[01:00:45] I tried it out and got unexpected error messages at Z22154 : https://tools-static.wmflabs.org/bridgebot/f4a56691/file_68390.jpg [01:02:50] I guess it's to do with how Python tries to deal with a list of Kleeneans at Z20683 (re @u99of9: I tried it out and got unexpected error messages at Z22154) [07:24:05] Was doing some tedious homework so made a wikifunction ([[Z22155]]) to automate it : https://tools-static.wmflabs.org/bridgebot/2cc95161/file_68392.jpg [07:25:08] Ironically making the function actually helped me understand the underlying concepts a lot better than just repeating it over and over!! [07:29:22] I think making Wikifunctions makes you have to think about something in a kind of computational and logical kind of way using a problem solving mindset [07:38:31] Re kleenians I think for NOT false should be true, true should be false and maybe should be maybe, for AND f+f=f f+m=f f+t=f (so anything with false is automatically false) m+m=m and t+t=t, t+f=f and t+m=m (so you need 2 trues to get a true result, otherwise it's false or maybe) [07:40:57] Or is a bit mORe interesting but I think anything with true is automatically true (so t+t=t, t+m=t, t+f=t), m+m=m and to be false you need 2 falses, so f+f=f but f+m=m and f+t=t [07:42:42] Nor should be exactly the same as or reversed so if OR gives true, return false, if OR gives false, return true but if it gives maybe just return maybe [07:43:54] What other logic gates are there to cover?? Only XOR comes to mind but I don't have much experience with physics/electrical engineering [07:54:12] @wiktionary is the telegram chat room for wiktionary projects [08:21:14] This isn't about wikitonary??? There was a discussion on kleenians above sorry if I was confused [08:21:24] About making functions involving kleenians [08:38:32] This is all correct. (re @MolecularPilot: Re kleenians I think for NOT false should be true, true should be false and maybe should be maybe, for AND f+f=f f+m=f f+t=f (so...) [08:42:44] There are 19683 possible binary functions of Kleeneans! Your theorising matches the descriptions at https://en.wikipedia.org/wiki/Three-valued_logic , so that is worth a read. (re @MolecularPilot: What other logic gates are there to cover?? Only XOR comes to mind but I don't have much experience with physics/electrical engi...) [08:50:39] Just wait until someone makes you a test for Chromium or Niobium or Molybdenum or ... (re @MolecularPilot: Ironically making the function actually helped me understand the underlying concepts a lot better than just repeating it over an...) [09:21:56] Speaking of context-free celebration due to passing random milestones - I just got thanked by the system for making my 10,000th edit! (re @vrandecic: Newsletter #187: [09:21:56] * With 2000 Functions into the new year: time for stats [09:21:56] * Next Volunteers’ Corner on 3 February [09:21:58] * Recent Change...) [09:35:18] Your random contributions are much appreciated! (re @u99of9: Speaking of context-free celebration due to passing random milestones - I just got thanked by the system for making my 10,000th ...) [09:43:22] It wasn't an on topic message, it was just to announce the birth of another channel (re @MolecularPilot: This isn't about wikitonary??? There was a discussion on kleenians above sorry if I was confused) [09:45:08] Yes. I found the fix. N-ifs doesn't work if the second list is not object-typed. (re @u99of9: I guess it's to do with how Python tries to deal with a list of Kleeneans at Z20683) [09:51:31] That would do it! To be fair, though, it’s the type inference heuristic that is understandably baffled. (re @u99of9: Yes. I found the fix. N-ifs doesn't work if the second list is not object-typed.) [09:54:09] It would be good to create one by one (re @u99of9: There are 19683 possible binary functions of Kleeneans! Your theorising matches the descriptions at https://en.wikipedia.org/wik...) [09:54:54] Not birth, just announcing its existence: I haven’t known for long (re @Sannita: It wasn't an on topic message, it was just to announce the birth of another channel) [09:55:13] Ha! You'll beat me to 100,000 with that zeal! I was interested that they don't even want to bother categorising them. (re @cvictorovich: It would be good to create one by one) [09:55:37] Similarly, where Boolean logic has 22×2 = 16 distinct binary operators (operators with 2 inputs) possible, ternary logic has 33×3 = 19,683 such operators. [09:55:59] Found this piece of text [09:58:31] 🤔 or is that fair? In any event, Z1-typed lists are in the “unsupported” category for type conversion 🤷‍♂️ (re @Al: That would do it! To be fair, though, it’s the type inference heuristic that is understandably baffled.) [09:59:40] Does that even make sense, mathematically? (re @cvictorovich: Found this piece of text) [10:00:23] I started replying but got too confused to express my confusion. I resolved to figure this out later and reply to you then 😁 (re @Al: 🤔 or is that fair? In any event, Z1-typed lists are in the “unsupported” category for type conversion 🤷‍♂️) [10:03:21] I’ll just 22114 you to avoid upsetting wikilinksbot 😎 (re @u99of9: I started replying but got too confused to express my confusion. I resolved to figure this out later and reply to you then 😁) [10:04:17] I don't understand why Z22189 should be possible in code, but Z22183 fails. We are providing the same information in both cases. Perhaps there is a code conversion improvement that can be made? [10:06:07] Very likely: just like Boolean (re @Al: Does that even make sense, mathematically?) [10:14:39] What would be the “improvement”? Changing the type of the list is just wrong, in my opinion. (re @u99of9: I don't understand why Z22189 works in code, but Z22183 fails. We are providing the same information in both cases. Perhaps ther...) [12:09:11] Todo: I’ll “explain” that there, sooner rather than later 🤔 (re @u99of9: I don't understand why Z22189 works in code, but Z22183 fails. We are providing the same information in both cases. Perhaps ther...) [12:48:54] 2^(2*2) and 3^(3*3) (re @Al: Does that even make sense, mathematically?) [12:56:53] I've pitched a Kleenean suggestion Z22143 for FOTW. Shorter and more typical than last time! https://www.wikifunctions.org/w/index.php?title=Wikifunctions%3AFunction_of_the_Week%2Fsubmissions&diff=162713&oldid=162702 [13:54:26] Thank you @u99of9 , that's a beautiful write-up! [13:57:27] Also, perhaps some type converters could be changed to show better errors (and indicate that it's the type converter doing it for what type, programming language, and input) (re @Feeglgeef: I think Error messages are a very important quality thing. A lot of functions either end up throwing an Error that nobody can un...) [15:03:23] Same for result validation functions [15:03:24] *T383326* [15:03:25] Hmm… child of *T385081* (re @Feeglgeef: Also, perhaps some type converters could be changed to show better errors (and indicate that it's the type converter doing it fo...) [15:54:26] [I’ve challenged this on Phabricator, just in case. To me, it seems more like a nested metadata deficiency.] (re @Al: Same for result validation functions [15:54:27] T383326 [15:54:28] Hmm… child of T385081) [16:34:15] Thanks, that makes mathematical sense now. 😎👍 (re @cvictorovich: 2^(2*2) and 3^(3*3)) [16:35:27] Forwarded from Nicolas: I disagree, why make an expensive call to Wikidata lexemes? [16:35:28] if anything, I would rather wait for the ability to call Wikifunctions (and even then, for simple regular conjugation, the gain is minimal ; could be very useful for Wiktionaries that don't have such model tho) [16:37:18] If only wiktionary is connected to wikifunctions… [17:45:30] be patient, it will come soon enough ;) (I have idea for Wikisources too) (re @cvictorovich: If only wiktionary is connected to wikifunctions…) [17:50:40] Details? [18:03:23] if you want detail, follow for instance (other Wikimedia project will come after this) (re @cvictorovich: Details?) [19:09:26] The recording of yesterday's Volunteer's Corner is now available on Commons: https://commons.wikimedia.org/wiki/File:Abstract_Wikipedia_Volunteer_Corner_2025-02.webm [22:23:11] “Topical”, even… on the question of practical uses of Kleenean, I’m reminded of *T373607, *which is probably a case of “not false” not being “true”. (re @u99of9: I've pitched a Kleenean suggestion Z22143 for FOTW. Shorter and more typical than last time! https://www.wikifunctions.org/wiki...) [23:27:36] I was trying to explain this on-wiki when this came up. [23:27:37] https://www.wikifunctions.org/wiki/Wikifunctions:Project_chat#Unable_to_serialize_Natural_Numbers? [23:27:39] It’s too early to say whether this will end up superseding the explanation I was failing to come up with! (re @u99of9: I don't understand why Z22189 works in code, but Z22183 fails. We are providing the same information in both cases. Perhaps ther...) [23:30:36] Yes. It's better to help the newcomer. I will probably (re-) learn something from it anyway. I know enough to get around this technically, but it's a big effort, and I feel like it shouldn't be! [23:35:05] The blurb at the top of https://www.wikifunctions.org/wiki/Wikifunctions:Catalogue/List_operations#Functions_with_list_outputs is where I currently go to remind myself how it works. Any improvement there would be appreciated. [23:41:47] 🤔 I think I wrote that… I wonder what it means! 😵‍💫 (re @u99of9: The blurb at the top of https://www.wikifunctions.org/wiki/Wikifunctions:Catalogue/List_operations#Functions_with_list_outputs i...) [23:46:20] yup https://www.wikifunctions.org/w/index.php?title=Wikifunctions%3ACatalogue&diff=120310&oldid=120064 (re @Al: 🤔 I think I wrote that… I wonder what it means! 😵‍💫) [23:59:59] I’ve understood what I meant, and I think it’s correct. It will become clearer! 😱 (re @u99of9: yup https://www.wikifunctions.org/w/index.php?title=Wikifunctions%3ACatalogue&diff=120310&oldid=120064)