[01:11:49] <harej>	 I see that Merge has been OOJS-ified. It is very pretty. Thank you.
[09:14:36] <hashar>	 CI is back around :-)
[09:14:44] <hashar>	 wmflabs had some issue preventing it from spawning/deleting instances
[09:53:26] <icinga-wm>	 PROBLEM - Host wdqs1001 is DOWN: PING CRITICAL - Packet loss = 100%
[10:10:33] <onobitno>	 Hello out there. I was wondering, how one would go about to migrate a whole lexicon with its data into wikidata? Some of our data is already there, but we could enrich it, but most of it is new. Thanks
[10:17:51] <icinga-wm>	 RECOVERY - Host wdqs1001 is UP: PING OK - Packet loss = 0%, RTA = 0.57 ms
[12:10:06] <hashar>	 hey there. I am going to cut the wmf branch over the afternoon
[12:10:30] <hashar>	 starting roughly now, I have no idea when the branch will actually be cut though
[12:11:30] <aude>	 i don't think we are making a branch this week, afaik
[12:47:23] <hashar>	 the cut script is running now
[12:47:34] <hashar>	 but I guess for wikidata/wikibase it points to some branch
[13:19:52] <yannf>	 what's the right way to add a reference in https://www.wikidata.org/wiki/Q23921645 ?
[13:19:55] <yannf>	 for the date of birth
[13:19:57] <yannf>	 cf. http://data.bnf.fr/12629705/andre_de_paniagua/
[13:24:55] * hoo waves
[13:26:24] <thedj>	 yannf: this sort of touches on that: https://www.wikidata.org/wiki/Help_talk:Sources#Authority_control_databases_as_sources.E2.80.94how_to.3F.21
[13:27:58] <aude>	 hoo: is there a reason allowDataAccessInUserLanguage is not set for commons beta wiki?
[13:28:45] <hoo>	 aude: It should be
[13:28:58] <aude>	 ok, found it
[13:28:58] <hoo>	 https://phabricator.wikimedia.org/rOMWC31459edf7f128e6a73ef5525e5c5fdbe32f8a3ae
[13:29:05] <hoo>	 ok :)
[13:29:20] <aude>	 is this something we set also for wikidata?
[13:29:31] <aude>	 i don't know if we announced that (or need to)
[13:29:48] <hoo>	 That is https://phabricator.wikimedia.org/T122670
[13:29:54] <hoo>	 but still blocked on doing it for test
[13:30:03] <hoo>	 we could maybe do it for test today, if Lydia is ok
[13:30:21] <hoo>	 and then announce a date for Wikidata some time
[13:31:28] <aude>	 ok
[13:33:55] <yannf>	 thedj, sorry, I don't get it
[13:34:16] <yannf>	 I mean, practically, what should be done?
[13:36:04] <yannf>	 I added retrieved+date and stated in
[13:36:04] <yannf>	 +data.bnf.fr (Q20666306)
[13:36:59] <yannf>	 I've seen others doing differently
[13:37:48] <yannf>	 i.e. imported from
[13:37:48] <yannf>	 +data.bnf.fr (Q20666306)
[13:37:54] <yannf>	 or
[13:38:14] <yannf>	 reference URL+url
[13:38:18] <yannf>	 or
[13:38:20] <yannf>	 both
[13:42:15] <yannf>	 anyone else here?
[13:42:34] <TriX1>	 hi
[13:42:37] <yannf>	 hi
[13:43:56] <TriX1>	 I'm from mh
[14:01:42] <thedj>	 yannf: imported from is for bots 
[14:03:16] <thedj>	 though ppl often use it because they copy what they see.
[14:03:48] <thedj>	 but imported from should be a subset of stated in really if I understood correctly
[14:05:12] <yannf>	 ok
[14:05:52] <yannf>	 thedj, and what about data.bnf.fr (Q20666306) vs. reference URL+url?
[14:07:01] <yannf>	 data.bnf.fr (Q20666306) seems useful to classify the source
[14:07:05] <thedj>	 on that front I really don't know :(
[14:07:14] <yannf>	 machine readable
[14:07:23] <yannf>	 but it is less precise for human
[14:07:29] <yannf>	 s
[14:08:38] <thedj>	 there is some circularity when it comes to quoting authority control.
[15:25:35] <Lydia_WMDE>	 yurik: i am reading your emails. just super swamped right now. sorry! :(
[15:25:57] <yurik>	 Lydia_WMDE, tis ok, let me know what you think
[16:11:44] <gothos>	 Hello! Is it possible to retrieve a list of all possible locations via some kind of API? I could also extract it from the JSON dump, but would probably first have to extract all subclasses of location etc.
[16:17:41] <DanielK_WMDE__>	 gothos: maybe you can find something here that you can adopt for your purposes? https://www.mediawiki.org/wiki/Wikibase/Indexing/SPARQL_Query_Examples
[16:18:15] <DanielK_WMDE__>	 gothos: a list of all locations probably has several million entries... is that really what you want?
[16:49:33] <gothos>	 DanielK_WMDE__: Thanks, I'll take a look, didn't even know that there was a SPARQL endpoint. And yes, it is what I want since I'm doing some entitiy/relation extraction on a dataset :)
[16:49:58] <DanielK_WMDE__>	 gothos: the response is probably so big that the request will time out...
[16:50:21] <DanielK_WMDE__>	 any analysis on the full dataset is best done on the dump
[16:50:32] <DanielK_WMDE__>	 perhaps use SPARQL to find all the relevant subclasses, and then scan the dump?
[16:52:42] <gothos>	 DanielK_WMDE__: Yeah, no worries. Just found the RDf dump downloads, I'll just import the data into a local virtuoso db
[16:52:53] <gothos>	 and run the queries on that
[16:53:22] <DanielK_WMDE__>	 if you have virtuoso sitting around anyway, sure, why not :)
[17:23:35] <justme>	 DanielK_WMDE__: I run wikidata on blazegraph, I tried with mongodb and simple queries work, but anything a bit more complex mongodb timesout and uses huge amount of memory ...
[17:24:12] <justme>	 has anyone tried other nosql / json document based db to import latest.json directly?
[17:24:36] <DanielK_WMDE__>	 i don't know
[17:26:31] <DanielK_WMDE__>	 NoSQL usually isn't great for transitive queries. Nor is SQL, for that matter :) You want a graph database for that, or a proper reasoner. A triple store is a little bit of both... 
[17:28:49] <justme>	 I haven't been able to comprehend sparql notion .. I tried some examples, but I don't understand the sparql concept yet .. 
[18:28:36] <DanielK_WMDE__>	 jzerebecki: can you tell Lucie how to fix this? https://integration.wikimedia.org/ci/job/mwext-testextension-hhvm-composer/3505/console
[18:32:33] <jzerebecki>	 DanielK_WMDE__, frimelle: error is unrelated to the patch. it seems AP depends on something that depends on Wikidata, which includes its own copy of AP
[18:33:47] <jzerebecki>	 DanielK_WMDE__, frimelle: uh it is ContentTranslation...
[18:34:52] <frimelle>	 jzerebecki, DanielK_WMDE__ uuh, I'll be in tomorrow probably. Can we have a look on that then?
[18:35:24] <frimelle>	 Is it kind of a dependency circle?
[18:37:26] <jzerebecki>	 frimelle: yes AP -> ContentTranslation (for https://gerrit.wikimedia.org/r/#/c/280950/ ) -> Wikidata (which includes the 2nd copy of AP). will revert that and remove CT from the dependencies of AP
[18:37:46] <frimelle>	 jzerebecki: Thanks :3
[18:39:00] <DanielK_WMDE__>	 jzerebecki: thanks for sorting this out!
[19:25:06] <goffreder>	 hello everyone. who should I speak to about an error in a interlanguage link?
[23:49:45] <Dispenser>	 "Jep" --Lydia_WMDE; J makes a different sound in English :-P