[06:08:58] hello everyone! [06:10:43] nikki: i assume if you are connected, you'll read this message when you are really here [06:10:58] nikki: I perfectly agree that wrong descriptions is a problem [06:11:33] nikki: but the scripts are used since 2013 for one and 2015 for the other, so we are speaking of thousands of items, if not dozens of thousands [06:11:45] nikki: so any correction will have to be done by bot [06:12:23] nikki: can you list all languages that you think should be verified? then we can ask on the project chat for people knowing these languages [06:12:30] nikki: thank you :) [07:13:19] Morning! [07:40:59] sjoerddebruin: :) [07:42:32] sjoerddebruin: i did good work on names, it'll go down again! [07:43:18] (especially P1950) [07:44:21] Great! [07:44:32] (and given names) [07:45:03] (it's slower for family names, now, to many "errors-made-only-once") [07:45:16] oh hello dachary! [08:32:07] \o [08:32:11] coffee time ! [08:46:22] dachary: tea time here :p [09:58:30] DanielK_WMDE: Hey, I want to join the tla meeting [10:00:32] nvm, it's okay [10:10:16] sjoerddebruin: urgh, seems like nearly 500 disambiguation pages were added as surnames during the week-end? [10:10:20] * Harmonia_Boulot weeps [10:10:26] By who?!?! [10:10:36] no idea [10:11:01] but Saturday, I had less than 1400 disambiguation page used as surname, and now 1900! [10:11:17] how can that be? [10:20:43] sjoerddebruin: we really need an AbuseFilter [10:21:01] something to warn people there are adding an item which isn't a family name [10:21:24] they* [10:21:26] urgh [11:20:45] Harmonia_Boulot: what is the approach for combined surnames of married people? https://www.wikidata.org/wiki/Q20734808 [11:22:01] sjoerddebruin: both, with "order in the serie" with the married one having a "beginning date" and "P31:married name" as a qualifier [11:22:14] :) [11:22:35] But first I have to look at the tussenvoegsels then. :( [11:24:12] If we go for the seperate property approach, how would we combine that with this then... [12:19:47] Amir1: ERROR: Step ‘Publish JUnit test result report’ failed: No test report files were found. Configuration error? [12:20:02] Amir1: I am guessing this still due to some configuration issues? [12:20:03] Yeah, It is in our repo [12:20:17] we need to fix our configs [12:20:19] On it [12:20:51] ah, ok, thanks. Just didnt know if this is something we should fix, or rather something "they" should do [12:50:20] Lydia_WMDE: Around? [12:55:07] Amir1: Lydia is in vacation :) [12:55:27] Yeah, I pinged and then I said oops :D [13:54:26] Amir1: More secret birthday presents, yay! 8-) [13:55:14] Thiemo_WMDE: \o/ :D [14:13:29] o/ [14:34:00] aude: here? [15:29:50] leszek_wmde jakob_WMDE around? [15:30:01] Amir1: yes [15:30:02] Amir1: yup! [15:30:07] https://gerrit.wikimedia.org/r/#/c/317490/ [15:30:17] Jenkins is happy now, thanks to releng people [15:31:00] Glorian_WMDE: don't forget to sign your messages onwiki. [15:32:53] leszek_wmde: Do you think we should merge this patch? [15:33:23] Amir1: looks good as far as I can tell. There are two differences when I compare it against WikibaseMediaInfo so I am not sure if those two bits are really needed [15:33:47] Amir1: plus I don't want if we already want to send notifications to this channel already at this stage [15:33:56] *i dont know [15:34:46] leszek_wmde: let me test that part [15:35:37] about spamming in the wikidata-feed I hope the jenkins part ends now but other than that it's part of wikibase and we should notify people we are working on it [15:37:07] Amir1: meh, forget what I said about the IRC channel. Totally missed that it will not spam in #wikidata [15:37:09] then its good [15:37:33] also just checked that returning true in onSetupAfterCache hook is what wikibase does, so I have no objections [15:37:57] looks good to me too! [15:38:40] Amir1: so I am only sure if AutoloadClass in extension.json are needed (MediaInfo does not have it) but it has been there, now its just fixed [15:38:50] Amir1: so to the point: should I give it +2 [15:38:52] ? [15:39:47] leszek_wmde: I removed something that I think was redundant let's see if jenkins gets sad or not [15:39:55] if it's happy please merge it [15:40:27] ok Amir1 [15:43:57] leszek_wmde: okay, jenkins is happy [15:47:10] Has Thiemo left the office already? [15:47:24] Amir1: and its merged. thanks! [15:48:15] leszek_wmde: thank you for merging \o/ Now I can site freely and work on Lemma and other parts, it's getting really fun [15:48:57] DanielK_WMDE: ^ [15:50:04] hoo: He has left already. Daniel is not here either [15:50:24] I see… thanks for checking [15:51:14] Where is Lydia btw? [15:51:53] sjoerddebruin: she is on vacation.. she'll be back next Thursday [15:52:13] Oh. Lea isn't on IRC right? [15:52:31] sjoerddebruin: I think she does [15:53:00] sjoerddebruin: I've just told her that you are looking for her [15:53:04] :) [15:53:12] Hello sjoerddebruin :) [15:53:13] Wait, do you have two Lea's? [15:53:30] sjoerddebruin: yes :) [15:53:52] I'm looking for Lacroix, the community communication one. :P [15:54:01] Yes it's me :p [15:54:18] Oh, very confusing then. [15:55:00] Oh wait, it's on your user page. :) [15:55:42] @nat/wmf/ ? :o [15:56:19] Nemo_bis: Yeah… the WMDE office ip addresses have that because otherwise we were exceeding Freenode's limits by far [15:56:46] Makes sense [15:57:32] Though public IPs for everyone would be nicer ;) [16:00:11] sjoerddebruin: I've just noticed your message.. thanks for reminding me regarding to the signing the wiki :) [16:02:58] Amir1: Did you have a look at https://gerrit.wikimedia.org/r/#/c/315324/12/repo/tests/phpunit/includes/Api/SubscribersTest.php@24 ? [16:04:28] hoo: in ores meeting, will fix it asap [16:04:51] Amir1: Cool… I'll have another look after that (and will hopefully be able to +2 it) [16:05:58] awesome [16:06:00] thanks [16:14:53] teaser for SparquickL 4 : BIND( CONCAT( '#', SUBSTR( MD5( xsd:string( ?language ) ), 1, 6 ) ) AS ?fill ) . [16:42:54] WikidataFacts, hi, re the query you made for me yesterday, it seems incomplete [16:43:07] https://en.wikipedia.org/wiki/Wikipedia_talk:WikiProject_Women_in_Red#List_of_streets_named_after_reknowned_women [16:43:43] in fact, I find that 1399 results for the whole world is very few [16:58:10] DanielK_WMDE: Could you have a look at my changes to that patch (from PS7 to PS8)? [16:58:15] I would like to move on there fast [17:10:18] hoo: done [17:15:34] hoo|away: which patch? [17:30:55] DanielK_WMDE: whoops https://gerrit.wikimedia.org/r/317127 [17:50:22] DanielK_WMDE: https://phabricator.wikimedia.org/T147307#2712931 [17:50:35] See comments from there [17:52:38] hoo: i did [17:54:52] DanielK_WMDE: Hm [17:55:00] yes, that also sounded weird to me [18:01:50] hoo: you mean the one-element special case? yea, let's not. consistency is good. [18:02:27] *cringe* what if someone wants to perform an operator on numeric values from wikidata ? :/ [18:03:27] dividing the literal 123456789 by the literal 987654 is not exactly the same thing as dividing the html element 123456789 by the html element 987654, no matter what foo and bar are [18:03:27] Alphos: What do you mean? [18:03:43] for instance, if you were to divide population by area to get a density of population [18:03:58] (in the case of a country) [18:04:10] Alphos: What are you talking about? In the query service that should be alright [18:04:25] I think we even have example queries that do such calculation [18:04:25] s [18:04:57] so lua would still get the literal values ? [18:05:06] Alphos: Oh, Lua [18:05:12] you have the full entity in Lua [18:05:28] Why would you use the html/wikitext representation for calculation? [18:06:00] aren't there cases where you want some values of another entity ? [18:06:35] Alphos: just ask Lua for the data of the other entity [18:07:03] i guess i'm just always cringy whenever i see a data provider return formatted data :/ [18:07:13] we are talking about *formatting* functions. they gfenerate human readable output. if you want data, use the data without formatting it. [18:07:22] hm [18:07:28] well that's kind of the point. it's not a data provider. [18:07:50] one reason to add the tags is to make this abundantly clear. [18:08:03] alright then [18:08:09] can you suggest a better name that would make this even clearer? [18:08:10] would anyone with privileges be kind and tell me who created item https://www.wikidata.org/w/index.php?title=Q27478560 [18:08:27] it is a deleted item... [18:08:29] Wesalius: You did [18:08:34] (diff) 14:52, 20 October 2016 . . Wesalius (talk | contribs | block) (159 bytes) (‎Created a new item: #mix'n'match) (Tag: Widar [1.4]) [18:08:42] ok, thanks [18:09:04] :P [18:09:31] Alphos: actually, if there is a need for a better data provider, so people stop (ab)using the formatter for this purpose, please let us know about it! [18:10:51] DanielK_WMDE "view" ? i'm a tidbit hazy right now, but if you insist on it being a formatted piece of data, it would somewhat make sense [18:11:06] even "dataview" [18:11:27] I'd rather like a verb. "render"? "show"? [18:11:41] well, view is a verb, ya know :p [18:11:56] yea, but the function doesn't view. [18:12:06] ...and dataview isn't ;) [18:12:22] formatter is a bit vague, isn't it ? [18:12:29] (or format) [18:12:35] even "print"... though this only returns the value, it doesn't commit it to the output immediately. [18:12:38] then again, view kinda is as well [18:12:52] "format" is actually fine with me. [18:13:28] then you should expect it to return truly formatted values for (for instance) numeric values [18:13:55] 123 456 789 for instance [18:14:01] (depending on locale) [18:14:49] Alphos: that is indeed the case [18:15:01] also for dates, etc [18:15:15] i assume ISO for dates ? :) [18:15:23] no, local format [18:15:31] same as the date parser functions [18:15:34] ...i think. [18:15:41] as long as it's not "2 016"... :p [18:15:54] hehe... that would be bad [18:15:59] Should be the same you also see on Wikidata with the given use lang [18:16:11] (roughly) [18:16:12] i'm actually not sure what the old Lua function does exactly. I hope no ISO... [18:16:27] oh, just thought of something else [18:16:33] "asHTML()" [18:16:36] DanielK_WMDE: Well, it does ISO [18:16:44] See DataAccessSnakFormatterOutputFormatTest [18:16:52] and yes, that's just crazy [18:16:53] *sigh* [18:17:05] well, it allows people to use the date formatting code they have [18:17:10] and it's ok as an option [18:17:12] hoo : wait, is wikibase written in java ? :p [18:17:13] but shouldn't be the default [18:17:35] Alphos: we call it phava... [18:17:39] ^^ [18:18:20] phava beans with a nice chianti, i hope ! [18:26:30] DanielK_WMDE: I start to wonder what we plan to do with mw.wikibase.formatSnak(s) [18:26:48] we probably also want to have a way to get format wiki there [18:27:25] oh fuck [18:27:34] these are already format wiki [18:27:47] luckily we didn't deploy yet [18:29:12] *sigh* [18:32:25] DanielK_WMDE: https://gerrit.wikimedia.org/r/317127 +2 now? [18:38:55] sjoerddebruin: I have seriously no idea what happened during the week-end but there is 500 disambiguation pages used as family names more than Saturday [18:39:15] like, did someone add family names without verifying? [18:39:43] Maybe some pages got an unrelated edit causing them to sync properly with the query service? [18:39:58] for 500 items? [18:40:15] Seems possible for me. [18:41:46] Harmonia_Amanda or maybe someone reverted your changes for some reason ? [18:42:19] Alphos: yno, I keep the family names in my watchlist [18:42:38] if someone had added back the wrong "disambig", I would have seen it [18:42:51] ^^ [19:00:19] DanielK_WMDE: Can I find an excuse to come visit WMDE for a few days? :D [19:00:34] Reedy: Wikidata birthday next week? [19:01:14] hoo: touche :P [19:01:18] Hmm [19:01:25] Maybe a little short notice [19:01:27] I wonder [19:01:47] <3 [19:02:39] I was kinda thinking a bit closer to Weihnachten [19:03:01] There is no deadline on the list, though. [19:03:02] I guess you can just come over whenever you want [19:03:15] heh [19:03:19] Any idea who I'd need to run it by? [19:03:28] I presume Lydia would be alright if she wasn't away [19:03:32] yeah [19:04:38] * Reedy adds it to his TODO list [19:05:01] You can ask Lea I think if you want to come that Friday. [19:05:33] I'm sure I can sudo ask [19:05:42] I wonder if it's actuall viable to do it [19:06:19] Yeah, we hosted various people in the office before [19:06:23] both WMF and volunteer [19:06:46] I spent a week in the office a few years ago :) [19:07:01] Including sleeping? [19:07:10] They put my up in a hotel :P [19:07:35] Reedy: When was that? Probably still the old office (or even office-2?) [19:07:43] Um [19:08:09] Before the Berlin hackathon when most people were in the Mercure (?) acrosss the street [19:08:21] Very near the office and technik museum [19:08:22] Ah, the 2012 one? [19:08:26] yeah [19:08:30] so that was office-1 [19:08:36] haha [19:08:50] but were literally across the corner now [19:08:54] * we're [19:09:33] Good luck finding a cheap hotel though. :) [19:09:42] Airbnb ;) [19:14:08] Don't have experience with that. :) [19:21:33] multichill: good evening! Things like https://www.wikidata.org/wiki/Q27526206 are really slowing me down, but worth the effort! [19:45:15] DanielK_WMDE: aude: Review of https://gerrit.wikimedia.org/r/317568 would be really appreciated [19:46:02] We don't have integration tests for these, so things went sideways there [20:48:12] sjoerddebruin: Thanks for your response on my talk page. Did you also create a new item for the other person? [20:55:07] nvm, I found it [21:08:00] Oh right, correct mix-n-match... [22:41:15] ^&& JOIN #wikimedia-ayuda &&^ [23:14:44] ^&& JOIN #wikimedia-ayuda &&^ [23:28:17] ^&& JOIN #wikimedia-ayuda &&^ [23:29:27] ^&& JOIN #wikimedia-ayuda &&^ [23:30:39] o_O