[14:05:47] Hi, all! How is Listeria working? [14:06:05] Why [[Wikidata:WikiProject Movies/Numbers/Most frequent P31]] aren't updating? [14:06:08] 10[1] 10https://www.wikidata.org/wiki/Wikidata:WikiProject_Movies/Numbers/Most_frequent_P31 [14:24:27] Hey everyone. :) [16:14:13] Lucas_WMDE: if/when you are not working, do you have any ideas for queries to show biology teachers? [16:14:40] I'm doing a presentation and I wanted to shpw them data they can relate to… [16:15:30] hm, we should have lots of interesting data for that, the problem is I don’t know much about biology so I can’t ask the interesting questions ;) [16:15:36] but I’ll try to think of something this evening [16:15:45] * Lucas_WMDE adds reminder to calendar [16:15:57] thank you :) [16:16:29] I actually don't know what data we have [16:17:09] we have tons of genes but do we have enough data to show correlations between genes and illnesses now? [16:17:33] I think someone had a query for that already [16:17:36] a huge graph in two colors [16:18:00] and one question I was asked is to show how a gene is present or not in the world [16:18:07] but I don't think we have that [16:18:25] I've never seen a human with its genes listed… [16:18:41] I know this data exists, just not on Wikdiatz [16:19:01] Wikidata* (wow that's a new one) [16:19:22] and of course they are asking me today for tomorrow [16:19:38] my fault I guess for saying I could do a basic presentation any time [16:22:00] hehe :) [16:22:07] but I’m not sure what it means for a gene to be present or not in the world [16:22:14] (which I can, but I can't think of cool working queries on any subject on the fly) [16:22:22] in populations [16:22:24] like, synthetic genes that aren’t found in any organisms? or only in lab rats? [16:22:25] like [16:22:39] this gene is present in 70% of Asian people [16:22:55] but only in 3% of Australians [16:23:39] oh, I see [16:23:46] question was how to model that, not just how to query for it [16:23:49] interesting [16:24:22] perhaps “prevalence” (P1193) with a slightly widened scope? (currently limited to diseases, it seems) [16:24:36] yes probably [16:24:54] it seems there’s some precedent for “country” qualifiers on that property [16:25:00] “ethnic group” could also work [16:26:38] I remember an problem about sickle-cell disease which showed that the sickle-cell trait was good against malaria [16:26:54] so more people had the sickle-cell disease in regions with malaria [16:27:31] I think that the kind of data they want [16:27:43] but I'm not sure it's the kind of data we have [16:38:37] * reosarevok kinda sorta agreed to lead a wikidata workshop. What have I done [16:39:07] Well, guess telling people "here's what I've been adding to Estonian things, now let's all do it together for an hour" is not that complicated :p [16:39:36] reosarevok: three years ago I said "hey wouldn't it be really cool if we made regular Wikidata workshops so people can know when they are and try to come?" [16:40:01] and now I'm doing a workshop every month in Paris for nearly two years [16:40:15] and I still have no idea how that came to pass [16:40:19] We (WMEE) just started that. Turns out when they suggested it, they forgot to mention that I'm probably the only person there who has been playing with WD enough [16:40:44] reosarevok: well in Paris we have… you [16:41:03] I think it went something like that for me too [16:41:14] So I guess that's me :p How do you start? Do you give a quick intro every time, or is it like "ok, this is event #2, we expect people to know the basics, if not come and talk to me"? [16:41:30] depends [16:41:42] Hello there! The IRC office hour will start in 20min on #wikimedia-office [16:41:43] now I mostly have people who came once or twice already [16:41:49] hi Auregann_WMDE! [16:42:06] Lydia and I will give news from the development team, and then we'll have an open discussion about strategies to address the growth of Wikidata [16:42:18] reosarevok: I adapt my speech depending on who is here [16:43:08] if they are starting from scratch, if they are semantic web workers who just don't know Wikidata, if they are Wikipedians who don't know anything about data… [16:43:21] it will really depends [16:43:27] :) Sounds reasonable [16:43:34] And what do you usually edit? [16:43:42] Names? :D [16:43:43] reosarevok: what they want [16:43:49] no, names are complicated [16:44:02] people are easier [16:44:13] or paintings [16:44:44] but mostly right now I work with projects like Women in Red or Les Sans Pages and make them work on women entries [16:44:52] because it's easy to do and useful too [16:44:54] BTW Harmonia_Amanda is there a news schedule of workshops for 2018? :) [16:44:56] Yeah. I was thinking of throwing the list of "items without statements and with etwiki sitelinks" at them, but they'll probably find a ton of crap that is hard to define, because it's hard for me :D [16:45:13] Auregann_WMDE: every third Friday of the month [16:45:19] Auregann_WMDE: until June [16:45:39] Auregann_WMDE: by the way, I have a workshop in Lausanne February 24th [16:46:19] reosarevok: you could work with Olympic Games data? [16:46:27] reosarevok: people like feeling relevant [16:46:44] and we will need some clean data like right now [16:47:03] reosarevok: things we are missing, like who coached who [16:47:10] That's why I'm thinking Estonian stuff specifically. I really should convert a couple templates to WD in etwiki [16:47:28] And then be like "look how it changes!" [16:47:30] :D [16:47:46] reosarevok: I think I identified all Estonian skaters with an ISU identifier [16:47:58] reosarevok: but I'm definitely missing who coached them [16:48:00] Harmonia_Amanda: Cool! I think it would be great to add the dates here https://www.wikidata.org/wiki/Wikidata:Events/Paris so people are sure about what date it is [16:49:18] Oooh and I just saw the event in Estonia =) is that you taking care of this reosarevok ? [16:50:19] Auregann_WMDE: done [16:50:33] Unfortunately it's too late for the newsletter (published on Mondays) but I'm going to tweet about this [16:50:38] Harmonia_Amanda: thanks :) [16:50:48] reosarevok: and we are missing birthdates and birthplaces [16:51:09] because it's not on WP and I can't understand Estonian texts [16:53:25] reosarevok: I don't know how Estonian Wikipedians took Wikidata but here in France I'm like always explaining that we are here to do human work, not bot work, and there are tools and gadgets to help us and to things for us [16:54:01] otherwise I have people adding labels manually, language by language, instead of using NameGuzzler [16:54:22] (no wonder they don't like editing Wikidata) [16:58:24] reosarevok: there is work to do about Estonian Olympians, I confirm [16:58:44] pretty sure you could do something wonderful with that [17:00:50] reosarevok: and you can have both easy cases to make them feel immediately useful and difficult problems to make them understand how we deal with complicated cases (like uncertain dates) [17:42:11] So much activity. :O [17:43:01] sjoerddebruin: I'm still in the state when I see the calendar and I'm like "I definitely need these properties before then" but then I don't ask for their creation [17:49:24] Auregann_WMDE: it's "us", I didn't suggest the event but will definitely help run it :) [17:49:53] Harmonia_Amanda: I really should look into NameGuzzler, shouldn't I :D [17:50:06] (I've been mostly ignoring missing labels, to be fair, not adding them one by one) [17:51:14] reosarevok: oh sweet summer child [17:52:00] reosarevok: you and I need to go to the same event once time and I'll take three hours to get you up to date to all marvelous tools and gadgets you need to know [17:52:19] Auregann_WMDE: I really need to do this workshop for confirmed Wikidatians [17:52:20] Sounds like a plan :) [17:52:24] like, it's a necessity [17:52:26] a vital one [17:54:27] Auregann_WMDE: and you know we have gadgets which don't even work anymore in the official gadget list in preferences? [17:54:48] Auregann_WMDE: we need a process to delete gadgets from the list not just for adding new ones [17:55:06] Harmonia_Amanda: indeed we do :/ [17:55:23] but should we delete immediatly the gadgets from the list? try to contact the creator first? [17:55:36] who created Ancestors? [17:55:57] the first version (officially listed) had not worked for months, maybe year [17:56:22] I don't know [17:56:23] Ancestors2 which was the fallback (not officially listed) has not worked for months either [17:56:34] OK, thanks for noticing [17:56:52] As far as I know there's no interdiction to remove gadgets from the list [17:56:57] and easyquery is still full of bugs [17:57:14] and made really less relevant now that we have a decent queryhelper [17:57:24] oh, while we’re on the topic of easyquery :) [17:57:31] https://www.wikidata.org/wiki/MediaWiki_talk:Gadget-EasyQuery.js#Use_wikibase-snakview-indicators – any admins who want to make that edit? :D [17:57:48] fixes at least one bug :P [17:57:58] it "works" but you need to understand SPARQL to make anything with it, so it's really more confusing than useful [17:58:26] * Harmonia_Amanda is tired of explaining to newbies that it doesn't handle dates, or qualifiers, or anything not really basic [17:58:32] Lucas_WMDE: did you actually refresh that? [17:58:51] sjoerddebruin: wat [17:59:01] I’m pretty sure I looked before I posted my second message [17:59:10] did someone fix it between that and me posting it in IRC? :D [17:59:22] ok, thank you d) [17:59:24] * :) [17:59:36] No, it was on my watchlist for some reason. [18:00:02] I'll go buy something to cook for tonight, write my name if you want me to answer you when i come back [18:12:28] so I come back and nobody has said a word [18:12:42] way to make me feels like I'm the only one talking here [18:14:48] Haha, I was there! [18:14:57] Your supermarket is close though. :O [18:15:22] I have a supermarked just across the road [18:15:30] which mostly have crisps [18:15:37] Healthy. [18:15:40] really limited choices [18:16:00] but still, I found pasta so [18:16:53] sjoerddebruin: and I don't even like 90% of these crisps [18:17:00] :O [18:17:14] I'm always wondering how they are surviving with only this [18:17:33] but there is a big one like a quarter hour from here [18:17:53] so I guess they mostly sell to people who are organizing a party last-minute [18:18:52] How are the opening times then? [18:19:18] 'till 22 pm [18:19:30] which would support the last-minute party theory [18:23:11] Now I want to go to the supermarket.... [18:23:50] But nah [18:23:52] sjoerddebruin: to find if yours to sell bad crisps? [18:23:57] too* [18:24:33] Nah, I want sweets. [18:26:10] so [18:26:19] I really really need to ask for these properties [18:26:28] that's my goal tonight [18:26:34] * Harmonia_Amanda will try [18:51:31] (["precision"] = 7, ["time"] = "-0760-00-00T00:00:00Z") => "8. century BCE" - what does -760 means in this case (rather than -800 with century precision)? [18:52:54] for proper formating is it correct to if precision=century => floor(year/100)? [18:58:34] how is it possible that someone set this value? is it a bug or a feature? [18:58:35] the string itself can contain anything, but with precision set to century the rest will be ignored for display [18:59:19] the API does not normalize datetime values afaik so they can be overspecified [18:59:54] IMHO it’s a feature, you can use it to get a more suitable value for the RDF export (especially the truthy statements) [19:00:10] if someone died 1367–1371, the best precision you can set is “century” [19:00:45] but by overspecifying the date to something like 1369 you’ll get a reasonable date in the RDF export, and it won’t look like the person died before they were born or something like that [19:01:11] you can even set that in the normal editing UI btw, edit a date as you normally would and then manually adjust the precision [19:01:24] sjoerddebruin: Lucas_WMDE https://www.wikidata.org/wiki/Wikidata:Property_proposal/Technical_Element_Score https://www.wikidata.org/wiki/Wikidata:Property_proposal/Program_Component_Score https://www.wikidata.org/wiki/Wikidata:Property_proposal/Deductions_(figure_skating) [19:02:10] Lucas, this is good point. thanks! I wasn't aware to this option :) [19:02:40] it’s kind of a dirty secret ;) [19:02:42] but I like it [19:02:49] though I don’t think many people use it like that [19:06:11] Harmonia_Amanda: is “number” the right datatype? on the help page it seems to link to “integer”, which doesn’t sound right if you need two decimal digits [19:06:33] Lucas_WMDE: what would be the right datatype then? [19:06:48] quantity, I guess [19:08:06] Lucas_WMDE: what is the datatype of https://www.wikidata.org/wiki/Property:P1351 ? [19:08:12] because this one work [19:08:24] Quantity [19:08:36] ok, I'll correct the demands [19:09:00] I think I’ve seen several other proposals use “number” too… I guess the property creators just use quantity instead? [19:09:16] anyways, I’m going home now, see you later for the biology queries :) [19:09:30] I went shopping and got free flowers, kek. [19:11:37] corrected [19:11:57] sjoerddebruin: just in time for saying that you want more precise skating data on WD [19:12:11] * Harmonia_Amanda pushes sjoerddebruin to the properties proposal page [19:12:19] That hurts! [19:12:50] I'm not sure, this channel has public logs so it could be seen as canvassing :( [19:13:14] well, I let message to all related wikiprojects [19:13:38] and I want feedback from absolutely everyone :) [19:13:53] the more the merrier [19:15:31] sjoerddebruin: by the way, if you have another way of handling this data, I absolutely want to know [20:57:05] oh no, Harmonia_Amanda is offline? [20:57:39] :O [21:21:27] WikidataFacts: did you ever try queries to replace https://www.wikidata.org/wiki/Property:P2021? [21:22:52] I had this a long time ago: https://twitter.com/WikidataFacts/status/728623992693731328 [21:23:17] though the query times out now, it seems [21:28:06] okay, but removing the /wdt:P279* from the scientific article class makes it fast again, and I think it should be unnecessary [21:28:28] sjoerddebruin: http://tinyurl.com/yb82fzs5 [21:29:08] I think it's a weird property. [21:31:15] I think it makes sense to have [21:31:35] we have plenty of people with P2021 statements who aren’t authors of anything, at least: http://tinyurl.com/ycsfeccq [21:31:47] so our coverage of scientific material isn’t enough to infer the number yet [21:33:07] Not enough? :O [21:35:27] nobody tell Daniel Mietchen ;) [21:37:34] TIL we have https://www.wikidata.org/wiki/Property:P66 [21:38:03] huh [21:38:34] It's for Chinese people, but that was edited out of the description before 2015 [21:38:46] less than 500 uses https://www.wikidata.org/w/index.php?title=Special:WhatLinksHere/Property:P66&namespace=0&limit=500 [21:40:20] Someone used it on some families... [21:40:44] With seems okay according to the constraints [21:41:09] wait, on family items? [21:41:29] but that doesn’t square with all the “item requires statement” constraints [21:41:33] even though the type constraint allows it [21:41:45] That's why we have https://www.wikidata.org/wiki/Wikidata:Database_reports/Humans_with_missing_claims/P66 [21:49:07] Aargh https://www.dropbox.com/s/cjh471jj4s638o7/Schermafdruk%202018-01-30%2022.48.56.png?dl=0 [21:49:52] oh. [21:50:25] "Oops, merged in the wrong item. Let me remove the valid sitelink and merge this disambugation page into some human!" https://www.wikidata.org/w/index.php?title=Q7812098&action=history [22:00:13] That https://www.wikidata.org/wiki/Wikidata:Database_reports/Constraint_violations/Mandatory_constraints/Violations should be zero makes me sad. :( [22:14:26] For crying out loud [22:15:50] well they would not be as big if statement ranks were taken into account [22:16:51] for instance https://www.wikidata.org/wiki/Q63514 has two ORCID ids but it should not appear on this report as one of them is marked as deprecated [22:17:34] so be merry, our data is not that bad \o/ [22:45:57] hi there [22:46:24] hello? [22:46:50] Hello? [22:47:17] :|