[00:27:06] is anyone else getting a broken wikidata logo? [00:28:15] the logo in the corner on wikidata.org is showing up fine for me, if that's what you're asking [00:29:28] strange, it's showing up blank for me. it's also trying to load the image from static-bugzilla.wikimedia.org, which is odd. [00:29:48] wtf [00:29:55] YairRand: Logged in or out? [00:30:01] logged in [00:30:19] also happening when in incognito mode [00:30:46] the relevant CSS seems to be .mw-wiki-logo { background-image: url(/static/images/project-logos/wikidatawiki.png); } [00:30:49] Try another browser [00:30:51] yeah [00:31:42] tried on IE, same isue [00:31:45] *issue [00:32:11] other Wikimedia projects have the logo working fine [00:32:28] including test.wikidata.org [00:32:50] "GET https://static-bugzilla.wikimedia.org/static/images/project-logos/wikidatawiki.png 404 (Not Found)" [00:33:18] https://www.wikidata.org/static/images/project-logos/wikidatawiki.png seems to be redirecting there [00:34:06] Why is static-Bugzilla involved in that? [00:36:13] do you ahve some kind of caching proxies in between you and WMF? [00:36:50] not as far as I know [00:37:05] Do you see a cached 301? [00:37:35] hoo: how do I check for that? [00:37:49] your browser's developer tools should tell you that [00:40:50] well, I cleared and disabled the cache, and the image is still failing to load [00:41:10] (couldn't figure out how to check whether the 301 was cached) [07:27:38] Lydia_WMDE: Page moves don't get processed again, I don't see hoo here... [07:28:27] sjoerddebruin: probably filled up the job queue so they are delayed again. will discuss if we can somehow give that more power. [07:28:55] Yeah. [07:36:31] sjoerddebruin: sent email to discuss options now [07:36:48] !lydia [07:36:48] \o/ [07:36:53] :P [07:37:23] Was working on German arena's. O2 stopped sponsoring them... [07:50:04] multichill: goodmorning. :) [07:51:54] hoi sjoerddebruin [08:05:41] https://www.wikidata.org/wiki/Special:Contributions/112.198.118.74 -_- [10:33:01] Jonas_WMDE_: Everything alright? [10:33:17] :/ [10:33:47] Lydia_WMDE: Is he in your walking range? ^ [10:34:03] sjoerddebruin: jep :) [10:34:04] wasup? [10:34:12] benestar: addshore: did you see the first message in the DataModel Serialization commit log? [10:34:16] The joining and leaving? :D [10:34:31] sjoerddebruin: ah :D he's setting up his computer atm [10:35:03] Ah, yes. [10:35:15] Little bit spammy. :P [12:29:56] hmm... isn't https://phabricator.wikimedia.org/T67722 really two issues, one about some formats not being parsed at all, one about some formats being ambiguous (and therefore parsed wrongly for some people)? [12:34:11] Jonas_WMDE: https://www.mediawiki.org/wiki/MediaWiki-Vagrant [12:36:38] benestar: this is what the post I was thinking off earlier http://martinfowler.com/articles/replaceThrowWithNotification.html [12:38:48] ty, will have a look [12:40:11] oh, I'm now getting the same issue that YairRand was earlier, where the logo doesn't work because it's trying to load it from static-bugzilla.wikimedia.org [12:40:15] benestar: did you get the phpstorm key yet? [12:40:34] nope :( [12:41:09] benestar: k, mailed it now [12:46:32] cute overload of the day: https://www.youtube.com/watch?v=V6ukn8aoUBY [14:31:39] benestar: should I instruct you how to do the rest of the backport and deploy? [14:32:17] if it's possible yes, but I guess you wanna do it now .. [14:33:07] benestar: click on cherry-pick in gerrit on the patch and fill in the name of the latest deployment "wmf" branch [14:33:13] jzerebecki: When will you leave for holidays? [14:33:21] hoo: 9th [14:34:11] Ok... I forgot a tshirt in Berlin... can you take that to Mexico for me, maybe? [14:34:12] hoo: we will disable property alias conflicts in a backport. should we swat that to group0 or let it ride on the train if possible? [14:34:17] Not sure when I'll be in Berlin again [14:34:25] hoo: probably. will try to fit it in. [14:34:55] If it's group0 I can make sure it's being pushed sometime before the train [14:35:12] jzerebecki: just let you do it ;) [14:35:31] hoo: ok [14:35:58] If were fast enough, we might be able to shove it into the early SWAT [14:36:21] benestar: BTW, the warning we added yesterday fired exactly once since :P [14:37:46] hoo: ok will try the swat. backport: https://gerrit.wikimedia.org/r/#/c/222121/ [14:38:03] hoo: btw, it's stopped now but a couple of hours ago I was getting the same weird issue with the logo that YairRand had last night [14:38:28] nikki: Very very weird [14:38:45] * nikki nods [14:38:58] Where are you physically located? (I don't need an exact place, just tell me the country or even continent) [14:39:03] germany [14:39:13] mh [14:39:53] hoo: o.O in what context? [14:40:12] hoo: we have 20min to the swat, need your +2 [14:40:51] The change looks somewhat scary to me at a glance [14:40:55] guess you tested it? [14:41:39] hoo: benestar tested it and I was siting next to him [14:41:44] After reading the bug that makes sense [14:45:34] jzerebecki: Will you update the build, or shall I? [14:51:21] * hoo does it, then [14:55:21] thx, was distracted with a meeting ;) [15:01:34] 32°C ... and I'm sitting underneath the roof :/ [15:07:35] hoo: will you do the submodule update, my checkout is a bit old? [15:07:49] Yeah [15:07:53] just waiting for jenkins [15:09:14] Thiemo_WMDE: :D [15:09:21] ! [remote rejected] HEAD -> refs/publish/production/jobrunnerWikibase (no changes made) [15:09:28] You did the exact same thing :P [15:09:43] ?? [15:09:59] The jobrunner patch you just rebased [15:10:08] I also rebased it [15:10:23] but failed to upload my version because you were faster ;) [15:10:45] hoo: havent done anything, only a rebase. [15:10:55] I know :) [15:11:01] And I did the same rebase [15:11:02] sorry, should have asked. [15:11:11] and apparently ended up with the exact same changes :D [15:11:34] now i feel stupid. :( [15:12:24] Why? I think it's funny that we ended up with the exact same diff [15:13:07] Slow jenkins is slow [15:31:01] benestar: jzerebecki: Deployed [15:31:43] hoo: not working [15:31:57] Uh [15:32:00] Property P82 already has label "Instance of" associated with language code en. [15:32:03] cool, thanks guys :) [15:32:17] I definitely did the submodule update [15:32:20] when trying to add alias CommonsName on https://test.wikidata.org/wiki/Property:P7 [15:33:13] also for conflict of an alias with an alias [15:33:35] Did that work locally? [15:34:17] hoo: we tested it [15:34:23] mh [15:34:36] we didn't remove label-label constraints [15:34:43] that's why it's failing [15:34:46] those still have to be resolved [15:35:34] Can you fix that? [15:35:46] We can do another deploy anytime before the train [15:35:55] Or revert, if needed [15:38:10] hoo: we can also remove the label uniqueness constraints but that doesn't make sense imo as they already exist [15:38:44] to be clear, the backported change only removes label-alias and alias-alias conflict detection [15:39:04] So it's ok now? [15:40:21] yes everything works as intended [15:41:48] Ok, great :) [15:42:01] In that case, I'll head for food [15:42:18] I'll be back in a bit and will be around for the deploy [15:42:55] Maybe only for the deploy... not sure, but definitely for that [16:16:14] Lydia_WMDE: btw, this came up a couple of days ago... interesting to imagine inline editing of wikidata-generated content, I wonder if it's important to let editors know how many articles they are vandalising at a time [16:18:32] Lydia_WMDE: full log here - the discussion is very broad, but several points of interest came up: http://bots.wmflabs.org/~wm-bot/logs/%23wikimedia-office/20150624.txt [16:19:07] DanielK_WMDE: Slightly offtopic... did you see my bug about scaling client change handling? [16:19:35] If change handling would work, that ticket has an answer to that question [16:19:44] hoo|away: yes, it's actually in the sprint (converted to an investigation)= [16:20:12] It's an investigation because I'm not sure that it needs fixing [16:20:25] but well, given the numbers, it probably does [16:22:49] hoo|away: seems to work well so far. but we should look into a batching mechanism anyway [16:23:20] anyone want to merge https://gerrit.wikimedia.org/r/222139 to fix our tests? ;) [16:23:24] Did anyone really change anything that has a lot of usages? [16:24:47] cheers hoo|away ! [16:26:17] Let's try changing an item that has ~8k 'X' usages and see what happens [16:26:21] on ruwiki, these are [16:29:03] sjoerddebruin: https://www.wikidata.org/wiki/Wikidata:Database_reports/Constraint_violations/P1970 looks quite fixable :-) [16:30:00] hoo: yaya [16:30:07] testing in production right? ;) [16:30:27] I'm just editing a Wikipage everyone can edit [16:34:08] benestar: http://wikiba.se/components/ [16:34:39] what about that? [16:36:21] benestar: has been updated, including the change you commented on [16:44:59] sjoerddebruin: https://gerrit.wikimedia.org/r/208397 <- hopefully fixing lagging page moves [16:54:12] Lydia_WMDE: Will that also speed up the time that a change on Wikidata gets processed on WIkipedia? [17:01:41] Could someone involved with DM3.0 please look into https://phabricator.wikimedia.org/T104405 ASAP [17:01:53] Might be more severe than I originally thought [17:01:56] and the train starts in an hour [17:03:02] Thiemo_WMDE: --^ [17:08:19] benestar: food arrived [17:11:34] multichill: I've asked the author of https://www.wikidata.org/wiki/Q15877689 on nlwiki to split his article. [18:02:05] hoo: will you update the build for the new dm release? [18:02:17] Yep, can do [18:06:24] hoo: will you need me for anything else (you can still ping me in textsecure or call if something araises)? if not I would go out. [18:06:40] No, I don't think so [18:06:45] Have a nice evening :) [18:06:48] Will you be here tomorrow? [18:07:01] For the Wikipedia deploy [18:19:29] hoo: I have something scheduled earlier, but not sure how long that takes. Will you be there anyway and you could phone me or should i watch it primarily? [18:20:02] I'm not going to be at home, but I will have my notebook with me and can watch it [18:20:35] ok cu [18:20:54] cu :) [19:00:33] benestar: https://github.com/wmde/WikibaseDataModelSerialization/pull/151 [19:01:31] benestar: if you tag, make sure to do x.y.z [19:01:34] not x.y [19:01:36] so 1.5.0. [19:01:38] err [19:01:40] 1.5.0 [19:01:50] sure [19:02:05] benestar: else the layout here will be less nice http://wikiba.se/components/ :D [19:02:15] Look at that silly DataValues Number... [19:02:50] it shows "build error" for me o.O [19:02:59] perhaps because of the broken version number ^^ [19:05:06] sjoerddebruin: Did you already create the second item? [19:05:20] multichill: Going to do that now [19:05:25] pywikibot.data.api.APIError: readonly: The wiki is currently in read-only mode [19:05:31] :-( [19:06:30] multichill: wikidata? [19:07:31] Yup and with WDQ lagging more than a day I have to be very careful with restarting [19:08:00] Oh, only 4 minutes ago [19:08:40] benestar: you released it, dun dun dun [19:09:21] DM serialization 1.5.0 released [19:09:53] addshore: yeah, now your evil code is going to be used in production o.O [19:10:12] addshore: so time for you to fix it in the api [19:10:12] evil, wmahahaa [19:10:17] is there already a patch on gerrit? [19:10:29] or did you start working on it already [19:10:29] ? [19:10:37] nope [19:11:13] Travis is buggy... keeps running into timeouts, loading stuff via composer fails, ... :/ [19:12:30] That is not Travis being buggy ;p [19:12:49] mh [19:12:51] github, then [19:13:01] http://docs.travis-ci.com/user/build-timeouts/ [19:13:10] travis_retry can be of use [19:13:20] I know I've put it in place for some things [19:14:53] addshore: did you install FAF? ;p [19:14:58] benestar: http://www.faforever.com/ [19:15:18] benestar: you can get it for only 10 EUR at http://store.steampowered.com/app/9420/ [20:41:59] benestar|cloud: Ran into another readonly, see https://phabricator.wikimedia.org/T100123 [21:25:16] Hm, the font size of the first heading changed... [21:25:36] yes [21:25:58] Weird padding under it imo [21:27:07] Oh, the sitelinks has different sizings too... [21:27:33] yeah, see https://www.mediawiki.org/wiki/Wikidata_deployment [21:31:29] hmph... now the "1 entry" bit is way too small [21:32:00] +1 [21:32:09] Not sure if my unofficial OCD can handle this. ;) [21:32:42] I predict some edits to my common.css page in the near future :P [21:33:36] benestar: ^ [21:33:53] hi Lydia_WMDE, what's up? [21:34:18] benestar: see last few sentences by sjoerddebruin and nikki [21:34:25] sjoerddebruin nikki: it's all my guilty :P [21:34:30] ;-) [21:34:37] It's hard to read imo. https://www.dropbox.com/s/y8k2tkr8z43dhnk/Schermafdruk%202015-07-01%2023.34.30.png?dl=0 [21:36:23] sjoerddebruin: do you know how to handle developer tools in your browser? [21:36:36] Do I look like a idiot? ;) [21:36:36] * nikki does [21:37:16] :P can you try to change the font size to 75%? [21:37:28] https://www.dropbox.com/s/ndov9vflwu61r1m/Schermafdruk%202015-07-01%2023.37.26.png?dl=0 [21:37:32] it's currently 67 % so maybe these tiny bits solve it [21:37:54] or 80%, ,90% tell me what's best in your opinino [21:38:14] * benestar thinks 75% should be fine [21:38:46] 75 is looking better yes [21:38:54] hm, it would be easier to compare with a screenshot I had of it before if it hadn't changed font too [21:38:59] I think 75% looks about the same, yeah [21:43:38] will the font be changing back to serif too, or is that something I should just go put in my common.css already? :) [21:44:34] nikki: it won't since vector defines

elements as sans-serif [21:45:29] k [21:45:38] I'll never understand that, but at least I can fix it for me :D [21:45:59] nikki: that's defined by Vector, not by us [21:47:38] I know, I was referring to why the people who made the decisions about vector made those decisions [21:47:56] Where is that alias coming from? https://www.dropbox.com/s/ju7mejqluz02d2z/Schermafdruk%202015-07-01%2023.47.50.png?dl=0 [21:48:58] I think it's setting the alias to whatever has been used to find the entity [21:49:01] in that case the id [21:49:03] but not sure [21:49:07] addshore would know [21:49:11] maybe benestar does [21:49:27] Ah, yes. [21:50:32] sjoerddebruin: yes, addshore worked on that and the idea is to show to the user why the entity was found [21:50:54] We should perhaps just remove the sentence before the value [21:51:58] at least no double values anymore [21:52:07] aliases [21:53:57] nikki sjoerddebruin: we could also increase the overall size of those sitelink headings [21:54:14] They don't look bigger than the text yes [21:56:14] that would work too