[00:00:21] yurik: the silent ones are the evil ones... [00:00:22] DanielK_WMDE: I have a fix, but I suspect more database leakage cause that test only errors isolated, but passes in the whole suite \o/ [00:00:36] that's why i like php so much... [00:00:53] it's more of an architectural problem [00:00:56] although nothing beats the if (array()) is false [00:00:59] mediawiki relies on global state a lot [00:01:05] stuff leaks state all over the place [00:01:08] hehe [00:01:11] yeah, true [00:01:25] but still, amazing php stuff is showing up all over the place [00:01:26] Today I had a look at the flow source code... not happy, either... [00:01:40] hehe... [00:01:49] aaanyway, daniel, thx for the advice, please please review the config and tell me early on what you think [00:01:54] they have two phpunits for the whole thing :P [00:02:00] i'm still debating how the extension framework should look like [00:02:18] because every extension will want to extend config system [00:02:23] yurik: i still think the text based approach will get in the way eventually. oh, well :) [00:02:40] DanielK_WMDE, we can always work around it - not an issue really [00:02:44] there's a nice lib/extension for data structure diffs (called Diff, appropriately) [00:02:55] config is an internal storage - not exposed outside of the extension [00:02:58] MariaDB [its_a_mw]> SELECT COUNT(*) FROM page site_identifiers; [00:02:59] 3 [00:03:03] * hoo cries out loud [00:03:07] but the visualization stuff for that is still much bound to wikibase :/ [00:03:27] DanielK_WMDE, how about we schedule half an hour when you are awake? [00:03:31] i could do it now of course [00:03:34] but don't want to stress you [00:04:37] i'd say tomorrow, but i'll only be there until 16:00 utc or so. [00:04:50] hmm.. well, when do yo ustart? [00:04:50] email? [00:04:55] nah, takes forever [00:05:07] i will probably be awake for a while [00:05:16] not too early, worked late today. 11 here [00:05:28] hehe [00:05:48] oh, only 4 hr diff? [00:06:15] ok, lets try tomorrow than, please read the RFC [00:06:20] try to ming me on the weekend in the evening, if you like. my laptop is usually online, sometimes i'm too :) [00:06:30] hehe [00:06:34] or that [00:06:44] i skimmed it [00:06:58] aaand? [00:07:05] (03PS1) 10Hoo man: Fix EditEntityActionTest if run on a repo+client wiki [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/101147 [00:07:27] looks good at a glance :) [00:07:37] you are soo helpful! [00:07:53] DanielK_WMDE: https://gerrit.wikimedia.org/r/101147 easy one... but the underlying problem sucks [00:11:17] (03CR) 10Hoo man: "Error this fixes:" [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/101147 (owner: 10Hoo man) [00:22:54] DanielK_WMDE: Any idea why ReferencedPagesFinderTest still leaks into the DB (although we recently added @group Database) [00:35:28] hoo: does it derive from MediaWikiTestCase? [00:36:05] DanielK_WMDE: No... doh... will fix that [00:36:10] Didn't realize [00:36:14] :) [00:36:29] just occurred to me... [00:37:05] anyway. good night! [00:37:09] good night ;) [01:02:22] (03PS1) 10Hoo man: Fix group Database tests which derived from PHPUnit_Framework_TestCase [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/101160 [09:19:53] (03CR) 10Aude: [C: 032] Fix group Database tests which derived from PHPUnit_Framework_TestCase [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/101160 (owner: 10Hoo man) [09:21:13] heh [09:23:13] * aude waves [09:23:17] (03Merged) 10jenkins-bot: Fix group Database tests which derived from PHPUnit_Framework_TestCase [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/101160 (owner: 10Hoo man) [09:29:45] Hoi ... back again. Had problems connecting to the Internet [09:33:13] ##Revi [09:33:32] what? [09:34:24] I am just typing your channel code to go to your channel. sorry for pinging you :P [09:35:02] you should use /join command to join. [09:37:40] (03PS1) 10Aude: Style cleanup in scribunto code [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/101185 [09:38:23] just don't break anything this time :P [09:38:26] (03Abandoned) 10Aude: Style cleanup in scribunto code [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/101185 (owner: 10Aude) [09:38:38] who? me? [09:38:40] [travis-ci] wikimedia/mediawiki-extensions-Wikibase#1416 (master - b282e04 : Marius Hoch): The build is still failing. [09:38:41] [travis-ci] Change view : https://github.com/wikimedia/mediawiki-extensions-Wikibase/compare/a22f7fb338e4...b282e047a955 [09:38:41] [travis-ci] Build details : http://travis-ci.org/wikimedia/mediawiki-extensions-Wikibase/builds/15389464 [09:38:52] I did not even break anything [09:39:10] aude... it's a reference to yesterday [09:39:24] Revi: :) [09:39:26] errr [09:39:29] rschen7754: [09:40:03] (03CR) 10Zfilipin: [C: 04-1] "You are improving, but there are still a few things that should be changed, see inline comments." (038 comments) [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/100591 (owner: 10Vldandrew) [09:40:24] WD:RFD is slightly backlogged [09:42:02] does ##Revi ping you [09:42:24] Revi [09:52:24] (03Abandoned) 10Tobias Gritschacher: Revert "Taking "movetoolbar" and "rankselector" out of experimental features" [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/100789 (owner: 10Tobias Gritschacher) [09:53:12] (03CR) 10Tobias Gritschacher: [C: 032] Nicer Ruby gemset name [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/100812 (owner: 10Zfilipin) [09:55:03] (03PS1) 10Aude: Fix covers tags [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/101188 [09:56:53] (03Merged) 10jenkins-bot: Nicer Ruby gemset name [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/100812 (owner: 10Zfilipin) [10:00:45] Jianhui67: Yes, it does ping me. [10:08:29] (03CR) 10Addshore: [C: 032] Fix covers tags [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/101188 (owner: 10Aude) [10:11:33] \o/ [10:14:09] [travis-ci] wikimedia/mediawiki-extensions-Wikibase#1417 (master - e06f9ef : jenkins-bot): The build is still failing. [10:14:09] [travis-ci] Change view : https://github.com/wikimedia/mediawiki-extensions-Wikibase/compare/b282e047a955...e06f9ef3f7a1 [10:14:09] [travis-ci] Build details : http://travis-ci.org/wikimedia/mediawiki-extensions-Wikibase/builds/15390922 [10:16:36] silly build [10:27:44] [travis-ci] wikimedia/mediawiki-extensions-Wikibase#1418 (master - abef936 : Addshore): The build is still failing. [10:27:44] [travis-ci] Change view : https://github.com/wikimedia/mediawiki-extensions-Wikibase/compare/e06f9ef3f7a1...abef93680247 [10:27:44] [travis-ci] Build details : http://travis-ci.org/wikimedia/mediawiki-extensions-Wikibase/builds/15391315 [10:36:48] (03PS1) 10Aude: Style improvements in Scribunto_LuaWikibaseLibrary [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/101190 [10:37:01] easy chocolate ^ [10:38:21] (03CR) 10Addshore: [C: 032] Style improvements in Scribunto_LuaWikibaseLibrary [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/101190 (owner: 10Aude) [10:38:30] :) [10:42:03] (03Merged) 10jenkins-bot: Style improvements in Scribunto_LuaWikibaseLibrary [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/101190 (owner: 10Aude) [10:58:19] [travis-ci] wikimedia/mediawiki-extensions-Wikibase#1419 (master - d7660d0 : jenkins-bot): The build is still failing. [10:58:19] [travis-ci] Change view : https://github.com/wikimedia/mediawiki-extensions-Wikibase/compare/abef93680247...d7660d026b6b [10:58:19] [travis-ci] Build details : http://travis-ci.org/wikimedia/mediawiki-extensions-Wikibase/builds/15392579 [11:18:38] (03PS3) 10Daniel Kinzler: (bug 58342) Inject TermBox when viewing [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/101064 [11:20:01] Lydia_WMDE: do you know why there are no properties "sibling" or "parrent" but specific ones like "brother", "sister", "father" and "mother"? In my opinion it would be smarter to have the generic ones and get the specific one implicitly by the gender of the corresponding item [11:20:35] because that does not work in some languages [11:20:45] same reason we have a sex property for people and animals [11:21:04] that has nothing to do with languages, its about the concept [11:21:19] it does [11:21:31] there is for example a language where there is not concept of grandfather [11:21:41] it is either the father of the mother or the father of the father [11:22:17] grandfather is one of the unnecessary properties [11:22:34] parrent of parrent [11:22:50] independent from specific gender [11:22:56] lbenedix: it'S only unnecessary if you can evaluate properties transitively. [11:23:04] which we currently can't. [11:23:22] my point is: there are differences in concepts accross languages [11:23:25] and wikidata has to handle them [11:23:42] lbenedix: also, it may be useful to give specific qualifiers or sources for the granparent relationship, without relyong on the sources for the parent relationship [11:23:47] it is possible to traverse the tree "by hand", or with js [11:24:12] lbenedix: possible, yes. useful, no. [11:24:12] was there a discussion about the generic "parrrent" property? [11:24:25] likely [11:24:35] is it usefull to have a grandgrandgrandgrandgrandfather property?! [11:24:52] just to model this relation? [11:24:57] lbenedix: no, because it's unlekey that an external source talks about that relationship [11:25:33] any relationship that is discussed directly, as such, in sources we want to use, needs to be modeled directly [11:25:33] (03CR) 10Aude: [C: 04-1] "not looked at everything in detail, nor try tests / manually yet but see comments" (033 comments) [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/101064 (owner: 10Daniel Kinzler) [11:25:48] the bible does ;) [11:26:15] Lydia_WMDE: can you find a link for this dicussion? [11:26:48] no sorry - too busy with other things atm [11:27:08] any hint, where I could look for it? [11:27:10] check the archive for property creations [11:27:21] on the property proposals page somewhere probably [11:27:39] or on the discussion page of the property [11:27:45] thx [11:28:18] lbenedix: the problem you observe here is an old never really resolved discussion in the semantic web community… you're searching for some tradeoff between just stating direct relations and doing everything else with reasoning at query time or materializing every however complex property over multiple hops at statement time. [11:28:58] in the end the only thing we can do is meet in the middle at what feels right [11:30:01] jorn, lbenedix: in wikidata, there's an additional factor to consider: we can only talk *about* direct relations (that is, give qualifiers and sources) [11:30:05] It's tired to remember so much QID while requesting deletion on my phone :( Revi [11:30:06] and grandgrandgrandgrandgrandfather doesn't ;) [11:30:13] so, whenever it is useful to give qualifiers and sources, we need a direct relation [11:30:25] DanielK_WMDE: yupp, i know [11:31:03] i'm no fan of reification either ;) [11:31:29] lbenedix: redundancy is not necessarily bad. redundancy can give depth and strength to a knowledge base [11:31:48] also it will be fun to do association mining on the multi hop ones [11:32:36] right, but it leads to inconsitency [11:32:55] jorn: you'd have to track the sources and qualifiers while doing that ( "A.b is foo, GIVEN x and y, TRUSTING w or v" ) [11:33:39] * lbenedix cant wait to see how you will do queries in wikibase [11:33:43] lbenedix: what woudl you rather have, consistent but incomplete and/or misleading, or inconsistent but informative? [11:34:04] consistency is NOT a goal of wikidata. in fact, modelling inconsistencies is a defined goal of wikidata [11:34:27] (of yourse, we shouldn't *introduce* inconsistencies. just model them) [11:34:32] best of both worlds would be great ;) [11:34:44] consistent and complete ;) [11:35:30] lbenedix: i just gave an outline of a logic for reasoning with qualifiers and sources: "A.b is foo, GIVEN x and y, TRUSTING w or v". Definine logical operators based on that should be intuitive. [11:36:13] but I cant imagine how this will scales on the json-blobs in a mysql database [11:36:23] note that inconsistencies are still possible (e.g. the same source making contradictory statements), but should be much rarer [11:36:59] lbenedix: oh, it doesn't. to apply reasoning, it needs to be loaded into a different kind of database. [11:37:32] would be interesting to investigate whether there already is a technology that can a) handel the complex logic and b) scales beyond a 100 million statements [11:37:57] ask the semantic web guys [11:38:19] no, they'll just say "oh, it can be modeled as triples". sure it can. but that's extremely inefficient [11:38:38] it's the AI people i want to talk to. this is about knowledge representation. [11:39:14] there is a overlap between ai and smw people [11:39:37] sure [11:41:06] will it be possible to have another database in the wikimedia universe? [11:41:58] yes it will be possible for us [11:43:05] but nobody promised that wikimedia will host a service that supports reasoning on wikidata. it would be cool, sure, but it's just a vague idea, and no commitment. [11:43:25] right - no reasoning from our side anytime soon [11:43:27] a *query* service, yes. but no reasoning in the foreseeable future [11:43:41] making that work is a very nice research project. i'd love to be part of that :) [11:48:03] (03PS1) 10Henning Snater: Toolbar edit group widget optimization [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/101198 [11:48:36] another question... as far as I know it is not be possible to have klingon labels and descriptions in wikibase... Are there plans to be indipendent from mediawiki for the languages? [11:50:29] accepted languages are determined by the language committee [11:50:48] speaking of which [11:50:56] I need a language committee, if I install wikibase on a local machine?! [11:51:07] DanielK_WMDE: aude: can one of you have a look at https://bugzilla.wikimedia.org/show_bug.cgi?id=57342 again? siebrand left a comment [11:51:15] no [11:51:19] but for wikidata [11:51:52] I think, I can't define my own languagecodes in wikibase [11:53:10] can't look right now but high on to-do :) [11:54:08] thx [11:55:32] (03CR) 10Tobias Gritschacher: Migrate simple RSpec browser tests to Cucumber. (0311 comments) [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/100591 (owner: 10Vldandrew) [11:56:49] (03PS1) 10Jens Ohlig: Implementation of snakformatted property table in Lua [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/101201 [11:56:58] (03CR) 10jenkins-bot: [V: 04-1] Implementation of snakformatted property table in Lua [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/101201 (owner: 10Jens Ohlig) [13:00:08] (03PS1) 10Henning Snater: Create only one rank selector menu [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/101202 [13:00:12] * Revi wonders if manually supervised bot needs approval :P [13:02:33] hi Revi :-) [13:02:39] :O [13:04:08] There's no mention about manually supervised bot @ [[WD:Bots]] [13:04:23] the WD:Bot page says "tools used to make edits without the necessity of human decision-making" - what I was doing has no element of automatic decision making [13:06:15] basically, it's a script which presents the labels from an item which has taxon rank but no taxon name, and asks if any of those are the taxon name [13:07:08] problem is, it was flooding RC. [13:08:57] ok, I'll file a bot approval? [13:09:45] :D [13:12:20] (03CR) 10Aude: Implementation of snakformatted property table in Lua (032 comments) [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/101201 (owner: 10Jens Ohlig) [13:16:37] (03PS5) 10Jens Ohlig: Add mw.wikibase.properties as a Lua function Patch set 4: Make tests run under Jenkins [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/100577 [13:19:53] (03PS6) 10Jens Ohlig: Add mw.wikibase.properties as a Lua function Patch set 4: Make tests run under Jenkins [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/100577 [13:19:54] (03PS2) 10Jens Ohlig: Implementation of snakformatted property table in Lua [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/101201 [13:21:45] (03PS3) 10Jens Ohlig: Implementation of snakformatted property table in Lua [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/101201 [13:21:52] (03CR) 10Daniel Kinzler: (bug 58344) use getUserLang for EntityView (039 comments) [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/100829 (owner: 10Daniel Kinzler) [13:22:21] (03PS3) 10Daniel Kinzler: (bug 58344) use getUserLang for EntityView [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/100829 [13:30:56] (03CR) 10Daniel Kinzler: (bug 58342) Inject TermBox when viewing (031 comment) [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/101064 (owner: 10Daniel Kinzler) [13:36:03] (03PS4) 10Daniel Kinzler: (bug 58342) Inject TermBox when viewing [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/101064 [13:37:23] (03PS3) 10Daniel Kinzler: Refactor SqlStore::doSchemaUpdates. [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/100361 [13:39:55] (03PS4) 10Jens Ohlig: Implementation of snakformatted property table in Lua [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/101201 [13:41:48] (03CR) 10Daniel Kinzler: Implement {{#infobox:...}} as a simple wrapper (031 comment) [extensions/Capiunto] - 10https://gerrit.wikimedia.org/r/100411 (owner: 10Hoo man) [13:43:17] just asking here, do you think flood flag is appropriate if I have a big deletion request of > 30 QID to request deletion and I don't want to flood rc? [13:43:45] I don't use bulk deletion request in RfD, but I use merge.js [13:43:57] which requests deletion 1 by 1 [13:45:46] rschen7754 [13:47:32] (03CR) 10Daniel Kinzler: [C: 032] Allow id and type params in wbeditentity data! [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/94775 (owner: 10Addshore) [13:47:51] (03PS7) 10Jens Ohlig: Add mw.wikibase.properties as a Lua function [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/100577 [13:48:14] (03PS5) 10Jens Ohlig: Implementation of snakformatted property table in Lua [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/101201 [13:49:18] (03CR) 10Daniel Kinzler: [C: 04-1] Fix EditEntityActionTest if run on a repo+client wiki (031 comment) [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/101147 (owner: 10Hoo man) [13:49:50] (03CR) 10Jens Ohlig: [C: 031 V: 031] Implement {{#infobox:...}} as a simple wrapper [extensions/Capiunto] - 10https://gerrit.wikimedia.org/r/100411 (owner: 10Hoo man) [13:51:06] (03Merged) 10jenkins-bot: Allow id and type params in wbeditentity data! [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/94775 (owner: 10Addshore) [13:52:53] (03CR) 10Daniel Kinzler: [C: 04-1] Add mw.wikibase.properties as a Lua function [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/100577 (owner: 10Jens Ohlig) [13:53:34] (03CR) 10Daniel Kinzler: Add mw.wikibase.properties as a Lua function (033 comments) [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/100577 (owner: 10Jens Ohlig) [13:53:43] (03CR) 10Daniel Kinzler: "see comments on PS6" [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/100577 (owner: 10Jens Ohlig) [13:54:00] (03PS8) 10Tobias Gritschacher: Browsertests: generate testclaims using the api [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/100359 [13:55:25] addshore: you join the daily? [14:00:30] (03CR) 10Daniel Kinzler: [C: 04-1] Implementation of snakformatted property table in Lua [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/101201 (owner: 10Jens Ohlig) [14:01:02] inductiveload: you forgot to transclude {{Wikidata:Requests for permissions/Bot/InductiveBot}} on [[Wikidata:Requests for permissions/Bot]] [14:01:08] so I did it for you :D [14:01:20] just FYI. [14:01:29] i just used the template generatory thing [14:01:36] thanks :-) [14:03:06] oh i see, didn't realise I had to do it like that [14:06:50] [travis-ci] wikimedia/mediawiki-extensions-Wikibase#1420 (master - bee475a : jenkins-bot): The build was fixed. [14:06:51] [travis-ci] Change view : https://github.com/wikimedia/mediawiki-extensions-Wikibase/compare/d7660d026b6b...bee475a740dd [14:06:51] [travis-ci] Build details : http://travis-ci.org/wikimedia/mediawiki-extensions-Wikibase/builds/15399776 [14:07:21] my question got ignored :( [14:52:04] Jianhui67: Bulks are far easier. I doubt you will be granted flood flag for RFD work as it is too great to out of the flags purpose. [14:53:39] has someone an idea what went wrong here: http://lbenedix.monoceres.uberspace.de/screenshots/awbt2fejo4_2013-12-13_15.53.15.png [14:53:47] "Exception Caught: Specified index is out of bounds" [14:54:58] it failed in chrome, worked in ff [14:55:08] another thing: the order of statements change after reload [14:56:06] Lydia_WMDE: ^ [15:07:17] feature request: order qualified claims by date (I added the "member of sports team" statements for Q7964894) [15:10:08] (03CR) 10Qgil: [C: 04-1] "Just a comment, if your first patch requires improvements you should keep ammending the Gerrit change https://gerrit.wikimedia.org/r/#/c/1" [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/101054 (owner: 10Aaron1011) [15:13:26] (03CR) 10Qgil: [C: 04-1] "Also a comment about the commit message: it lacks a description of the changes this patch brings, and a reference to the bug report it wan" [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/100209 (owner: 10Aaron1011) [15:15:39] aude: what'S the page you collect notes for the next deployment on? [15:16:20] DanielK_WMDE: I have a question about modelling something [15:16:23] Q207120 [15:16:25] https://www.mediawiki.org/wiki/User:Aude/Deployments [15:16:35] he was a player and a trainer for fc st pauli [15:17:29] there is a claim "member of team" "st pauli" (from 1989, to 2002, occupation player) [15:17:41] i want to add that he was a trainer [15:17:51] in another time [15:18:03] add another claim [15:18:25] i cant add another claim "member of team" "st pauli" [15:18:35] sure, why not? [15:18:45] right [15:18:48] interesting [15:18:55] reloading helps [15:19:21] lbenedix: oh, you can't in one go? that's bad, please file a bug report. [15:19:44] there was some confusion about whether that should be allowed or not, so there might be code left somewhere that assumes it isn't. [15:20:11] adding qualifiers while adding a claim is not possible as well, right? [15:20:25] but you can have two claims with the same main snak. you can even have two completely identical claims, though there may be cases where the duplicate would get dropped. [15:20:40] lbenedix: not yet, i think [15:21:57] adding qualifiers is not easy... this explains why the grey line here is allmos 0: http://page.mi.fu-berlin.de/benedix/wikidata/stats/img/items_statements_qualifiers.png [15:23:06] here the same graph with log scale: http://page.mi.fu-berlin.de/benedix/wikidata/stats/img/items_statements_qualifiers_log.png [15:23:49] aude: hey :-) [15:24:04] hi [15:24:04] aude: so beta should now let you access the beta public URL from some instances [15:24:10] \o/ [15:24:15] that was https://bugzilla.wikimedia.org/show_bug.cgi?id=45868 [15:24:22] sorry it took so long :D [15:24:26] thanks for the puppet patch :) [15:24:35] * aude would have struggled to come up with something like that [15:25:02] lbenedix: i think the main reason is that qualifiers have to be added manually, while "main values" can often be imported from wikipedia [15:25:08] Tobi_WMDE: ^ [15:25:20] qualifiers can be imported as well... [15:25:36] for football players it would be easy to add start and end dates from the infobox [15:25:40] hashar: can beta instances also access each other's database? [15:25:52] yeah that should be fine DanielK_WMDE [15:25:56] it's api requests [15:26:05] aude: huh? [15:26:10] that was the problem and site links does some normalization and stuff via api [15:26:16] see MediaWikiSite in core [15:26:22] could probably be improved [15:26:40] aude: i know, i just want to make sure we don't have a simmilar problem with cross-wiki db access [15:26:48] because we also need that [15:27:07] no, we shouldn't [15:27:09] lbenedix: but i think few people (bots) do that. go right ahead and start :) [15:27:19] or at least easier to fix if that's the case [15:27:28] but stuff like central auth works so it should work [15:27:50] aude: hashar: so how far are we with getting a beta-client hooked up with the beta-repo? [15:27:56] maybe I'll start this...I have a botaccount from the hackathon ;) [15:28:13] http://wikidata.beta.wmflabs.org/wiki/Q7143 [15:28:19] err should be https :) [15:28:21] btw. ordering statements is a usabilityhell: http://lbenedix.monoceres.uberspace.de/screenshots/fsgg1o8x7d_2013-12-13_16.27.05.png [15:28:23] it works! [15:28:23] Tobi_WMDE: I have no idea what you are talking about :D [15:28:31] Tobi_WMDE: done [15:28:46] =) [15:29:27] http://en.wikipedia.beta.wmflabs.org/wiki/David_Boswell_Reid [15:29:39] aude: cool [15:29:43] :) [15:29:54] https://wikidata.beta.wmflabs.org/w/api.php [15:29:56] badges! [15:29:57] \o/ [15:30:08] I don't see the data-item link on http://en.wikipedia.beta.wmflabs.org/wiki/David_Boswell_Reid though [15:30:18] it has site link [15:30:33] i see data item also [15:30:40] aude: yes, saw that. but also only after purging [15:30:44] we'll need to setup change propagation [15:31:04] involves the cron job, so i need to see where/how to do it [15:31:25] i dont understand why its not possible to add sources for qualifiers [15:31:36] \o/ [15:31:38] a qualifier is a claim about a claim [15:32:07] https://gerrit.wikimedia.org/r/#/c/101209/ is the patch btw [15:32:24] that makes it work, and is see a related patch for parsoid [15:32:28] i* [15:32:47] aude: hashar: thx for the fix [15:32:50] and another patch [15:32:51] aude: so the firewall rules I have applied has fixed the issue with beta wikis not being able to fetch from the beta wikidata repo ? [15:32:55] that improves the situation a lot [15:33:01] hashar: yes! :) [15:33:04] Tobi_WMDE: to be honest, I haven't done a lot. aude did it all ! [15:33:09] nah [15:33:15] I just wrote down a bunch of non sense that nobody care about [15:33:18] * aude can't do all this in puppet or it would be a challenge [15:33:24] ah yeah that [15:33:28] well it is like using a hammer [15:33:34] not that useful of a skill :-D [15:33:35] heh [15:33:46] * aude still learning more about puppet :) [15:33:49] it is much harder to figure out the plan of the whole house that is going to be build with that hammer. [15:33:53] what to put where [15:33:55] so basically, I have hit something with a hammer [15:33:57] yeah [15:34:03] hashar: you can fix a lot of things with a hammer [15:34:07] and now you are a dmiring the whole house that got build somehow I have no idea how [15:34:08] e.g. computers… [15:34:12] lazowik: definitely :-D [15:34:17] glad it unlocked something [15:34:19] :) [15:34:26] and I am very happy aude wrote that email asking for status [15:34:31] looks like it helps parsoid also [15:35:05] might [15:35:22] not sure to be honest, iirc parsoid is made to use the private ip address. But we never knwo [15:35:30] hmmm, ok [15:36:41] (03PS4) 10Daniel Kinzler: (bug 40157) store entity status in page_props. [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/99385 [15:36:55] (03PS5) 10Daniel Kinzler: (bug 40157) store entity status in page_props. [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/99385 [15:37:46] (03PS4) 10Daniel Kinzler: (bug 47135) Make row IDs use BIGINT. [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/99637 [15:38:26] (03PS5) 10Daniel Kinzler: (bug 45529) use composite indexes on wb_terms. [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/99660 [15:38:38] (03PS6) 10Daniel Kinzler: (bug 45529) use composite indexes on wb_terms. [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/99660 [15:38:45] (03PS4) 10Daniel Kinzler: Refactor SqlStore::doSchemaUpdates. [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/100361 [15:39:58] (03PS4) 10Daniel Kinzler: (bug 58344) use getUserLang for EntityView [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/100829 [15:42:53] DanielK_WMDE: there are bots adding qualifiers: http://page.mi.fu-berlin.de/benedix/wikidata/stats/img/who_did_what_added_bots_vs_humans_relative.png [15:51:36] I'm flooding rc :( [16:16:00] (03CR) 10Jeroen De Dauw: [C: 04-1] (bug 58344) use getUserLang for EntityView (031 comment) [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/100829 (owner: 10Daniel Kinzler) [16:20:24] (03CR) 10Jeroen De Dauw: (bug 58344) use getUserLang for EntityView (031 comment) [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/100829 (owner: 10Daniel Kinzler) [16:25:14] addshore: you win the lottery! https://travis-ci.org/wikimedia/mediawiki-extensions-Wikibase/builds/15399776 [16:25:20] First passing test on travis for some time [16:25:55] In fact the only build on master on the first page that did not fail [17:02:38] hola wikidataticians [17:03:31] :p [17:05:55] Lydia_WMDE: per the profiling question yesterday, I have found this: https://git.wikimedia.org/tree/operations%2Fpuppet.git/19185d5359b31ace6c0922114abae20800657b9d/files%2Fcgi-bin%2Fnoc (the code for the report.py) [17:06:17] greg-g: thanks! i will let daniel know [17:07:24] * aude suspect what we want is all the ganglia etc stuff [17:08:19] e.g. click data browser on http://gdash.wikimedia.org/ [17:08:28] which we can't access [17:09:13] aude: I was specifically asked about that report.py, but yeah, working on that too :) [17:09:14] it can probably be made to track wikidata specifically [17:09:18] ok [17:09:35] i don't think we are very aware of the other stuff avaialble :) [17:48:40] hi [17:49:14] i'm hacking some javascript extension, is there an easy way to set the label into edit mode? [18:47:41] jorn: something like $('.wb-ui-labeledittool input').trigger('input'); [19:05:36] Lydia_WMDE: so, Ori says that adding the ability for it to be WD specific will probably be hard, but he offered his assistance to you all [19:05:52] where "it" == that profile.py [19:48:26] greg-g: thanks for the update [19:50:54] Hi. I am new to Wikidata. I see that Firefox' entry on Wikidata contains the version number for the stable version of Firefox. Is there a some reason as to why this information is not used in the article on Firefox on the English Wikipedia? Has it perhaps just not been done yet (but should and can be)? [19:52:25] * lbenedix wonders why a version number is a String in Wikidata [19:52:47] Perhaps because it sometimes contains letter [19:52:50] letters [19:52:55] good point [19:52:58] but I don't know the data types on WD [19:53:29] the datatype can be seen on the property pages: https://www.wikidata.org/wiki/Property:P348 [19:54:09] Yes [19:54:43] but do you have an answer to my question? Is WD ready for such use? [19:54:56] as far as I know, yes [19:55:17] I see. [19:55:31] it should be possible to use wikidata claims in templates on the wikipedias or other clients [19:56:05] excellent. I might start working on doing that on the Danish Wikipedia then [19:56:11] but I've never done this [19:56:13] for versions numbers [19:56:15] for now [20:00:14] with {{#property:P348}} you can include the version number in the wikipedia [20:00:29] Excellent [20:00:34] Thank you [20:00:50] (03PS1) 10Aude: Fix covers tag for NumberValidator [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/101360 [20:01:08] https://de.wikipedia.org/w/index.php?title=Mozilla_Firefox&diff=125417237&oldid=125385804 [20:01:39] I'm not sure how to access the qualifier that sais it was released on 2013-12-10 [20:01:55] I'll look into it [20:01:59] Thanks for your help [20:02:51] no problem! Thanks for your question, so I learned how to access WD from Wikipedia! [20:32:03] [travis-ci] wikimedia/mediawiki-extensions-Wikibase#1421 (master - b455bd3 : Translation updater bot): The build was broken. [20:32:03] [travis-ci] Change view : https://github.com/wikimedia/mediawiki-extensions-Wikibase/compare/bee475a740dd...b455bd364433 [20:32:03] [travis-ci] Build details : http://travis-ci.org/wikimedia/mediawiki-extensions-Wikibase/builds/15418287 [20:35:28] Lydia_WMDE: Definitely not a short list from me: https://www.mediawiki.org/wiki/User:Howief/WMF_Product_and_WikiData [20:35:49] James_F: ohhh dear :P [20:36:09] Lydia_WMDE: And I've probably not thought of a few dozen others. :-) [20:36:10] we will have to battle that out in SF [20:36:13] * James_F nods. [20:36:14] haha [20:36:15] right [20:36:23] And that's just for VE. [20:36:42] you have more? don't scare me! [20:36:44]