[00:00:24] <harej>	 As Monty Python would say: "First shalt thou take out the Holy Pin. Then, shalt thou count to three. No more. No less. Three shalt be the number thou shalt count, and the number of the counting shall be three. Four shalt thou not count, neither count thou two, excepting that thou then proceed to three. Five is right out. "
[00:02:13] <WikiGnom>	 :-D
[00:12:03] <WikiGnom>	 Thanks again, bye!
[09:10:53] <addshore>	 aude: FYI https://phabricator.wikimedia.org/T116728 :D
[09:50:13] <aude>	 addshore: move all the things! :D
[09:50:18] <addshore>	 yup!
[10:27:01] <Adrian_WMDE>	 jzerebecki: https://phabricator.wikimedia.org/T110533
[11:04:47] <DanielK_WMDE_>	 benestar: hey! i want to schedule a discussion about derived values in the data model for tomorrow. if possible, i'd like you to be part of the discussion. what time would work for you?
[11:43:02] <_joe_>	 just an heads-up, I just migrated the wikidata cronjobs to HHVM, I don't see any issue atm, but please let me know if something fishy happens.
[11:43:54] <aude>	 ok
[11:43:56] <aude>	 thanks :)
[11:45:16] <_joe_>	 aude: the time of the migration is in the SAL, for reference
[11:45:21] <aude>	 ok
[11:59:21] <Thiemo_WMDE>	 jzerebecki: https://integration.wikimedia.org/ci/job/php-composer-test/20351/console
[12:06:02] <addshore>	 SMalyshev: around? :)
[12:07:13] <_joe_>	 addshore: it's a bit early for him I guess :)
[12:07:24] <addshore>	 you never know ;)
[12:08:03] <addshore>	 I asked Ironholds what timezone he was currently in a few days ago, and he replied at 4am ish I believe :D
[12:51:12] <nikki>	 I don't understand sparql...
[12:52:20] <nikki>	 why does "0000-01-01T00:00:00Z" count as year 0 but not "+0000-05-17T00:00:00Z" :/
[12:52:33] <Lydia_WMDE>	 frimelle: do we have a ticket for making article placeholders show up in search engines already? 
[12:54:13] * hoo waves
[12:54:17] <hoo>	 only here for a short bit
[12:54:22] <hoo>	 anything that needs CR?
[12:54:47] <aude>	 hoo: want easy chocolate?
[12:54:49] <aude>	 https://gerrit.wikimedia.org/r/#/c/250953/ :)
[12:55:00] <aude>	 not urgent
[12:56:59] <hoo>	 rsync office:/dev/bowls/chocolate/0 ~/mouth
[12:57:10] <hoo>	 Will have a look ;)
[13:00:05] <jzerebecki>	 nikki: how did you test for year 0?
[13:01:59] <nikki>	 originally I tried various things like FILTER (!?date > "0001-01-01T00:00:00Z"^^xsd:dateTime), then FILTER (year(?date) = 0)
[13:02:59] <jzerebecki>	 addshore: we could additionally copy our metrics to graphite without changing retention...
[13:03:01] <nikki>	 now I just tried FILTER(REGEX(?date, "0000")) that seems to match "+0000-05-17T00:00:00Z" and not "0000-01-01T00:00:00Z", which I guess might be what I wanted, but I still don't understand why it would return one and not the other
[13:03:15] <addshore>	 jzerebecki: yes
[13:03:31] <addshore>	 But also save stuff for longer
[13:03:43] <jzerebecki>	 addshore: will you? might be nice to have them around for use in grafana
[13:03:56] <addshore>	 jzerebecki: yeh, also doing it in graphite is trivial! :)
[13:05:24] <jzerebecki>	 nikki: mh. year() should work.
[13:08:58] <nikki>	 I'm trying to find things like https://www.wikidata.org/wiki/Q6767140, when I tried with year(), with http://tiny.cc/7h6j5x, I got two things which were both 1 bc, not 0
[13:09:29] <nikki>	 but both the 1 bc and 0 ones seem to have 0000 as the year when I output the date >_<
[13:32:35] <jzerebecki>	 DanielK_WMDE_: for date of birth this is the wrong xsd type, right?: wd:Q6767140 wdt:P569 "+0000-01-11T00:00:00Z".
[13:34:01] <frimelle>	 Lydia_WMDE: no ticket for search engines yet, will make one!
[13:34:33] <DanielK_WMDE_>	 jzerebecki: yes - it's kept as a string, because it cannot be converted to proleptic gregorian. Year -0000 does not have a well definined meaning in our internal representation, we use -0001 for 1 BC.
[13:35:36] <Lydia_WMDE>	 frimelle: merci :)
[13:36:37] <frimelle>	 :)
[13:39:15] <jzerebecki>	 DanielK_WMDE_: oh so that is what happens on IllegalValueException :)
[13:39:57] <DanielK_WMDE_>	 jzerebecki: i guess so. the alternative would be to skip the statement completely.
[13:40:22] <jzerebecki>	 DanielK_WMDE_: no its good
[13:49:54] <tobias47n9e>	 https://www.wikidata.org/wiki/Wikidata:Wikimania_2016 <--
[13:53:09] <jzerebecki>	 nikki: its a broken date value, so it will be represented in sparql as a string instead of a date, to find these use FILTER (datatype(?date) != xsd:dateTime)
[13:54:36] <jzerebecki>	 nikki: BCE 1 is stored in wikidata as -1
[14:03:06] <jzerebecki>	 nikki: BCE 1 is stored in sparql 1.1 as 0
[14:18:59] <benestar>	 Lydia_WMDE: found the bug with the broken toc on mobile and submitted a patch 
[14:19:01] <benestar>	 however, I still need to get in touch with jon to discuss the search thingy
[14:19:17] <Lydia_WMDE>	 benestar: cool! :)
[14:31:44] <addshore>	 benestar: !
[14:35:26] <tobias47n9e>	 https://www.wikidata.org/wiki/Wikidata:Wikimania_2016 <---
[15:38:22] <sjoerddebruin>	 So, replacing coordinates of sport teams with a item for the sport facility. :)
[16:42:30] <john___>	 hey
[16:53:53] * john___ slaps tonythomas around a bit with a large fishbot
[16:55:27] * tonythomas dodge
[17:18:24] <addshore>	 jzerebecki: we may end up with a seperate graphite instance and statsd instance in the analytics clsuter ;)
[17:19:27] <jzerebecki>	 addshore: can I make a graph show up in grafana that has data from both graphites?
[17:19:53] <addshore>	 I really have no idea at all about that :P
[17:20:08] <addshore>	 you would be able to make an graph in shiny pulling data from both
[17:23:32] <addshore>	 jzerebecki: https://github.com/grafana/grafana/issues/35 https://github.com/grafana/grafana/commit/36f56f3695b454e64d14148b029b3176779f37e4
[17:23:44] <addshore>	 grafana now supports multiple graphite backends since 1.4
[17:44:22] <jzerebecki>	 addshore: what version is running on grafana.w.o?
[17:44:32] <addshore>	 2. something now I believe
[17:45:34] <jzerebecki>	 thx
[17:49:51] <mutante>	 jzerebecki: ii  grafana                        2.5.0
[17:50:07] <mutante>	 on krypton
[21:25:41] <multichill>	 Help me out here. Who was running the viaf import bot again that imports other authority control based on viaf?
[21:59:35] <sjoerddebruin>	 multichill: Wasn't it a Krbot thing again?
[22:00:00] * nikki was thinking krbot too but doesn't really know
[22:03:49] <sjoerddebruin>	 Yep. https://www.wikidata.org/w/index.php?title=Q20468627&type=revision&diff=238677056&oldid=237913092
[22:14:11] <multichill>	 sjoerddebruin: I meant the other way around
[22:14:17] <multichill>	 So when it has viaf, add something else
[22:14:33] <sjoerddebruin>	 Ehm, you? :P
[22:14:40] <sjoerddebruin>	 Never seen that recently.
[22:14:43] <multichill>	 Created a bunch of new items, see https://www.wikidata.org/wiki/User:Multichill/KMSKA_creators
[22:14:53] <multichill>	 Noticed that quite a few of the new ones with viaf have ULAn too
[22:14:54] <sjoerddebruin>	 Yeah, saw a merge already.
[22:15:47] <multichill>	 Made a mistake in the dupe checking 
[22:15:48] <multichill>	 Fixed now
[22:17:39] <multichill>	 sjoerddebruin: https://www.wikidata.org/w/index.php?title=Q3384934&type=revision&diff=268186896&oldid=267445061 could be done by a bot
[22:17:52] <sjoerddebruin>	 Yep