[01:02:30] I'm now reasonably happy with the accuracy of Z32668. But it takes over 5 seconds to determine that people aren't positions... Traversing hierarchies is costly. Maybe we can add a circuit-breaker for common classes. Or it would be even better if caching was in place for calls within calls (e.g. [human is not a subclass of position] can be cached and will be called on [01:02:30] for every specific person). [01:03:53] This will be called twice to obtain "Denny has a cat." to ensure it shouldn't be "The Denny has the cat." (re @u99of9: I'm now reasonably happy with the accuracy of Z32668. But it takes over 5 seconds to determine that people aren't positions... T...) [03:52:24] David or @vrandecic please can you set Z32758 as the equality function for Z31 [04:15:47] Al can you have a look at Z32765? I think it's failing to compare a language with a language reference. I feel like we've had this kind of issue before, but I thought it was resolved. Is this a v2 issue? [05:02:12] Z32749 (re @Al: I had a bit of a poke around and it seems that we might be able to rely on item aliases to determine whether the English label f...) [05:46:46] 0400 [05:46:53] that was weird [05:47:30] anyway hi I'm trying to figure out abstract wiki right now [05:48:06] does anyone have a good example of a fully functioned page that I can look at? [06:01:17] [[abstract:Q16]] (re @feedmepaperr: does anyone have a good example of a fully functioned page that I can look at?) [06:02:00] For some value of "fully" [06:02:12] I mean something a bit more substantial than just a single sentence [06:04:15] I'm working on [[abstract:Q17198982]] [06:04:29] I wrote the enwiki version of that page [06:10:11] I started to write [[abstract:Q711]], but yeah there aren't very many "language" functions yet from what I can tell. Most every article is in the for "X is Y". [06:12:48] seems like it, yeah [06:13:00] that's what we get for being early adopters [06:15:57] yup. i'm jumping in from enwiki with no experience with wikifunctions, so I'm trying to somewhat educate myself. very confusing as things stand, but I'm sure it'll be more clear as it becomes more accessible to people down the line [06:18:07] hopefully [06:22:11] not technically wrong but that can't be the only thing to say about birds 😭 : https://tools-static.wmflabs.org/bridgebot/87f7ca16/file_78922.jpg [06:27:20] better? :) : https://tools-static.wmflabs.org/bridgebot/342d749a/file_78923.jpg [06:33:45] yknow what, sure [07:02:13] As someone who has a *lot* of experience on Wikifunctions, you're not wrong. The other good news is that the AW rollout shows us what we need, and it's not necessarily what we expected. (re @M: yup. i'm jumping in from enwiki with no experience with wikifunctions, so I'm trying to somewhat educate myself. very confusing ...) [07:03:10] done (re @u99of9: David or @vrandecic please can you set Z32758 as the equality function for Z31?) [07:03:25] Awesome. I'm really glad this is being worked on. (re @u99of9: As someone who has a *lot* of experience on Wikifunctions, you're not wrong. The other good news is that the AW rollout shows us...) [07:14:22] I'm limited to English and infrastructure though, and there are much better English linguists than me. So I have typically focused on infrastructure. But now that I can see it, I'm dipping my toes into what I think AW requires of English. (re @M: Awesome. I'm really glad this is being worked on.) [07:16:54] Should there be a general, multilingual Z10771 or Z22511? I see a Japanese function adds full stops "。" manually as string. [07:18:07] btw I dont think the two should be used in real sentences [07:18:17] https://www.wikifunctions.org/view/en/Z22511?call=%7B%22Z1K1%22%3A%22Z7%22%2C%22Z7K1%22%3A%22Z22511%22%2C%22Z22511K1%22%3A%22iPhone+runs+iOS.%22%7D [07:28:49] I don't think this one needs to be general, because each language constructor can call the specific one used in their language. But I think it will eventually prove useful to have one for Japanese. And that's an excellent starting function if you haven't tried Wikifunctions much. (re @OverflowCat: Should there be a general, multilingual Z10771 or Z22511? I see a [07:28:49] Japanese function [07:28:49] adds full stops "。" manually as string.) [07:33:23] for example : https://tools-static.wmflabs.org/bridgebot/db714f3b/file_78927.jpg [07:35:13] looks like transform should be applied at the lexeme level [07:35:58] Then it looks like you want to make a more clever English one? (re @OverflowCat: for example) [07:36:39] I don't know -- is there even a property in Wikidata to specify the capitalized version of a word? [07:37:07] I leaned long ago not to start the title of your papers with a lowercase greek letter! (re @OverflowCat: for example) [07:41:29] Not that I'm aware. But you could improve your example by avoiding capitalising when there is already another capital in the first word. (re @OverflowCat: I don't know -- is there even a property in Wikidata to specify the capitalized version of a word?) [08:55:32] It’s not resolving the argument reference (or a function call) for the Z31K1 of the constructed Z31. I don’t know whether that’s specifically v2; I don’t think I ever tried it with v1. (re @u99of9: Al please can you have a look at Z32765? I think it's failing to compare a language with a language reference. I feel like we've...) [08:58:12] So instead I could echo the language into the construction? Then is it a bug in its current form? (re @Al: It’s not resolving the argument reference (or a function call) for the Z31K1 of the constructed Z31. I don’t know whether that’s...) [09:02:02] No, it doesn’t evaluate a function call there either. (re @u99of9: So instead I could echo the language into the construction? Then is it demonstrating a bug in its current form?) [09:02:52] So, yes… I’d say it’s a bug somewhere. (re @u99of9: So instead I could echo the language into the construction? Then is it demonstrating a bug in its current form?) [09:06:08] okay, I'll file a report [09:22:43] I can certainly see the sense in that when there’s retrieval from Wikidata… in fact I thought it was in place (for some reason) 🤔 David? (re @u99of9: I'm now reasonably happy with the accuracy of Z32668. But it takes over 5 seconds to determine that people aren't positions... T...) [09:30:56] *T421468* (re @u99of9: okay, I'll file a report) [09:31:15] the /view/en paths on Abstract Wikipedia have been re-configured, and a whole set of errors should be gone now [09:31:53] I've already trained myself not to click them! (re @vrandecic: the /view/{lang}/ paths on Abstract Wikipedia have been re-configured, and a whole set of errors should be gone now) [09:35:19] Sorry! [11:08:39] Is there a specific place for proposing new functions for abstract wikipedia or should I just be bold and create them myself? [11:14:07] Feel free to be bold! And feel free to create a page on-wiki for “desired outcomes” (I don’t know what you’d call it). It would be the AW equivalent of [[f:Wikifunctions:Suggest a function]], I suppose. (re @Gears: Is there a specific place for proposing new functions for abstract wikipedia or should I just be bold and create them myself?) [11:39:39] Great, thanks [11:52:03] I've created [[abstract:Abstract Wikipedia:Suggested functions]] [15:51:31] Not merely “a lot” but, by a long way, “the most”! (re @u99of9: As someone who has a *lot* of experience on Wikifunctions, you're not wrong. The other good news is that the AW rollout shows us...) [17:37:21] Well, no. We have caching of Wikidata entities that have been retrieved, but not of the results of nested function calls. There is a proposal for doing this in the context of the V2 composition language, when it's a bit more mature, and it's regarded as a relatively high priority. (re @Al: I can certainly see the sense in that when there’s retrieval from Wikidata… in [17:37:21] fact I [17:37:22] thought it was in place (for some reason)...) [17:55:32] Ah, yes… that’s what I was thinking of. It’s probably repeated traversals into code that’s mounting up, even though the Wikidata item is the same in each case. 🤔 (re @David: Well, no. We have caching of Wikidata entities that have been retrieved, but not of the results of nested function calls. Ther...) [18:16:00] I'd put you second. (re @Al: Not merely “a lot” but, by a long way, “the most”!) [22:36:35] Could a functioneer update the first two test cases of Z32825? The "profession" parameter just needs to be changed to be a list containing a single item reference rather than the reference itself. [22:42:41] ✅ Done (re @Gears: Could a functioneer update the first two test cases of Z32825? The "profession" parameter just needs to be changed to be a list ...) [22:43:11] Thank you. [22:47:59] If you'd like to request functioneer on [[WF:RFP]] I would support it, you have sufficient contributions on AWP and WF. (re @Gears: Thank you.) [22:48:36] Ah cool, I wasn't sure whether I met the criteria yet [22:53:05] There’s a 48-hour wait, so it pays to ask early 😎 (re @Gears: Ah cool, I wasn't sure whether I met the criteria yet) [22:53:12] Fair enough [22:53:15] Submitte [22:53:18] d [23:30:00] does there exist a plural form of Z32145 ? [23:38:20] No. (https://www.wikifunctions.org/w/index.php?search=%3A+%22Z8K2+Z89%22+%22Z17K1+Z6091%22&title=Special:Search&ns0=1) (re @wikilinksbot: Z32145 – sentence case section title from Wikidata label) [23:39:20] alr then i'll try my shot at making one. probably will fail [23:41:46] No reason why any such failure would be long lasting! 😏 (re @M: alr then i'll try my shot at making one. probably will fail) [23:42:15] :-) first time for everything