[01:30:44] I never understood why it seems to be so acceptable. it seems strange to me that commons won't accept various file formats out of principle, because they're not free enough, but then (the last time I checked, at least) seemed to have no issue with people uploading images generated by AIs trained on copyrighted images (re @ebraminio: thank you for summarizing it 🙏 somehow that [01:30:44] "Particularly, there may be infringement claims on the final work based on copyright...) [01:31:25] Ege There is some discussion and related links at https://www.wikifunctions.org/wiki/Wikifunctions:Project_chat#Plan_for_function_categorization%3F [02:36:09] I'm hoping that we'll set up a SPARQL endpoint at some point, but it's not an immediate concern. And I love seeing exactly this kind of experimentation. (re @ebraminio: I'm feeling some kind of tool that can enlist all the functions for power users and maybe query/filter/modify/review them later ...) [05:49:56] just wrote a new experimental tool that duplicates a page, https://www.wikifunctions.org/view/en/Z10091?withJS=MediaWiki:Gadget-duplicate.jsif it sounds ok I can expose it as a gadget that users can enable in gadgets section of preferences, if not, I can hide the tool from users : https://tools-static.wmflabs.org/bridgebot/5d3f716e/file_50742.jpg [07:31:53] on the topic of function categorization and others, we could have a structured data tab? [07:36:35] https://www.wikifunctions.org/wiki/Wikifunctions:Project_chat#Structured_data [08:36:15] parseFloat doesn't have i18n support, I wish we won't see more use of it in the absence of a proper number typehttps://www.wikifunctions.org/wiki/Z10604 : https://tools-static.wmflabs.org/bridgebot/99e6c504/file_50747.jpg [09:25:51] does javascript support using those numerals for actual numbers? (re @ebraminio: parseFloat doesn't have i18n support (doesn't support languages use ۱۲۳ or other digits for their numerals instead of 123), I wi...) [09:29:01] getting the vector-2022 skin on mobile is really annoying [09:51:31] timeless and even monobook are actually responsive enough to be readable on my phone, and I'm completely baffled that the wmf has spent years working on this skin and it still can't display a page on a phone [10:15:24] https://www.wikifunctions.org/wiki/Wikifunctions:FAQ/bn can someone help me understand why this page has a [[Category:Help|*]] categorization and where? It seems I cannot remove it [10:15:58] (I'm trying to re-categorize help pages with my volunteer account, to better help sorting translations, system can be definitely improved!) [10:19:20] I found it: https://www.wikifunctions.org/w/index.php?title=Wikifunctions%3AFAQ%2Fbn&diff=35279&oldid=35214 (re @Sannita: https://www.wikifunctions.org/wiki/Wikifunctions:FAQ/bn can someone help me understand why this page has a [[Category:Help|*]] c...) [10:19:45] thanks! (re @Ameisenigel: I found it: https://www.wikifunctions.org/w/index.php?title=Wikifunctions%3AFAQ%2Fbn&diff=35279&oldid=35214) [11:54:26] 8635 [13:39:05] fwiw we have an open task T343283 to redirect to m.wikifuncions... when visiting from mobile, and load the minerva neue skin. but yeah, i hear you (re @Nikki: getting the vector-2022 skin on mobile is really annoying) [13:40:31] It does for turning number into string, (213).toLocaleString('fa') guess there would be someway parse them also : https://tools-static.wmflabs.org/bridgebot/36788e8e/file_50754.jpg [13:43:55] and indeed no one wants to actually apply number operation on such digits directly, we should use the difference numbers and strings have in programming for the propose [13:47:30] what I mean is that if you can't use them as actual numbers (in something like `if (foo < 1) { ... }`, then using `parseFloat` isn't removing any functionality we would have if we were using actual numbers [13:51:08] thanks, I thought there would probably be a ticket already :) and yeah, I was frustrated when I opened the project chat link and it didn't display very well but I also couldn't really scroll to the bottom to get to the link to the mobile site (page is too long and the sidebar was flickering too much >_<) (re @internetam1n: fwiw we have an open task T343283 to redirect to m.wikifu [13:51:08] ncions... when visiting from mobile, and load the minerva neue skin. bu...) [13:57:37] the main page isn't translatable 😔 [15:28:44] exposed the first gadget created specifically created for https://www.wikifunctions.org/wiki/Wikifunctions:Project_chat#%22Duplicate%22,_a_new_gadget [15:28:44] [15:28:46] feel free to bring your feedbacks about the tool, I found that such a tool can be useful personally when I wanted to created multiple objects that had similar functionalities but sure it should be used with caution, like any other thing in a wiki (re @ebraminio: just developed a new experimental tool that duplicates a function, https://www.wikifunctions.org/view/en/Z10091?withJS=MediaWiki...) [16:33:45] Something like brainfuck? (re @ebraminio: using the tool and searching for "Z16K2" I've learnt we have lambda 'implementation' and found this beauty along the way) [16:34:49] yes: [[w:JSFuck]] (re @infovarius: Something like brainfuck?) [16:49:05] Hello everyone, I've got a question about the senses part in lexemes, when we have a word in x language, and we're entering a sense for that in y language, do we write the explanation, or just the translation? [16:49:07] [16:49:08] Like in L7 , I see that for many languages, there's the word "cat" in their own languages, not an explanation of what a cat is [16:49:59] I normally only add a translation, or at least only enough text for disambiguating the translation [16:50:47] in the original language, you need to avoid using the lemma because that's what you're trying to define, but in another language, you don't need to do that [16:51:47] I kinda also see the value in adding an explanation in said language instead of just the translation, though I'm not sure [16:54:13] e.g. if Spanish "gato" had the meanings "domestic mammal" and "particular type of person" (which are meanings of "cat" in English) then the English glosses on the two Spanish senses might be "cat (domestic mammal)" and "cat (particular type of person)" (re @Ege: I kinda also see the value in adding an explanation in said language instead of just the translation, though I'm not sure) [16:55:08] and hopefully we'd have enough data to be able to get to the corresponding senses on the english lexeme, if someone wants an explanation rather than a translation [17:07:23] guess it uses known and funny JS corner cases about using + with arrays (re @infovarius: Something like brainfuck?) [17:08:07] found this beauty also, 32768 long short descriptionhttps://www.wikifunctions.org/wiki/Z10123 : https://tools-static.wmflabs.org/bridgebot/6e42e2ba/file_50764.jpg [17:12:02] if there is a public domain dictionary with explanatory definitions in another language, it can be useful to add as a gloss quote as in L1136691-S1 (re @Ege: I kinda also see the value in adding an explanation in said language instead of just the translation, though I'm not sure) [17:12:54] yes, on the Turkish lexemes I've added senses to most have such a gloss quote on them [17:13:16] What's a gloss quote? [17:14:00] a quotation ("quote") of a dictionary definition ("gloss") [17:15:11] Oh yeah, I used that a lot for what I was doing, it's generally from the TDK, which is licensed. Sometimes I rephrase it but not always [17:15:58] e.g. on L728512-S1 or L311212-S1 [17:19:03] I find this dictionary ("Perso-Arabic Loanwords in Hindustani") to be a good example of the usefulness of gloss quotes in establishing a chronology for the use of a word : https://tools-static.wmflabs.org/bridgebot/0e884272/file_50765.jpg [17:21:38] I believe use of `parseFloat` and similar will be hurtful for the reasons I've described in https://www.wikifunctions.org/wiki/Wikifunctions:Project_chat#Let's_don't_hardcode_a_specific_locale_in_Wikifunctions (re @Nikki: what I mean is that if you can't use them as actual numbers (in something like if (foo < 1) { ... }, then using parseFloat isn't...) [17:25:03] I can explain more how that case applies here and what should be done instead but perhaps I'm overreacting and I can just let things go to what happens and then do something about [17:25:08] does that mean you think we should never support a real number type, only strings? [17:26:24] reverse way, when we are operating in real number we should have a clear boundaries between input string which is in a specific locale, and, real numbers type (re @Nikki: does that mean you think we should never support a real number type, only strings?) [17:26:56] but if it's a real number, people will be forced to use ascii digits, and that's what you want to avoid [17:27:44] that could be adjusted with some UI work (e.g. using some digit equivalences on the client side) (re @Nikki: but if it's a real number, people will be forced to use ascii digits, and that's what you want to avoid) [17:28:50] so that I could type ১২৩.৪৫ in the text box and before it gets sent to wikifunctions.org the number gets converted to 123.45 [17:29:03] but it still means someone has to make it do the conversion before calling the function, since the function itself won't understand other digits [17:29:20] maybe I should explain this way, there should be a clear boundary between logic layer, that uses real numbers, and presentation layer, that uses users or locale specific string (re @Nikki: but if it's a real number, people will be forced to use ascii digits, and that's what you want to avoid) [17:30:09] By the way I accidentally opened up the Lexeme discussion here instead of the Lexeme group, I noticed now [17:30:19] As far as I understand at the very moment the same is done for strings and bools - the function gets e.g. true and not Z41 [17:30:53] once a floating point datatype gets introduced, the little input boxes on a function page could have this digit conversion logic someplace, so that the need for parseFloat() and other equivalents thereof within Wikifunctions functions is removed (re @Nikki: but it still means someone has to make it do the conversion before calling the function, since the function itself won't underst...) [17:33:49] so it shouldn't matter if people are temporarily using parseFloat, because it's not intended to be the proper version of the function [17:35:23] indeed, somehow I'm overreacting [consciously] over this slippery slope because of things I've seen elsewhere (re @Nikki: so it shouldn't matter if people are temporarily using parseFloat, because it's not intended to be the proper version of the fun...) [17:36:11] say if someone was cared enough to make the decision happen with lua module initially introduced in mediawiki we hadn't the issues we have in our local wiki [20:38:49] I think it's good to point out this things now. It's the best time for it.