Jump to content

Support

From translatewiki.net
(Redirected from User talk:WelcomeMessageBot)
Support — Problems? Questions? Suggestions? Write your message below. — See also the FAQ or try to catch us in IRC or in Telegram.
Archives
2023
  • Jan
  • Feb
  • Mar
  • Apr
  • May
  • Jun
  • Jul
  • Aug
  • Sep
  • Oct
  • Nov
  • Dec
2024
2025
  • Jan
  • Feb
  • Mar
  • Apr
  • May
  • Jun
  • Jul
  • Aug
  • Sep
  • Oct
  • Nov
  • Dec


Enable translation to the southwestern/"Chippewa" dialect of Ojibwe (ciw)

I would like to translate into the southwestern dialect of Ojibwe

ISO 693-3 code: ciw

Native name: Anishinaabemowin or Ojibwemowin (the specific region would be zhaawani-ningaabii'ani-ojibwewakiing)

JGHFunRun (talk) 20:03, 11 December 2024 (UTC)[reply]

Do you actually know this language? Amir E. Aharoni (talk) 17:39, 25 December 2024 (UTC)[reply]
I am learning, but I'm not particularly great ATM (although I try to only provide translations for which I am confident). If necessary I can wait until someone else steps up or until I meet the required fluency; it likely is a good idea to wait until then now that I think about it. JGHFunRun (talk) 18:50, 14 January 2025 (UTC)[reply]
Yes, let's do that. According to the page Translatewiki.net languages, there should be at least one person with native or comparable proficiency willing to translate into the language. Amir E. Aharoni (talk) 20:53, 14 January 2025 (UTC)[reply]
@JGHFunRun: It would also be a good idea to hold off on creating articles on the incubator in ciw since you don't know the language. It creates issues down the line ranging from making the language community avoid it to the whole Scots Wikipedia mess. It can also cause problems with datasets should someone who cannot assess the quality of the Wikipedia texts in ciw decides to collect the texts for natural language processing applications, perpetuating the problem. Anyway you look at it, it's not a good idea and you (and all of us too) should stick to languages we know well when there is not an active community of native speakers who are willing to fix the texts. - Yupik (talk) 21:32, 17 January 2025 (UTC)[reply]

Translated untranslatable messages

{{Msg-meta}} (used in {{Msg-mw}} etc.) uses MediaWiki:Quotes (“test”) to render messages. As far as I can tell, it seems to work in English as well as other languages to show text wrapped in some variation of "quote" marks. Despite that, if you go to Special:Translations/MediaWiki:Quotes, it tells you that it is not actually a translatable message. Current translations can therefore only be seen at Special:PrefixIndex/MediaWiki:Quotes.

This behaviour is different from Special:Translations/MediaWiki:Parentheses, Special:Translations/MediaWiki:Brackets etc. which do display the translations. Why is Quotes in particular handled like that, and is there any reason why it does not behave like other similar optional messages? stjn[ru] 17:29, 18 December 2024 (UTC)[reply]

Maybe Quotes (page does not exist) should be added to MediaWiki:Betawiki-messages? That way it will be shown to translators when translating the Translatewiki.net group (there are already a few messages there used in TWN templates), and Special:Translations/MediaWiki:Quotes will work as it should. Jon Harald Søby (talk) 16:01, 25 December 2024 (UTC)[reply]

Request to Activate Lampung Api language

Dear Translatewiki Team,

I am writing to request the activation of the Lampung Api language on Translatewiki.

ISO 639-3: ljp

English Name: Lampung Api

Native Name: Lampung Api

It has approximately 1,3 million speakers across Indonesia.

It is written in Latin script and Lampung script (Native name: Had Lampung/Kaganga) which is not yet added to Unicode standard.

Portal: https://translatewiki.net/wiki/Portal:Ljp

As a contributor and advocate for the Lampungese community, I believe enabling this language will significantly support localization efforts and promote the preservation and usage of Lampung Api in various digital contexts. The activation of this language will allow contributors to translate software interfaces, helping to make technology more accessible to Lampungese speakers worldwide. If there are specific requirements or steps needed to proceed with this activation, please let me know, and I will gladly assist. Thank you for considering this request. I look forward to your response and the possibility of working together to bring the Lampung Api language to Translatewiki.

Best regards, Dede DedeKurnn (talk) 05:01, 23 December 2024 (UTC)[reply]

Hi,
Thank you for the request.
How different is this language from Komering (kge)? There is a Wikipedia in it at https://kge.wikipedia.org . Amir E. Aharoni (talk) 18:17, 25 December 2024 (UTC)[reply]
Komering (kge) is a dialect of Lampungic language family, and sometimes included in Lampung Api, sometimes treated as an entirely separate language. Komering people see themselves as ethnically separate from, but related to, Lampung people. Both Lampung Api and Komering is mutually intelegible, but somewhat different in spelling, pronunciation, and vocabulary. Both use different kind of traditional script for writing, Lampung use Kaganga/Had Lampung, while Komering can be written with Jawi (Arabic script) and Komering script. You can read more about the details here: https://en.wikipedia.org/wiki/Lampung_language DedeKurnn (talk) 00:30, 26 December 2024 (UTC)[reply]
We also have Wikimedia Incubator program for Lampung Api and Lampung Nyo, and as a native speaker I actively write articles in Wikimedia Incubator for Lampung Api language.
https://incubator.wikimedia.org/wiki/Wp/ljp/Main_Page DedeKurnn (talk) 00:45, 26 December 2024 (UTC)[reply]
@Amire80: Just wanted to add as an active editor (under User:Swarabakti) in kge.wp. Lampungic itself is pretty diverse dialectologically in terms of phonology, lexicon, and even morphology (but less so in syntax). That said, studies in early 2000s (cf. Anderbeck 2007 and Hanawalt 2007 in the "Lampung language" article on en.wp) have clarified the relationship between these varieties and proposed the three-way division currently adopted by ISO 639-3 (down from 9 separate codes). You can read them to get a clearer picture of the differences but in short:
  • Nyo dialects clearly constitute a distinct, cohesive dialect cluster within Lampungic, but the other varieties form a large dialect chain stretching from Kayuagung down to Kalianda, with no clear boundaries between them.
  • Within this dialect chain, Hanawalt (2007) considers Komering to be sufficiently distinct on both linguistic and sociological grounds, but possibly also Kayuagung (formerly coded [vky], currently under [kge]), Daya, and Jabung (the last two now under [ljp], though some Daya people consider themselves ethnically closer to Komering).
  • Anecdotally, I can say that the phonological, lexical, and morphological differences between these varieties are almost comparable to e.g. that of distinctly coded languages within the Southern Batak subgroup (Toba [bbc], Angkola [akb], Mandailing [btm]), only that the clustering is less clear — a classic case of dialect continuum, really.
Note also that both Lampung Api and Lampung Nyo enjoy greater institutional support in Lampung Province than Komering does in South Sumatra Province; since there are 1) province-wide mandatory school subjects on the two varieties, which are (AFAIK) considered equally "standard", 2) an undergraduate minor and a graduate study program on Lampung language in the local state university, and 3) an active contemporary literary scene with regular publication (cf. the "Lampung literature" article on id.wp). Jawadywn (talk) 19:21, 30 December 2024 (UTC)[reply]
Thanks for the details. We'll add it soon. Amir E. Aharoni (talk) 02:47, 1 January 2025 (UTC)[reply]

OpenHistoricalMap and Fuzzybot

FuzzyBot just imported new texts for old messages with many errors, one example here:

« Projects:Ohm-api.notes.rss.title/fr » : historique des versions — translatewiki.net (see the 2 newer versions: openhistoricalmap becomes openstreetmap).

This message is not marked on translatewiki as untranslated or outdated so this bad import is invisible on translatewiki.

Apparently it is not the first time FuzzyBot creates this mess (see first and second version in the link).

I have no time to read all 2901 messages of this projet to fix them so I suggest reverting these latest auto changes and doing something to prevent another big mess of Fuzzybot. Mahabarata (talk) 21:42, 26 December 2024 (UTC)[reply]

It's not FuzzyBot's fault. Upstream has done these changes: https://github.com/OpenHistoricalMap/ohm-website/pull/261#issuecomment-2562714833. Nike (talk) 08:16, 27 December 2024 (UTC)[reply]
Ok, thanks for this explanation and sorry for my wrong accusations. Mahabarata (talk) 13:23, 27 December 2024 (UTC)[reply]
@Mahabarata, I second you. It takes two steps to prevent readers given uncorrect translations.
Before translation, we need a step to verify if the updates at external sites is reasonable to users including Wikimedians.
My worry is why translators are the first and the last to wipe clean "the bottom" of FuzzyBot and the mess? Is there anybody else who will receive pings and notes what is going on? My inbox has messages like below, and both are imported not matching sentences from what you call the source, but by our Bot:
> There are new phrases awaiting review at TranslateWiki:
> == !recent (ja; untranslated messages) ==
> * 1288:Ohm-site.copyright.legal_babble.more_1_1_html
> * 1288:Ohm-site.welcome.rules.para_1_html
-- Omotecho (talk) 22:11, 30 December 2024 (UTC)[reply]
@Mahabarata @Omotecho: On behalf of the OHM project, I apologize for the mess that this caused. The development team has been discussing some approaches for avoiding these problems in the future. If all else fails, there's also the nuclear option of starting the fork over from scratch. If we do that, we'll make sure to preserve existing translations as much as possible. Thank you for your patience and especially for helping us out with these translations. Minh Nguyễn 💬 18:59, 14 January 2025 (UTC)[reply]

Database error due to 300+ messages on this page

After clicking on New messages, which shows the content of this page as I'm watching it, the following database error occurred. 

[7d7cbc0c40dc390d9c5ca38a] 2024-12-28 09:35:32: Fatal exception of type "Wikimedia\Rdbms\DBQueryError"

I'm hoping this isn't something that happens to everyone clicking here and watching this page (usually with <20 messages) Cigaryno (talk) 09:40, 28 December 2024 (UTC)[reply]

@Cigaryno: Special:NewMessages is part of the LiquidThreads extension, which we used to use as the main discussion platform here, but we have now moved back to using normal wiki talk pages (enhanced by the DiscussionTools extension). Most of the pages that used to use LiquidThreads have been archived, except mainly some user talk pages. You can safely ignore that link – if you want to follow any specific talk pages, just used the watch feature, and if you want to follow specific discussion, use the subscribe feature next to the header for that topic. Jon Harald Søby (talk) 15:26, 28 December 2024 (UTC)[reply]
It happens for everyone visiting Special:NewMessages (see above), but, as Jon says, very few people will want to open it in the first place. —Tacsipacsi (talk) 15:05, 29 December 2024 (UTC)[reply]
Okay, thanks for this. Afterall, Echo is what can notify about this page if watched. Cigaryno (talk) 16:03, 30 December 2024 (UTC)[reply]

FuzzyBot mulfunction and mass imports jambles (OHM)

Who stops or roll backs those imported contamination by FuzzyBot? What is wrong with its codes? There are mass destruction by it, at least for OpenHistoricalMap: a sample below. Newly imported sets reads: "(Importing a new version from external source)". Actually the FuzzyBot has been contaminating and inserting jambles.

  • check from or earlier than timestamp 14:45, 26 December 2024.

Here is a sample:

Happy holidays, Omotecho (talk) 19:09, 28 December 2024 (UTC)[reply]

Nothing is wrong with the code of FuzzyBot, simply someone changed/created a lot of translations in the OHM source code. See above. —Tacsipacsi (talk) 15:07, 29 December 2024 (UTC)[reply]
Forget that someone... The rightiousness of our code does not proof we are safe from problems troubling us, as far as it is our FuzzyBot. Wonder if you are suggesting we would need somebody to judge the imported contents are correct/reasonable, or not? -- Omotecho (talk) 22:17, 30 December 2024 (UTC)[reply]
As explained in the the thread above, the changes were too massive to review. The other option would be holding back all translation updates, or disabling OHM translations temporarily. Nike (talk) 09:42, 31 December 2024 (UTC)[reply]

[mui] Language name and mainpage

Currently the native name is written as "Baso Palembang", although most sources on this language use "e" instead of "a" in the first syllable to represent the schwa [ə] sound, cf.:

  • Zulkifly, H. U. A. (2007). Kamus sederhana bahasa Palembang / Komus ringkes baso Pelembang [A simple dictionary of Palembang language]. Tangerang: DPC Kerukunan Keluarga Palembang.
  • Amin, Baderel Munir et al. (2010). Tata bahasa dan kamus baso Pelémbang [Grammar and dictionary of the Palembang language]. Palembang: Yayasan Madrasah Najahiyah.
  • Tadjuddin, Zuhdiyah Malik et al. (2024). Buku baso Pelêmbang: Sekolah Dasar (SD) / Madrasah Ibtidaiyah (MI) kelas 4 [Palembang language textbook: Primary School / Islamic Primary School, 4th grade]. Depok: Multisarana Nusa Persada.
  • Panji et al. (2024). Komus lengkep baso Pelembang [A comprehensive dictionary of Palembang language]. Palembang: Anugrah Jaya.

The second "e" vowel is an [ɛ] and is variously represented as "é", "ê" or "e" (in which case it would be the same glyph as the schwa). Following the regional Language Agency recommendation (Trisman et al. 2007, Pedoman ejaan bahasa Palembang), I prefer using "é" to represent this sound (thus "Pelémbang"), since "ê" or "e" are more commonly used to represent schwa in Indonesian linguistic tradition. The baso 'language' qualifier can be excluded as it is pretty clear that it refers to the language variety in this context.

Also, for the MediaWiki translation of "Mainpage/mui", formerly a literal, almost 1:1 translation Laman Luan is used, but this is a term that is virtually unknown to most speakers. I tried changing it to the more idiomatic translation Muaro Lawang ('front door') but apparently I need extended permission to do that... Jawadywn (talk) 20:07, 30 December 2024 (UTC)[reply]

Thanks for the detailed explanation and sources. We'll change the name soon—it may take a few days because of the holiday season.
I've updated the "main page" messages.
Happy new year! :) Amir E. Aharoni (talk) 02:46, 1 January 2025 (UTC)[reply]
Thank you and happy new year, too! Jawadywn (talk) 06:47, 1 January 2025 (UTC)[reply]
Actually, I have a couple of questions.
I found "Pedoman ejaan bahasa Palembang mention" online, and I see that it mentions the letter é, but doesn't use it when actually writing the name "Pelembang", for example, on pages 33 and 42. So maybe it's not necessary at all?
Or if a 2024 school book (Tadjuddin, Zuhdiyah Malik et al.) says "Pelêmbang", then maybe that is better? My inclination is usually to go with what school books suggest, but I understand that it's not always the best thing. Amir E. Aharoni (talk) 00:45, 5 January 2025 (UTC)[reply]
@Amire80: Alright, here's the thing. The Pedoman says that the usage of the accent mark is optional to distinguish between pepet [ə] and taling [e]/[ɛ] (they consider the latter two sounds as the same), cf. in page 3 where they use it in éjo [ɛd͡ʒɔ] 'to struggle', séwét [sɛwɛt] '(a piece of) cloth' and sengé [səŋe] (not [səŋɛ]) 'grin (n.), to grin (v.)'.
As for the schoolbooks, they actually distinguish three sounds spelled with "e" in the Pedoman, which are:
  • "e" for [ə],
  • "ê" for [ɛ], and
  • "î" for [e].
So the examples above would be written as êjo, sêwêt, and sengî. Similarly, they also distinguish between "û" [o] and "o" [ɔ], e.g. in jerû [d͡ʒəro] 'inside' and jero [d͡ʒərɔ] 'deterred', both of which would've been written jero if we follow the Pedoman.
If we consider popular usage, most speakers barely use any diacritics at all, but if they do, they will usually only use "é". Google Search isn't a representative corpus but a quick browse there for the terms "Pelémbang" with an accent "é" has around 5.380 results, while "Pelêmbang" with a circumflex "ê" has only 7 results, all referencing the newly published schoolbooks.
FWIW, last November, me (as User:Swarabakti) and the local community of Wikimedians in Palembang actually managed to collaborate with Prof. Zuhdiyah (Malik Tadjuddin is a patronym) and the other authors of Palembang-language schoolbooks and dictionaries in a workshop for Wiktionary and Lingua Libre.
During the event, we actually brainstormed several alternatives for the orthography to be used in Wikimedia projects, ranging from not using diacritics at all (as is still common among the speakers), using only one glyph with diacritic to separate the taling sounds [e~ɛ] (which would be "é", following the option given by the Pedoman and the default diacritic people use when in need), or using three glyphs with uniform circumflex diacritics ("ê", "î", "û") as in the schoolbooks.
Eventually we settled with a pragmatic compromise of using the same orthography as the schoolbooks, but with "é" instead of "ê" for [ɛ] (see for example the Indonesian Wiktionary entry for ajén 'to respect'). This is not without precedent, as a similar system ("é", "î", "û") is used in Amin et al. (2010), of which Prof. Zuhdiyah herself is actually a co-author.
So yeah. I think what I did not convey previously is that the language is still in the process of standardizing the written form, and there is no single agreed orthography as of now. While the Zuhdiyah et al. (2024) schoolbooks' version has the most potential to be the standard, it is not (yet) adopted officially; currently the local govt only mandate "Palembang language" as a subject without requiring a specific standard, syllabus, or materials to be taught... Jawadywn (talk) 16:26, 8 January 2025 (UTC)[reply]
OK, I guess it's convincing enough :)
We'll do it soon. Amir E. Aharoni (talk) 19:17, 8 January 2025 (UTC)[reply]

Ottoman Turkish language

I really want Ottoman Turkish to come back because I saw it enabled on Wikidata but not on TranslateWiki. BlueHatMario (talk) 02:41, 1 January 2025 (UTC)[reply]

Being enabled on Wikidata doesn’t mean that it was ever enabled on Translatewiki, nor that it should be. This is especially true for old languages: Wikidata accepts extinct languages, as data like inscriptions may be written in them, but (as far as I remember, I don’t find the policy at the moment) Translatewiki supports (with a few exceptions) only living languages, since that’s what helps making software accessible for people – probably all or almost all people who would use the software in Ottoman Turkish can also use it in Modern Turkish as well. As far as I understand w:Ottoman Turkish#Legacy, while it’s taught in schools, this teaching is more like Latin and Ancient Greek in European schools, not like a living foreign (or native) language. —Tacsipacsi (talk) 10:39, 2 January 2025 (UTC)[reply]
If Ottoman Turkish does get enabled on TranslateWiki, then the fallback language should be Turkish. BlueHatMario (talk) 14:00, 2 January 2025 (UTC)[reply]
Ottoman Turkish will not get enabled on translatewiki for localization. The policy is on Translatewiki.net languages: it must be a living language and a general purpose language. It's enabled on Wikidata, but not for user interface, but for collecting historical data like names and words. Amir E. Aharoni (talk) 15:30, 2 January 2025 (UTC)[reply]

Balkan Gagauz Turkish

I really want Balkan Gagauz Turkish to be used because it's not enabled or used on TranslateWiki. BlueHatMario (talk) 15:52, 2 January 2025 (UTC)[reply]

Can you write this language? Amir E. Aharoni (talk) 16:05, 2 January 2025 (UTC)[reply]
adımda ailêm arkadaşım artırmak âmcası attıktan ağlıyordu ağacın altında alırdım âdâm attıktan atmosfêreydi sıkıntısı arkadaşlarında gêçirmeyi yüreyîme sêvdiim dêiştirmek âldıımdâ dêprêsiona yüreyime yoktu ormanın kênârında yürüyordum eğlencelî sêçti yolcularla erkeğê arkadaşım dêprêsiona dêprêm etmeye değerlerini dêprêsiona düşünyorum öğrencîler yêtiştirdîi yürütêç erkeklêrin ôlduğu profesyonêl yolculûklarını yoluylâ reâktörü temîz têşekkürlêr yêtiştirdîi yêtiştirmê tarafından tariflêrini sâdecê tükênmez teşviktê ölüm gürültülü giyilên temîzlenir temîz têşekkürlêr tercîh sâdece çıllık saldırgân sektörleylê dêiştirdim terörlerî rastgelê rengî güreşî yumuşak güçlüydü görüyorum onnarı uykuya dêprêm kullanılır kızgın kürtlêr kârarları katılmıyorum katlardan kahvaltısı küçük köyümuzdê kultürümdê kahvaltı küresel ôlduğun korkutcu BlueHatMario (talk) 18:24, 2 January 2025 (UTC)[reply]
This language is very similar to Azerbaijani/Turkish and I can say that I didn't understand anything from this text. It looks like it is random words. Nemoralis (talk) 00:58, 5 January 2025 (UTC)[reply]
It's not. I got them from a memo. BlueHatMario (talk) 02:25, 5 January 2025 (UTC)[reply]

Text replacement request

Can someone please replace all the instances of msg-mw to msg-wikimedia in message documentations related to Translating:Wikipedia Year in Review? See, for example, Wikimedia:Wp-year-in-review-badge-template-wizard-desc/qqq. A bunch of them are currently in Category:Message maintenance/msg-mw because of this typo, and I don’t want to do it by hand because it’s a lot of messages. But I also don't have access to ReplaceText myself. stjn[ru] 15:52, 3 January 2025 (UTC)[reply]

@Stjn: Done Done Jon Harald Søby (talk) 17:37, 3 January 2025 (UTC)[reply]
@Jon Harald Søby: a number of these also seem to have names like Wikimedia:Badge-100k-desc/qqq and Wikimedia:Talk-2/qqq, are they not part of the group or are those names with no prefix erroneous? stjn[ru] 18:06, 3 January 2025 (UTC)[reply]
Ah. It seems to be another case of ‘non-translatable messages being left dormant after they stop being translatable’. Maybe we should do something about it already (a check for some could be added to {{Msg-meta}} directly if I knew what to check for, since /en subpage also exists in these cases). stjn[ru] 18:42, 3 January 2025 (UTC)[reply]

On every message there is link to "Ask for more information". It suggest you can use it to ask for explanation. But in MediaWiki messages it leads to phabricator where we can report bugs with them to have, for example, rewording original text or add better documentation, but it is not support centre for questions.

The title should be reworder or target page changed to this or purpose of the place should be clarified, especially it is about questions or reporting bugs.

Example problem: phab:T382748 Woytecr (talk) 15:22, 5 January 2025 (UTC)[reply]

For reference, a discussion about the links to Phabricator/the wording of the links has been opened in phab:T383488. Best, a smart kitten (mw // phab // talk) 11:44, 16 January 2025 (UTC)[reply]

User:Hadĵnix contribs

Please delete his contributions his knowledge of Serbian language is not good. He has a problem with grammar and properly forming sentences. Milićević (talk) 11:03, 6 January 2025 (UTC)[reply]

OK, this can be done easily using the Nuke function. However, I wanted to check something first: There are relatively few of them, and it was long ago. Are they all completely bad? If they are not all bad, perhaps the bad ones could be simply fixed? They are all in the IABot project.
If it's all awful, I can delete. Amir E. Aharoni (talk) 18:00, 6 January 2025 (UTC)[reply]
Please delete, i already fixed a lot of his translations on MetaWiki and MediaWiki.org and have no desire to clean up after him anymore Milićević (talk) 18:47, 6 January 2025 (UTC)[reply]
OK, I deleted and removed translation rights. Amir E. Aharoni (talk) 20:56, 6 January 2025 (UTC)[reply]

Update MediaWiki Credits

Can we please update Translating:MediaWiki/Credits? It is updated ~5 years ago. Nemoralis (talk) 00:01, 7 January 2025 (UTC)[reply]

Hopefully @Nemo bis can do it. In this edit, he left "instructions" on how to do it, but I think you'd have to download all (?) supported extensions for that to work, and I'm not sure I want to do that, hehe. Jon Harald Søby (talk) 10:25, 7 January 2025 (UTC)[reply]
We have all the extensions cloned on the translatewiki servver. Nike (talk) 10:32, 7 January 2025 (UTC)[reply]

Bug: no documentation and English text

Something happened, and now, when I edit the page in the wiki editor, neither the original text in English nor documentation is displayed! maro21 (talk) 22:50, 7 January 2025 (UTC)[reply]

@Maro21: Hi! I'm afraid you have to be a bit more specific. Which page? Jon Harald Søby (talk) 08:46, 8 January 2025 (UTC)[reply]
Any page with a translation string. For example [1]. There is "Information about message (contribute)" and "Message definition (Url Shortener - User interface)" below. But it works now, thanks for fixing :). maro21 (talk) 17:50, 8 January 2025 (UTC)[reply]

Category:Supported projects

Is there any reason why some projects in this category are under * and some under letters A,B,C...? maro21 (talk) 13:01, 11 January 2025 (UTC)[reply]

JSON file format

Hi there, I look for the official format guideline of the JSON files compatible with translatewiki.net. Any idea where I can find that ? (A dev working for Lingualibre has submitted a nested i18n/en.json, and I believe this won't be supported by Gitlab/Github translatewiki bots.) Yug (talk) 19:14, 12 January 2025 (UTC)[reply]

That is fine if nestingSeparator is configured (see https://www.mediawiki.org/wiki/Help:Extension:Translate/Group_configuration#FILES). Nike (talk) 19:21, 12 January 2025 (UTC)[reply]

Correction Request for Ukrainian Translation of Romansh Language Name

Hello, I would like to kindly request the correction of the Ukrainian localised name for the Romansh language in the system interface across Wikimedia projects.

Currently, the name displayed—such as on Wikidata—is "ретороманська", which is incorrect. It should be updated to "романшська".

Thank you very much in advance for helping me with this. Gzhegozh (talk) 01:12, 15 January 2025 (UTC)[reply]

Can you please request this in Phabricator? I see that you already have an account there.
It would also be nice if you could give some references why the name "романшська" is better. It's OK if the sources are in Ukrainian—I can read it.
Thanks! Amir E. Aharoni (talk) 01:17, 15 January 2025 (UTC)[reply]

Grammatical support to Toki Pona

I wish to request some technical support with dealing with grammatical information relating to names in toki pona (tok). toki pona is an isolating language, so it does not need any grammatical support for its morphology. however, a problem that we in the translating community have encoutered is for names.

Introduction and issues

brief introduction: instead of proper nouns, toki pona has proper adjectives, this means that a name is accompanied by a head word or phrase (called the headnoun) that describes the name with an entity: if you have something named "Sydney" (Sini), it may be a place (ma Sini), a person (jan Sini) or a football group (kulupu musi Sini) or something else entirely. this is often backtranslated as "the thing named Sydney"). it is typically ungrammatical to translate without that headnoun. for wikis, this means that we have to make a choice about how the name the wiki otherwise everything is either has (1) no headnoun or (2) too many headnouns.

the actual name part of name is the latter part of the phrase, found with capital letters (toki pona is otherwise all lowercase). names may take different headnouns even when referring to the same entity in order to showcase different aspects of it. for example, with Wikipedia (Wikipesija):

  • lipu ("book, document") — as the wiki is something you read
  • ma ("land") — this is a common metaphor, that the wiki is a place to edit
  • ilo ("tool, intrument") — this refers focuses on the aspect of the wiki as a tool for something
  • kulupu ("group") — this refers to the group that edits the wiki

there is additionally another issue with dealing with wiki names that use words that are already in toki pona, such as the partner wiki to lipu Wikipesija, that is sona pona (literally meaning "good knowledge"). you can't refer to the website in toki pona as sona pona because it would be interpreted literally, you can just say lipu sona pona because again it would be interpreted literally as "the good knowledge book/website". in writing, people trying to preserve the name will either wrap it quotations (lipu "sona pona") or capitalise it similarly to a proper name (lipu Sona Pona).

one more thing, toki pona and other languages, like English, treat toki pona names differently, all I described works for toki pona as names are fluid and can change headnoun, however, when talking about toki pona things in English, one typically picks one headnoun and treats the whole phrase as a name (i.e. in English, it is only called lipu Wikipesija, but in toki pona, it varies). I wonder if there is a way to set the wiki name depending on the language as that would be helpful.

Proposal

with the issue described, let me propose a possible solution using the GRAMMAR function as follows:

{{GRAMMAR:name|1= |2= }}
  • the first argument is the desired headnoun
  • the second argument is the name of the wiki, which can be pulled from SITENAME
  • internal logic:
    • if all the words in the site name start with capital letters, return $1 $2
    • if any of the words in the site name start with lowercase letters, return $1 "$2"

in this method, lipu Wikipesija should be renamed to just Wikipesija for MediaWiki, but otherwise, no further specific changes are needed outside of translating the software itself.

Examples:

  • {{GRAMMAR:name|lipu|Wikipesija}} gives lipu Wikipesija
  • {{GRAMMAR:name|lipu|sona pona}} gives lipu "sona pona"
  • {{GRAMMAR:name|ma|My Test Wiki}} gives ma My Test Wiki
  • {{GRAMMAR:name|ilo|my Test Wiki}} gives ilo "my Test Wiki"

pinging the rest of the people working on translating MediaWiki into toki pona: @Eic17H, @.hecko, @Ookap, @Tamzin, @Tbodt. Juwan (talk) 19:30, 17 January 2025 (UTC)[reply]

not sure about "any of the words" (vs just the first word); it messes with some half-named like (toki) Elena majuna, and i think i might be fine with e.g. lipu Wiki of Testing not having quotes?
also what about caseless writing systems? initial non-letters, like in my username? .hecko (talk) 19:50, 17 January 2025 (UTC)[reply]
good point, I hadn't considered that well enough, reworked for humans:
  • if the word is the the Latin script and has a capital/lowercase version:
    • if the first word in the site name start with capital letters, return $1 $2
    • else, if the first word in the site name start with lowercase letters, return $1 "$2"
  • else, if the word starts with a non-letter symbol (number, punctuation mark), return $1 $2
  • similarly if the word is any other script other than Latin, also return $1 $2
possible reworked for machines:
  • if the word matches [A-Za-z], return $1 "$2"
  • otherwise, return $1 $2
Juwan (talk) 20:20, 17 January 2025 (UTC)[reply]
This (including the more detailed outline just above) sounds good to me! I do want to say that in my experience the style of lipu "sona pona" is much more widespread than that of lipu Sona Pona. However, it seems quotation marks will be preferred for lowercase words, which is good. I think this system is a great idea and am excited for its implementation. Ookap (talk) 23:52, 17 January 2025 (UTC)[reply]

Number support for Twnmp-stats-number-k

the toki pona translation for the string Twnmp-stats-number-k (“$1k”) cannot be added. toki pona has different number systems and numbres cannot be abbreviated with e.g. "k". the simplest option is to just write the number in full. Juwan (talk) 21:35, 19 January 2025 (UTC)[reply]

If there is no word or abbreviation for "thousand", you can simply write three zeros, e.g. Amir E. Aharoni (talk) 04:06, 20 January 2025 (UTC)[reply]

The menu bar is translated, but it doesn't appear in the menu

This is a string for template https://meta.wikimedia.org/wiki/Template:WMF_Navbar :

https://translatewiki.net/w/i.php?title=Special:Translate&showMessage=wm-commonstrings-news-events&group=ext-wikimediacommonstrings&language=uk&action_source=search&filter=&optional=1&action=proofread

The string for the menu bar is translated, but it doesn't appear in the menu, for example at https://meta.wikimedia.org/wiki/Wikimedia_Foundation/uk we still see text "News and Events" instead of "Новини та події".

What should I do to fix it? Perohanych (talk) 10:07, 24 January 2025 (UTC)[reply]

You did everything correctly. It will probably be deployed to Wikipedia on Thursday, January 30. Amir E. Aharoni (talk) 20:22, 24 January 2025 (UTC)[reply]
Retrieved from "https://translatewiki.net/w/i.php?title=Support&oldid=12902012"
pFad - Phonifier reborn

Pfad - The Proxy pFad of © 2024 Garber Painting. All rights reserved.

Note: This service is not intended for secure transactions such as banking, social media, email, or purchasing. Use at your own risk. We assume no liability whatsoever for broken pages.


Alternative Proxies:

Alternative Proxy

pFad Proxy

pFad v3 Proxy

pFad v4 Proxy