[01:08:11] Meanwhile I found a way to make a new type of function-checking-function, (example here Z13419), but unfortunately the orchestrator is too slow in even the simplest cases. [01:52:01] And I actually miss a proper system to track Wikipedia articles that don't yet exist or that have issues. Wikipedia has backlog categories, and some people manage wiki pages or spreadsheets of missing articles, but it's not really a _proper_ system. (re @Jan_ainali: Yes. And the same applies to Wikipedia articles, I guess.) [01:52:47] But I don't miss it as much as I miss a proper system for managing missing Wikidata label translations or lexemes. [02:59:14] Can I get my functioneer status extended? It’s expiring soon and I’ve placed a request on wiki [03:19:14] Are all wikifunctions stateless functions? [03:37:17] Yes I think so. (re @wmtelegram_bot: Are all wikifunctions stateless functions?) [03:37:31] kk [03:38:42] I just re-read https://www.wikifunctions.org/wiki/Wikifunctions:About . Can anyone tell me how this is done: "every implementation can be annotated with metadata and cryptographically signed" [03:47:01] That page also fails to mention that WF is supposed to be multilingual (not just in the programming language sense). When I gave a brief talk about WF in my department, my title was "a language-independent programming language" which I think is a cute way to think of it (although I know it conflicts with https://www.wikifunctions.org/wiki/Wikifunctions:What_Wikifunctions_is_not) [08:32:58] I did the bold and edited the page to include references to multilingualism (re @Toby: I just re-read https://www.wikifunctions.org/wiki/Wikifunctions:About . Can anyone tell me how this is done: "every implementati...) [09:02:45] The Abstract Wikipedia primary goals are more explicit: [09:02:45] [09:02:46] • Allowing more people to read more content in the language they choose. [09:02:48] • Allowing more people to contribute content for more readers, and thus increasing the reach of underrepresented contributors. [09:02:49] [09:02:51] The problem with these words, in the context of Wikifunctions, is that contributors can freely choose the human language of their contributions, whereas the choice of programming languages will always be less universal. Nevertheless, these goals very much apply to Wikifunctions and could usefully be re-stated here. I may be a little bold myself later… and I note a “secondary goal”: [09:02:52] “Allowing the development and sharing of functions in the user’s native languages, instead of requiring them to learn English first.” [09:02:54] [09:02:55] https://meta.wikimedia.org/wiki/Abstract_Wikipedia/Goals (re @Sannita: I did the bold and edited the page to include references to multilingualism) [09:06:37] I made a small change, because a more extensive rewrite would have required community consensu [09:07:14] I'm not against it, it's just that I thought of a simple fix, while waiting for some more extensive rewrite [09:14:47] Yes, it is a good small improvement. Thank you. And I intend to be only a little bold myself. (re @Sannita: I'm not against it, it's just that I thought of a simple fix, while waiting for some more extensive rewrite) [09:19:46] Please can we have a volunteer's corner sometime that doesn't start at 5:30am Sydney time? [09:21:17] I think it's too late for March, but we can try to find a solution for April [09:21:39] I was already evaluating that [11:04:44] Tweaked: https://www.wikifunctions.org/w/index.php?title=Wikifunctions:About&curid=2771&diff=83811&oldid=61147 [11:24:37] in fact, I see it's already there: https://www.wikidata.org/wiki/Lexeme:L31687#S1 (re @Nicolas: In the sense, you link to the item for cat) [11:26:29] It was just an example. (re @Nicolas: in fact, I see it's already there: https://www.wikidata.org/wiki/Lexeme:L31687#S1) [11:40:26] sure, but it already works in most the cases [11:40:27] [11:40:28] and then you can do a SPARQL querry (like that for instance https://w.wiki/9JYh ) or even a Wikifunctions (not yet, but soon-ish hopefully) [11:43:27] It is an interesting example, because we see a picture of a common stingray (a different sense, apparently) and that Wikidata item appears to have no label in Russian (although it links to the appropriate article in the Russian Wikipedia). It would be interesting to know whether any Punjabi senses link to items without Punjabi labels 🤔 (re @amire80: It was just an example.) [11:59:21] quite a few, adding labels can be tie consuming but I'd like to get around to adding more at some point. also the pnbwiki people like to blank the Wikidata labels whenever they link an article (re @Al: It is an interesting example, because we see a picture of a common stingray (a different sense, apparently) and that Wikidata it...) [12:11:33] or they seem to go through phases of it at least certain topics seem to be deemed more contentious than others [12:14:52] That image would probably be correct only if there's another word before it. I'm trying to find the policy about multi-word lexemes, and I'm failing. (re @Al: It is an interesting example, because we see a picture of a common stingray (a different sense, apparently) and that Wikidata it...) [12:15:41] Are the equal sign on sense 3 and 5 correct or are these copy-paste errors? (I don't know Russian at all.) (re @Al: It is an interesting example, because we see a picture of a common stingray (a different sense, apparently) and that Wikidata it...) [12:16:55] One lemma can have multiple senses if they have the same etymology and forms. (re @amire80: That image would probably be correct only if there's another word before it. I'm trying to find the policy about multi-word lexe...) [12:18:15] It's like "lion" and "sea lion" in English. Should there be an image of a sea lion on the "lion" lexeme? [12:18:22] I don’t know Russian either, I’m afraid. The word is used in the Wikipedia article, but whether it is legitimate to use it unqualified for “stingray” I cannot say. (re @Jan_ainali: Are the equal sign on sense 3 and 5 correct or are these copy-paste errors? (I don't know Russian at all.)) [12:18:38] Maybe there should be, but I can't find the policy. [12:19:31] this is generally language-specific and there is a general lack of interest in working on English lexemes and documenting them at the moment [12:21:01] No, "sea lion" should be a separate lemma using P5238 of L3578 and L17815 (re @amire80: It's like "lion" and "sea lion" in English. Should there be an image of a sea lion on the "lion" lexeme?) [12:25:13] An example of the implemetrability I spoke about yesterday: it's very hard to read messages with P, Q, and L numbers. The bot does not help much, because I need to go back and forth between two messages. Is it really so hard to write it like this: [12:25:14] [12:25:15] No, "sea lion" should be a separate lexeme using "combines lexemes" (P5238) of "sea" (L3578) and "lion" (L17815)? (re @Jan_ainali: No, "sea lion" should be a separate lexeme using P5238 of L3578 and L17815) [12:25:41] Policy wise, and in English for example, there shouldn't be an equal sign there, but as I don't know Russian, for all I know it is a regular letter. (re @amire80: Maybe there should be, but I can't find the policy.) [12:26:19] This happens a lot in discussions about Wikidata and Lexemes and Wikifunctions. Evidently, lots of people here find it convenient. But not everyone. (re @amire80: An example of the implemetrability I spoke about yesterday: it's very hard to read messages with P, Q, and L numbers. The bot do...) [12:37:15] It’s more convenient on-wiki, where the language-neutral IDs can resolve according to the current language. In an English-language message, it is perhaps better to use the English labels, but the writer may not actually know what they are. (re @amire80: This happens a lot in discussions about Wikidata and Lexemes and Wikifunctions. Evidently, lots of people here find it convenien...) [12:43:11] Raised in https://phabricator.wikimedia.org/T358571 (re @vrandecic: That's weird. I thought the value would be key label for Z14K2, Z14K3 and Z14K4. I think that might be a bug.) [12:43:43] I forgot to mention Z numbers, which are the worst. (re @amire80: An example of the implemetrability I spoke about yesterday: it's very hard to read messages with P, Q, and L numbers. The bot do...) [12:44:19] Do we know why they like to blank the labels? (re @bgo_eiu: quite a few, adding labels can be tie consuming but I'd like to get around to adding more at some point. also the pnbwiki people...) [12:47:55] I can only guess, but for example in this case they seem to have decided this dialect of Sindhi shouldn't exist. since none of the mother tongues of Pakistan have official status people can get very invested in these kind of language disputes https://www.wikidata.org/w/index.php?title=Q61137225&oldid=1926016670 [12:47:55] [12:47:57] (Sindhi forms a dialect continuum with Punjabi) (re @Al: Do we know why they like to blank the labels?) [12:48:32] Deleting the label sounds like vandalism to me. [12:51:16] sure, I agree. what I'd like to see in the future is a merger with pawiki so that the space to have these kinds of discussions exists [12:51:51] that's something that lexemes have the potential to make a lot easier [13:12:23] If Wikidata always acquired or presented the lemma of a linked sense as an alias, would that make the problem better or worse? (re @bgo_eiu: I can only guess, but for example in this case they seem to have decided this dialect of Sindhi shouldn't exist. since none of t...) [13:13:13] better certainly (re @Al: If Wikidata always acquired or presented the lemma of a linked sense as an alias, would that make the problem better or worse?) [13:13:57] Is there a Feature Request for that? (re @bgo_eiu: better certainly) [13:21:29] I don't think so but in this case adding languages other than English as fallbacks for Punjabi on Wikidata, which I requested a year and a half ago, would be more immediately useful (re @Al: Is there a Feature Request for that?) [13:23:41] And simpler too… is the obstacle technical or political? (re @bgo_eiu: I don't think so but in this case adding languages other than English as fallbacks for Punjabi on Wikidata, which I requested a ...) [13:24:54] no idea, it is not clear who has the ability to make this change on Wikidata [13:26:09] Is the Phabricator ticket not assigned to anyone? (re @bgo_eiu: no idea, it is not clear who has the ability to make this change on Wikidata) [13:27:30] Doesn't look like it https://phabricator.wikimedia.org/T317152 [13:32:58] Perhaps someone with authority would like to subscribe to that 🙄 (re @bgo_eiu: Doesn't look like it https://phabricator.wikimedia.org/T317152) [14:45:30] I added comments to https://phabricator.wikimedia.org/T301712 and https://phabricator.wikimedia.org/T343565 for now. (re @vrandecic: Ah, yeah, that's right, we're missing a progress dashboard. I sure can see progress being done, but yeah, a dashboard would be n...) [15:28:21] The Best Practices document says: [15:28:21] “In general, while individual words that aren't merely inflections of other words might warrant lexemes, non-idiomatic phrases typically do not warrant them, since they may be treated as the sum of their parts. (https://en.wiktionary.org/wiki/en:WT:SOP) [15:28:22] ◦ This does not necessarily discount the addition of non-idiomatic meaning senses to lexemes which do have idiomatic meanings, however, and which have those idiomatic meanings as senses already.” [15:28:24] [https://www.wikidata.org/wiki/Wikidata:Lexicographical_data/Best_practices] [15:28:25] [15:28:27] Thus the English “black cat” should not be a lexeme (and is not) whereas “red herring” should be [L584591 but there should be no hyphen in the lemma] (re @amire80: Maybe there should be, but I can't find the policy.) [18:16:01] @mahir256 I love it when the magic “just happens” 🙏 Thank you! https://www.wikidata.org/w/index.php?title=Lexeme%3AL584591&diff=2088783313&oldid=1482073046 [22:27:39] So then IMO "sea lion" should get it's own, because it's not the sum of the parts. It's not a lion that lives in the sea. (re @Al: The Best Practices document says: [22:27:39] “In general, while individual words that aren't merely inflections of other words might warran...) [22:35:21] Thanks @mahir256 https://www.wikidata.org/wiki/Lexeme:L1267725 (re @Toby: So if I understand this right, "sea lion" should get it's own, because it's not the sum of the parts. It's not a lion that lives...) [23:34:41] New topic. Perhaps I've misunderstood, but it seems like the order of arguments chosen for a function specification actually matters. In particular the first argument seems special, in that when the function is used for result validation in a test (of another function), the test result can only be fed into the validating function as the first argument. This limitation might mean [23:34:42] that either: when we set up a function, we need to choose carefully which argument should be first; or we end up making multiple versions of a function with the order switched (as was done here: "contains" Z12696, vs "is listed in" Z13381) [23:37:53] (Discussion prompted here by Al https://www.wikifunctions.org/wiki/Talk:Z13381) [23:53:00] Yes… I wouldn’t want to make a habit of it! But I don’t think that “thinking carefully” is necessarily the answer (just this once) 😏 (re @Toby: New topic. Perhaps I've misunderstood, but it seems like the order of arguments chosen for a function specification actually mat...)