 July 2013 (UTC)

Collective plural forms[edit]

E.g. ždrijebe/ždrebe and ždrebad. I think that we should have separate lemma entries for both (given that collective plurals are formed in a special way, and that most dictionaries have them as separate lemmas), but with collective plural forms listed as normal "plural" in the declension template of the main lemma, with a special note (w.g. via the title= parameter) that we are dealing with a noun having collective plural. The collective plural lemma could then have definition line like I've put for ždrebad, something like:

  1. Collective form of ždr(ij)ebe.

That would eliminate duplication of the definition lines, as well as needlessly split morphological information. When a user looks up a word having a collective plural, he expects to have all that in a single place, and not be clicking around comparing tables and cases and endings.. --Ivan Štambuk (talk) 11:01, 10 July 2013 (UTC)

substantivized adjectives (colors, languages)[edit]

A while back I converted all of the language names (in -ski) as a special cases of adjectives, on e.g. [[srpskohrvatski]], after having a discussion about it on my talkpage with Krun. Similar thing could (and IMHO should) be done for colors as well. I added a nominal sense for crvena at [[crven]], and now [[crvena]] is simply treated as an inflected form of crven. There are more such cases in Category:sh:Colors. HJP, Речник српскохрватскога књижевног језика, Речник српског језика (Матица српска, 2007) all do it that way, and I suggest we do it as well. --Ivan Štambuk (talk) 16:36, 17 July 2013 (UTC)

Missing pages[edit]

From Special:WantedPages. DCDuring TALK 17:16, 22 October 2013 (UTC)

Headword module[edit]

@Biblbroks, @Ivan Štambuk I've started writing a Module to handle automatic transliteration in headwords. What do we need our headword templates to do other than what they were doing already? I remember seeing a discussion about this somewhere but I can't find it now. Crom daba (talk) 03:31, 3 November 2016 (UTC)

@Crom daba This is great. Please make test cases to make it more convincing for both ways.
The module and editors need to be aware of the fact that while Cyrillic to Roman conversion in Serbo-Croatian can be 100% reliable, in Roman to Cyrillic conversion "nj", "lj" and "dž" may be converted to one of њ/нј, љ/лј, дж/џ pairs. I think such cases should have an additional parameter, manual conversion or similar. In any case, the manual intervention is required. One of the places I mentioned this was in Wiktionary_talk:Votes/pl-2016-01/Automated_transliterations. Repeating the ping @Biblbroks, Ivan Štambuk. --Anatoli T. (обсудить/вклад) 02:31, 4 November 2016 (UTC)
I set it up so that backtick ` can be used between the constituents of the potential digraph, like so:
Lua error: bad argument #3 to 'gsub' (function or table or string expected, got nil)
This will be meticulously documented before putting the Module into use. What I'm having a problem with currently is this (test) template: User:Crom_daba/sh-noun when I try to use it in the sandbox it doesn't detect the argument given in the template definition itself. Might you know how to make this work? Crom daba (talk) 19:44, 4 November 2016 (UTC)
I just noticed that í didn't transliterate properly, I don't know why I thought that accented letters transliterated without being treated as separate characters. Crom daba (talk) 19:47, 4 November 2016 (UTC)
That's why it's best creating the test module and adding a number of cases with diacritics. Then you'll have a solid proof that it's going to work. Sorry, I can't help with your question. Please try WT:GP. Good luck! --Anatoli T. (обсудить/вклад) 23:31, 4 November 2016 (UTC)
I've been searching through discussions for sh headword templates requirements but haven't found anything particularly relevant. Sorry. Anyway, good work you've done. Thanks. --biblbroksдискашн 21:05, 9 November 2016 (UTC)
CodeCat took over the project, all credit goes to her. Crom daba (talk) 00:57, 10 November 2016 (UTC)