User talk:Kephir

Philosophy is a battle against the bewitchment of our intelligence by means of language.

Old discussions: User talk:Kephir/old

Hi!

You get this message because you are an admin on a Wikimedia wiki.

When someone edits a Wikimedia wiki without being logged in today, we show their IP address. As you may already know, we will not be able to do this in the future. This is a decision by the Wikimedia Foundation Legal department, because norms and regulations for privacy online have changed.

Instead of the IP we will show a masked identity. You as an admin will still be able to access the IP. There will also be a new user right for those who need to see the full IPs of unregistered users to fight vandalism, harassment and spam without being admins. Patrollers will also see part of the IP even without this user right. We are also working on better tools to help.

If you have not seen it before, you can read more on Meta. If you want to make sure you don’t miss technical changes on the Wikimedia wikis, you can subscribe to the weekly technical newsletter.

We have two suggested ways this identity could work. We would appreciate your feedback on which way you think would work best for you and your wiki, now and in the future. You can let us know on the talk page. You can write in your language. The suggestions were posted in October and we will decide after 17 January.

Thank you. /Johan (WMF)

18:14, 4 January 2022 (UTC)

I could be wrong, but it looks like about 340 of the 373 entries in Category:Pages with module errors are there due to your edits to this module a couple of days ago. Could you look into it? Thanks!

You were not. Fixed for now. Though I have run into some entries like [[갅]] or [[갆]] that might need to be revisited by someone more knowledgeable in Korean. Specifically, the jamo decompositions may need to be checked.

Thank you! I was, however, right about being wrong: there were 4 Polish entries in that count that were actually due to edits to Module:pl-noun by Tweenk. I just left a message on his talk page.

Well, that depends on how you define about. Almost all sometimes means "all except finitely many".

I was hoping to add some images to the Appendix:Unicode/CJK_Strokes page. A few of the entries are currently illustrated with highlighted strokes of 永, but I've created images that highlight all the strokes with the examples listed in the Unicode block chart. But I see the page is built with "invoke:character list". I poked around a bit and couldn't figure out what I should be editing to replace the 永-images on the list.

Kephir hasn't been active here for the last few years. This is the kind of thing to post at the Grease pit.

Thanks for that. Can you also explain the -ów bit, if possible, please?

w:Kraków#Etymology calls this an "archaic possessive form". Given this vagueness and weak sourcing, I would take it with a grain of salt (note, however, *-ovъ). In modern language, -ów is a common genitive plural suffix (genitive case fulfilling the possessive role), hence it can be understood as "[the town] of the Kraks (servants/descendants/etc. of Krak)". In our Proto-Slavic noun declension tables, however, the genitive plural suffix tends to be *-ъ (and the genitive dual *-u). I would love to see why this is so, how it compares to other contemporary Slavic languages, and how -ów came into being.

I have been hardly seeing you lately. Busy in real life? Fine by me if you do not want to elaborate on it too much, just wondering whether you expect to keep coming here.

Sorry I missed your response from nearly three years ago. I forgot about liquid threads. You haven't been around since September last year, so I don't expect to see you very much. :-(

It returns an error saying "missingparam: One of the parameters rcid, revid is required".

Perhaps you should exclude the sandbox from Special:AbuseFilter/33 "Bad Unicode".

...and a few other testcases pages are in Category:Pages with module errors, although there's language in MediaWiki:Scribunto-common-error-category that puts most testcases pages into the hidden subcategory. Can you figure out how to exclude these pages, too? Thanks.

I think those may be a different type of module error than the ones normally found in testcases pages: User:Vitalik changed the directory structure without bothering to change the "require" statements to point to the new locations. I fixed a batch of them, but got tired of it. You would think that someone doing extended heavy module-programming work would check Category:Pages with module errors every once in a while, but I guess they would rather let us do it for them...

Hi. Can you look at why the ['кь'] = 'ḳ̍' rule is ignored in the headword line of ажакьа? The addition of a stress sign plays a role.

"к" is the third letter from the start. The mw.ustring.gsub(text, '..', tt) line takes successive pairs of characters and substitutes them according to the mapping; here, they are "аж", "ак", and "ьа". Since "кь" is on an odd position, it is not matched. I vaguely remember some other module being bitten by the same problem before.

My solution cheats a bit; it just restricts the pattern to pairs which start with consonants that have entries in the table. If you ever need to process overlapping pairs, or ones with some other consonant letters, it will no longer work.

I see, thanks.

Some languages are written in multiple scripts. Assume we have an automatic transliteration module for each script. Is there a way to make, for example, Ossetian to be transliterated with Module:os-translit when written in Cyrl and with Module:Geor-translit, when written in Geor? I understand that I can copy the content of the latter into the former, but that would lead to duplication and dis-synchronization over time.

That is what require is for. Just check whether the script is "Geor", and if so then return require("Module:Geor-translit").tr(text, lang, sc). Is that something you are actually trying to do?

Yes, I am trying to do it. I would like to start with Kipchak (Module:qwm-translit), which can be written in Latin and Armenian. How would the code look like if I want to skip transliteration if it's Latn, and forward to Module:Armn-translit, if it's Armn. There are many more languages in the pipeline.

FYI I commented with a suggestion on the PAM filter.

Thanks, but I somewhat doubt the condition you proposed is feasible to implement. (Dropping conditions may help, actually…)

I noticed after you edited Module:character info and Module:character list, the character links in the subpages of Appendix:Unicode are broken (as they no longer show up as links but just as displayed glyphs). Was this the intended behavior of those changes? Just curious.

Yes, it was indeed intentional. I have done it, in fact, after I noticed you creating entries for Unicode code points, giving the Unicode character name as the definition. I think such entries should not be created, and other editors have agreed with me in a few discussions about the topic: Talk:⦰ and Wiktionary:Beer parlour/2015/January#Is documenting all Unicode characters within the scope of Wiktionary?; although there was some disagreement about what should be put in their stead.

I am of the opinion that:

• Wiktionary is a project that documents natural languages, not any particular computer encoding of them. There are already many websites presenting the Unicode character database in human-readable form, and there no real benefit to creating poor competition to them.
• Unicode code point names do poorly as definitions of glyphs they represent, and should not be remotely treated as such. They are as just as much of technical artefacts as the code point numbers, combining classes, bidirectional classes, etc.
• Emoji, dingbats and other symbols should be subject to the same attestation criteria as any other lexical item. It is better to be incomplete than unreliable.

So you also think it's reasonable to simply break linking with *everything* in Appendix:Unicode, even though it also contains single-character words such as in the CJK Unified Ideographs range? Then really, why have such an appendix at all if there is no individual character linking? I'm not really following your line of reasoning entirely.

If I tried the equivalent of {{#ifexist:X|[[X]]|X}}, it would run into expensive function limits quite quickly and fail to render. So it really is either all or nothing.

I have no idea why you would like to browse CJK characters through the Unicode appendix anyway; it is not like it is the most convenient way to do it, given their pretty much random placement.

Just because one person finds a method inconvenient doesn't mean everyone does. Unicode is a standard, and its ordering matters to things like CJKV characters; also, the placement is in radical-stroke order, not "pretty much random" as you attest. Perhaps you don't know as much about these things as you think you do?

Okay, I admit that I made that statement based on an extrapolation from my observations about placement of letters in Latin and Cyrillic scripts, which at this point are decided on a "wherever it fits" basis. I realise the analogy may not carry through completely, and Unicode Inc. may even attempt to maintain some sane ordering within individual allocation ranges. I can grant you that. (Even though you cannot rely on this property on a large scale: U+20000 has fewer strokes than U+4684. From that perspective, the placement really is random.)

On the other hand, this is still not the most convenient way to browse CJK characters, given that you only have a meaningless number next to the character itself in a smallish font, and a completely nondescriptive character name of the "CJK UNIFIED IDEOGRAPH-XXXXX"…

Anyway, there are some other reasons why I decided removing all links from that appendix was an overkill after all. I restored the links in Appendix:Unicode. Still, in the future please do not create entries of the kind that I mentioned.

I was just looking at the listings at Appendix:Unicode and it looks like Module:Unicode data needs to be updated to include the new Unicode ranges (including CJK Unified Ideographs Extension E). The new ranges are listed at http://unicode.org/versions/Unicode8.0.0/ . Cheers!

Done. (You could have done it yourself… I left every ingredient here.)

Re your edit summary, it's to remove pages from Special:WhatLinksHere/Template:acronym-old, per the post-deletion instructions.

Fix those instructions then. A redlink to a template can be clicked to reveal the deletion log, which saves some time (compared to typing the template name in the address bar).

Now, redlinks to deleted dictionary entries should be removed, sure. But that is because readers are generally not interested in log entries.

I'd like to build a gadget similar to Quiet Quentin which would allow an editor to automatically import usage examples from tatoeba.org (short description here: Tatoeba-gadget). How can I build a gadget which makes cross-site requests? Is there a process for whitelisting? I guess it needs the Access-Control-Allow-Origin header set.

Hi, I am trying to wikify the list here https://en.wiktionary.org/wiki/Wiktionary:Frequency_lists/Esperanto since the edits require modifying 1000 words at a time, the system is flagging me for vandalism. May I receive some help and or guidance on how to proceed from here?

I made {{pl-decl-phrase}}. Let me know whether the input format is acceptable.

Dobre na początek, choć mnie przyszło do głowy co innego. Raz, w tabelce dobrze byłoby linkować całe odmienione formy, a nie tylko pojedyncze wyrazy. Dwa, formy odchylające się od wzorca można podawać jako parametry całego szablonu przecież, a nie samej odmieniarki (|vocs=Słodka Aniu, nie f!A!ni!vocs=Aniu).

Zastanawiam się też, czy nie dałoby się zrobić, żeby wzorzec deklinacji rzeczownika był wyznaczany przez moduł automatycznie. Pełnego automatu, jak z przymiotnikami, raczej się nie zrobi, i oczywiście wyjątki się będą trafiać wcale nierzadko, ale wydaje mi się, że rzeczownik w mianowniku wraz z rodzajem-ożywionością (m-pr, m-an, m-in, f, n, i osobny znacznik dla przymiotników, bo w końcu sarna → sarny, ale ciężarna → ciężarnej) powinien w większości przypadków wyznaczać jednoznacznie wzorzec odmiany. Wtedy można by zrobić jeden szablon, który może odmienić wszystko, i któremu podawałoby się całe wyrazy. Wtedy wszystko byłoby prostsze i bardziej czytelne.

Wyszukiwarka jest w stanie znaleźć formy występujące w tabelce z odmianą, np. po wpisaniu "elektrycznych zespołów trakcyjnych" w wynikach pojawia się "elektryczny zespół trakcyjny", więc nie wiem, czy tworzenie stron dla tych form jest sensowne.

Jeśli chodzi o formy oboczne, to można całość napisać tak, żeby parametry typu |vocs= były obsługiwane zarówno dla poszczególnych składowych jak i dla całego wyrażenia. Często jest tak, że np. w wyrażeniu przymiotnik-rzeczownik tylko rzeczownik ma formy oboczne - wtedy używając form obocznych dla składowych można zaoszczędzić sobie pisania.

Kwestia 'bardziej automatycznej' deklinacji rzeczowników jest niezależna od samego pl-decl-phrase. Pierwszy parametr to jest po prostu nazwa wzorca deklinacji (elementu tablicy patterns) w Module:pl-noun, więc jeśli wymyślimy jakiś bardziej ogólny wzorzec, to od razu będzie można go użyć w pl-decl-phrase. Natomiast można zrobić coś w tym rodzaju:

{{pl-decl-phrase|f!kiełba!s|adj!wyborcza}}

Dla takich parametrów szablon mógłby sam zgadywać, że przymiotnik powinien mieć żeńską odmianę, bo w wyrażeniu występuje rzeczownik rodzaju żeńskiego.

Hi. You can unflood-flag me again if you fancy. Regards

 Done. Cheers.

Kephir, your bot reverted my edit. The language is Slovene not Serbo-Croatian and I also added a correct definition. Please revert the definition page to include my adds.

Thanks

No, the entry is for Serbo-Croatian. If it also exists in Slovene then there should be a separate entry for that, you shouldn't just go around removing stuff.

I'm not "just going around changing stuff" as you put it. The Slovene translation link sent me to the wrong page, which isn't immediately obvious in the Wiktionary Android App. Only after using a web browser is it obvious. I will create a Slovene page and I assume the translation link will then work.

I'm not "just going around changing stuff" as you put it. The Slovene translation link sent me to the wrong page, which isn't immediately obvious in the Wiktionary Android App. Only after using a web browser is it obvious. I will create a Slovene page and I assume the translation link will then work.

That was no bot, and you are on the way to being blocked for edit warring.

The definition was not in English, and you destroyed a valid language header. To add a definition in another language, add a new header.

Hey! Sorry, for the last revert. For some reason, the Urdu spelling in hi-noun didn't take the Urdu font specifications and it looked really odd and blocky. Now it looks better, however it's bolded and so is the Hindi spelling in Urdu entries. It makes it hard to read (we intentionally increased the size of Urdu and Hindi font specifications so that they are more legible, but making them bold makes both hard to read). Can you make the text regular and not bold?

This should do it.

That's not exactly what I had in mind and your edit only worked on Arabic language and Devanagari script. What I meant was that in the actual templates themselves, when the spelling in the other script is provided, it shouldn't be bolded when it is wikified. Otherwise, as a headword, it's fine to be bolded.

I do not get it. If emboldening hurts legibility, why allow it anywhere? And if it does not, why make an exception for script variants?

I'm sorry if I confused you. I think it was just my browser that's been acting funny the last few days.

Care to lend a hand?

Could you move back Kucing to kucing. Someone moved it and I don't know how xx --Malaysiaboy (talk) 01:18, 17 January 2015 (UTC)

Done. (I have also edited a few related entries which apparently mischaracterised relations between these variously spelled words.)

Thanks a lot for helping!