[07:02:52] is there a way to modify existing catalog at mixnmatch? [07:03:15] or do I have to reupload and delete (how then?) the original one? [07:12:34] Wesalius: you can ask Magnus Manske [07:12:48] it's his tool [07:13:03] what is the best way to contact him? [07:13:05] Wesalius: if there is a way to do that, I'm interested too [07:13:10] hmm twitter? [07:13:15] he answer on twitter [07:13:17] I dont use twitter [07:13:18] :/ [07:13:21] or his discussion page [07:13:27] why wouldn't that work? [07:13:47] I dont use twitter, I can do his talk page [07:14:24] yes, i meant, the best way to contact a Wikimedian is always their talk page no? [07:15:09] I guess... but I can imagine Magnus is bombarded by questions [07:15:15] since his tools are so popular [07:20:00] Harmonia_Amanda: you can subscribe to https://de.wikipedia.org/wiki/Benutzer_Diskussion:Magnus_Manske, he might answer there [07:21:28] I don't speak german, so I'm never on de... [07:21:34] but thank you :) [07:22:12] me neither, but I contacted him there in English [07:22:16] since its his home wiki [07:22:34] and his babel states en-3 [07:23:59] :) [07:47:30] I really don't understand Infovarius [07:49:28] ? [07:50:06] https://www.wikidata.org/w/index.php?title=Wikidata_talk:WikiProject_Names&curid=19546578&diff=392107151&oldid=391601104 [10:08:24] . [10:26:22] leszek_wmde: hey let's talk [10:26:41] Amir1: oh, yeah, I was distracted, sorry [10:27:13] no worries [10:27:16] No rush [10:27:36] leszek_wmde: tell me when you're done [12:29:10] * DanielK_WMDE waves [12:30:29] Lydia_WMDE, leszek_wmde, etc: I'm back, and I'll be catching up with my emails today. Let me know if there is anything urgent i should look into. I'll be in the office on Monday on my regular schedule. [12:35:16] DanielK_WMDE: yay! I think there is nothing super urgent. There are things we'd like to discuss with you on Monday or so, but no emergency stuff [12:48:31] leszek_wmde: cool. meanwhile, enjoy this nice summary of team dynamics: http://maximized.soup.io/post/691461428/Bild [12:49:00] just to keep everyone motivated ;) [12:51:04] DanielK_WMDE: we probably weren't going that fast as we would like but I am not thinking about buying a bar yet [12:53:47] Is Thiemo still in the office? [12:53:59] frimelle: ^ [12:57:01] hoo: he's working from home today [12:57:16] ah right, he often does taht Friday's [13:00:25] DanielK_WMDE: I guess you're super eager to get more stuff to look at on your plate today? :/ [13:01:00] I have a few things where we have different opinions yet or need more input which I would like you to look at [13:01:19] Probably okay to defer that to Monday, though [13:12:59] Amir1 leszek_wmde: I went a bit overboard there with the copying from mediainfo on the lexeme entity patch. did we decide how to proceed with forms, senses, lemma etc for now? [13:13:38] jakob_WMDE: we've got a basic sense on how to do it [13:13:41] my actual question: which ones of those should be included in the intial patch? just lemma? non of them? [13:13:56] lemma, language and lexical category [13:14:07] Thiemo_WMDE: Here? [13:14:07] senses and forms for later [13:14:11] alright, thanks! [13:14:11] jakob_WMDE: no forms, sense for now. lemma would be a term as https://phabricator.wikimedia.org/T148827, also other things mentioned there [13:14:17] hoo: yea. [13:14:29] Regarding EntityIdSiteLinkFormatter [13:14:52] I think we should move that to client [13:15:02] and override the client data type definition accordingly [13:15:26] Or is there a reason no to? I suggested that earlier but didn't get a reply to that suggestion [13:15:29] leszek_wmde: right, so in my patch i just leave them out for now? [13:15:38] jakob_WMDE: yes [13:16:00] leszek_wmde: alright, thanks! [13:16:02] Amir will add lemma etc. And forms and senses will follow later when we know more [13:16:17] got it! [13:19:30] Thiemo_WMDE: ^ [13:48:06] hoo: write me an email with links, i'll look at them tonight or over the weekend. [13:49:00] DanielK_WMDE: Ok, will do [13:49:28] leszek_wmde, Amir1: you added language and lexical category to the baseline? have you decided on how these will be represented? [13:50:22] lex cat should be an item ref. with the language, i'm not sure at all. [13:50:37] DanielK_WMDE: hey, We talked about it a little bit [13:50:47] DanielK_WMDE: well, so far we only said they' would be items. We had no good idea how the represent them really [13:51:28] you decided to use items als ofor the language? i like this conceptually, but i also see problems... [13:51:34] let's talk about that on monday [13:51:40] also, wil lthe lemma be mono-lingual or multi-lingual in the baseline? [13:52:19] i'd propose multi-lang, as terms usually are. We can still restrict it to a single language in the UI and API for now. But the model should be multi-lang I think, for forward-compat. [13:53:31] lemma should be monolingual, as in the proposal. Well, technically we thought it could be a term, so it would have a lang code, so it would be basically what you just said [13:53:54] DanielK_WMDE: I tried to follow the proposal which said the lemma should be just an string and language should be just one [13:54:04] or I misunderstood the proposal [13:55:06] DanielK_WMDE Amir1: gotta go offline now. Amir knows all we talked about so far, just in case. See you guys on Monday [13:55:39] o/ [13:59:35] Amir1: a lexeme is always tied to a single "language", but we may have different spellings or scripts for the lemma. that is still a bit unclear. For the baseline implementation, it does not matter much, but once we deploy, this should no longer change. [14:00:46] Amir1: for practical reasons, I suggest to treat the lemma as a Term object, which is text+language. It's ok for now to have just on Term Object, not a TermList, like we do for labels. [14:01:26] but we may end up having a TermList for labels, e.g. for swiss spelling of german words, or for roman and cyrillic representations of serbian words. [14:01:54] DanielK_WMDE: hmm, so you think we should start with TermList from now? [14:01:59] It's okay for me [14:07:45] DanielK_WMDE: Mail sent [14:13:12] Amir1: I think we should use Term definitly. Let's decide on TermList soon. [14:13:31] Okay [18:14:11] sjoerddebruin: I mixed up the weeks. Thought the Utrecht Wikidata thing was tomorrow :P [19:19:04] There's another office hour on potential upgrade to Creative Commons 4.0 is in #wikimedia-office now if you are interested! (It doesn't really impact you guys, but you might want to join anyways.) [19:22:50] so we'd be moving from CC-0 to CC-4.0 ? :p [19:23:03] * Alphos ducks, covers, and chuckles :p [19:27:23] Alphos: Haha, good one [19:29:24] Alphos: wow, so we’re way behind the Wikipedias? Why are we using an outdated version? ;) [19:31:07] Alphos, ;) [19:35:47] WikidataFacts we'll get to have 4.0 when we've settled the disambiguation pages vs name entities, and when we have a clean ontology for french subdivisions. til then, GO BACK TO YOUR ROOM, NO DESERT FOR YOU ! [19:36:33] but by then, version 5.0 will be out :( [19:45:22] only 5.0? that seems optimistic :P [19:51:09] well, 3.0 held up for almost nine years, and GPLv3 has been going strong for a bit longer – I’d say licenses seem to be maturing