[08:31:15] morning y'all [08:31:32] morning phuedx_car [08:31:58] the amount of work i do from the car, it might be worth buying a portable wifi ap [08:32:14] that said, my iphone works just fine [08:32:24] 4g coverage in the uk is great [08:32:38] nice :) [08:32:52] i've found 3g to be good too (not for meetings obvs) [08:33:50] yeah, meetings wouldn't be viable [08:34:02] face-to-face that is [08:35:34] good morning :) [08:36:34] morning phedenskog [08:38:32] joakino: I'm curious, was you able to reproduce the poor performance on your phone when you turned on the other applications? [08:41:16] and would either of you phuedx or joakino wanna team up with me to make a new Grafana dashboard for mobile later this week? [08:41:18] phedenskog: didn't try, moved to other things after reading some google searches and seeing the open bug on android/chrome https://code.google.com/p/android/issues/detail?id=81950 https://lastpass.com/support.php?cmd=showfaq&id=8166 [08:41:47] but everything got better when i disabled accesibility and removed lastpass, so i'm assuming that was the culprit [08:42:07] phuedx: I got my pi now, waiting on the antenna ... [08:42:13] SOON [08:42:20] :D [08:42:31] joakino: yeah, but I meann users use lastpass and have accesibility enabled. [08:42:33] * phuedx is travelling at the moment, but would like to see you go through the set up [08:43:34] ah reading the issue now [08:43:37] sad :) [08:43:41] yeep [08:50:24] phuedx: lets do it when you have time there's no rush but would be cool if we do it together [09:28:30] phuedx: sup with https://phabricator.wikimedia.org/T135554#2343809 :p [09:28:39] is it good or not? [11:41:52] heading out for lunch in a bit [13:00:55] joakino: my internet connection dropped so i couldn't update the comment [13:01:22] ie support is borked due to the mw.popups.createSVGMask failing [13:43:42] phedenskog: the graph you posted on T133565… [13:43:43] T133565: Images do not load until scroll event completes (not during scroll) on iOS - https://phabricator.wikimedia.org/T133565 [13:43:44] HOW!? [14:12:08] that is, how did you generate it? [14:25:04] niedzielski: hey, do you have a minute for a batcave chat about the most-read endpoint? [14:25:57] niedzielski: brb, gonna order another coffee [14:27:07] mdholloway: sure. i'll set up shop there for whenever you get back [14:27:15] niedzielski: omw [15:02:24] ping jhobs [15:02:43] (meetin' ) [15:03:11] phuedx: oh my bad thought this one was gonna be optional [15:03:13] brt [15:06:28] phuedx: it's webpagetest [15:06:35] rly? [15:06:35] let me show you tomorrow [15:06:41] i hadn't seen that output [15:07:19] it's the waterfall graph [15:08:57] hrrm [15:09:01] i might be misremembering [15:48:11] bearND: should i remove wikidata description from my endpoint patch for now? it's rfr otherwise [15:53:46] bearND: lmk, i'll be back shortly [15:53:54] * mdholloway -> coworking [15:57:40] maxbinder, jdlrobson, jhobs, joakino, bmansurov, dr0ptp4kt: thanks for that meeting [15:57:45] i found it incredibly useful [15:58:02] phuedx: glad to hear, happy to talk process anytime :-D [15:58:24] thanks to the team for suggesting it during retro [16:08:08] mdholloway|biab: I'd say leave the WD description in for now since we're not storing the results in Cassandra yet. gwicke: When do you expect change propagation of WD descriptions to work? [16:32:47] bearND: i'll change the /routes file for the aggregated endpoint to aggregated.js to avoid the conflict with your featured.js [16:33:45] mdholloway: thanks [16:35:12] bearND: I just chatted with Petr about this; it turns out that there is an end point for wikidata sitelinks already, which we can use [16:35:21] he is planning to tackle this in the next days [16:36:03] gwicke: very nice! Awesome! [16:36:23] the general logic to page through such a list is already in place, it's mainly a matter of configuring it [16:36:28] mdholloway: ^ [16:38:22] bearND: gwicke [16:38:29] excellent news! [16:38:34] (sorry, bad at irc) [16:38:38] ;) [16:38:57] bad at irc'ing while eating a salad, anyway [16:39:19] bon appetit ;) [16:39:30] thanks :) [16:39:34] * gwicke suddenly feels hungry [16:44:53] phuedx: same here [17:01:28] phuedx: joakino standerp [17:03:53] maxbinder: having router issues, go on without me [17:04:00] joakino: k [17:04:05] joakino: plz email [17:47:42] jhobs: hey [17:47:47] -1ed your patch - but not far from completion [17:59:36] jdlrobson Hi :) [17:59:48] i audephone ! good name for this channel ;-) [18:00:06] Heh [18:00:22] Not sure I can easily answer on phabricator from my phone [18:00:38] But about vagtant [18:00:45] Vagrant * [18:01:37] If you disable the wikidata role can you still access the wiki? [18:02:46] Or in local settings set wmgUseWikibaseRepo to false [18:05:38] audephone: last time i disabled wikidata role i broke it so i was keen to avoid that [18:05:55] Maybe just change the setting then [18:05:57] but i was a little confused to how i would be able to access it. Setting set wmgUseWikibaseRepo to false seems better [18:06:10] You have to find it unfortunately [18:06:43] Then move the main page to a wikitext namespace or just delete [18:06:52] And recreate [18:07:25] Somehow it shouldn't be hard for mediawiki installer to do this [18:10:56] audephone: that looks much more straightforward [18:10:58] doing as we speak [18:11:11] Ok [18:11:53] jhobs: did you get my message above? [18:12:37] jdlrobson: sorry, IRC didn't make a noise [18:12:41] * jhobs reads up... [18:12:52] I'll take a look [18:13:03] audephone: sort of worked. I dont get error any more but no main page either :) [18:13:09] no big deal [18:13:12] That's ij [18:13:14] Ok [18:13:32] You can recreate it in the project namespace if you care [18:13:50] And update the mainpage system message [18:13:54] jdlrobson: does that step exist and if not, how would one verify that in selenium? [18:14:03] jhobs: see my follow up patch [18:14:07] oh k [18:14:20] https://gerrit.wikimedia.org/r/#/c/292190/ [18:14:24] See MediaWiki:Mainpage on wikidata as example [18:14:45] when we push to stable that step can become empty / be removed [18:15:54] jdlrobson: gotcha. Want me to just make my patch depend on yours, or copy over the same step & LocalSettings? [18:16:04] jhobs: you can squash mine into yours no big deal [18:16:13] i just wnat to make sure you understand the changes and agree with them [18:16:47] ok cool. I'm not sure how to do that properly in gerrit though. Or can I just git review -d yours and merge the branch into mine and gerrit will take care of the rest? [18:17:05] i'll look a bit closer, but preliminary inspection makes sense [18:17:55] jdlrobson: ^ [18:20:40] dbrant: sweet! [18:21:36] mdholloway: thx! [18:36:11] jdlrobson: hopefully I did that right [18:38:44] bmansurov: would you be able to monitor tomorrow afternoon's swat for the 1% hovercards on hu.wikipedia.org? [18:38:55] dr0ptp4kt: yes [18:39:29] bmansurov: thx. jhobs ^ would you please ready the 1% patch for hu.wikipedia.org and get it on the swat calendar for tomorrow's afternoon swat? [18:39:57] dr0ptp4kt, bmansurov: got it [18:40:07] jhobs: thx [18:40:27] dr0ptp4kt: so jhobs gets it to the calendar, and I'll be available during deploy right? [18:40:44] yes [18:40:48] ok [18:44:16] jdlrobson: want to meet now instead of waiting? [18:44:33] jhobs: bmansurov thx [18:44:47] sure dr [18:44:49] dr0ptp4kt: [18:44:54] tracy? [18:45:00] jdlrobson: i'll take dr :) [18:45:06] jdlrobson: tracy it is [19:03:03] bmansurov: do you know the variable to set for the percentage deployment of hovercards? I'm having trouble finding it and am a bit confused by "wmgPopupsBetaFeature" as well [19:03:44] and/or do you have 5min for a quick hangout? [19:20:51] dr0ptp4kt: we're enabling Hovercards for 1% of users on huwiki, but the EL sampling is supposed to be at 10% of those, yes? (That's the default set on the extension.) [19:21:14] the card simply says "Schema:Popups should continue logging at some percentage for Hovercards-enabled and non-Hovercards-enabled alike." [19:24:17] jhobs: yep. as i understand there's no need to change the config variables for el sampling. the change should be confined to enabling the experiment. then users will have an equal probability of being sampled in, irrespective of their hovercards on/off status. [19:24:35] dr0ptp4kt: ok cool so we're on the same page then [19:24:38] s/config variables/config variable/ (there's just the one global in the extension itself as i understand [19:25:00] jhobs: yep, i think so! it's mainly enabling the experiment via the mechanism phuedx provided [19:27:21] jhobs: i'll have time after break [19:27:25] bmansurov: think i figured it out, but please review https://gerrit.wikimedia.org/r/#/c/292206/1. Not 100% certain I understand how some of the magic variables work [19:27:28] dbrant: looks like grooming is mostly gonna be triage today. You guys good without me? [19:27:36] jhobs: ok [19:27:49] maxbinder: yep, sure [19:52:51] jdlrobson: jhobs bmansurov do you plan to be at the backlog grooming meeting this afternoon with maxbinder and me? if we don't have quorum, we can skip this one instance. jhobs would you like us to get this started one hour earlier in general? if so, i think maxbinder and i would need to reschedule a regular meeting or two to make it fit. [19:53:23] to be clear, i think we should keep holding two instances until we've cleared the to-triage column. but hopefully that's temporary [19:53:28] dr0ptp4kt: i was going to join, but i'm happy to skip [19:53:31] (like not a lot more than a quarter) [19:53:42] jdlrobson: jhobs ^^ you? [19:53:57] Can do yes [19:55:53] dr0ptp4kt: I think we can still find stuff to do even if the whole team isn't there. Are you looking for more time to catch up on memorial day cruft? [19:56:23] dr0ptp4kt: especially since this morning was fairly groomless [20:06:54] bearND: mdholloway: single-character fix -- https://gerrit.wikimedia.org/r/292212 [20:07:21] dbrant: awesome. have you tested it? [20:07:23] dbrant: saw that :D looking now [20:07:32] bearND: yep, that does it. [20:08:01] dbrant: that looks good to me [20:08:02] the ">" notation selects only the immediate child tag, instead of all (sub)children tags. [20:10:21] dr0ptp4kt: i was planning to be there, yes. I don't mind it being later since I generally just leave an hour earlier the following day or friday [20:10:31] dr0ptp4kt: i'm ok with skipping this one if you two have conflicts though [20:11:06] no conflicts. bmansurov maxbinder jhobs i'll plan joining (cc jdlrobson ) [20:14:55] bearND: mdholloway: when's the next deployment of the service to production? (and does this require re-generation of any kind?) [20:15:48] dbrant: bearND: sometime in the next 45 minutes, and i think so, yes [20:17:33] dbrant: yes, it requires re-generation [20:33:15] dr0ptp4kt: in hangout [20:33:54] maxbinder: i'll be there shortly. thanks for waiting [20:34:00] k [21:37:24] * mdholloway -> home