[07:53:31] yo [10:13:33] Haaapy Friday! [10:52:36] happy friday olliv [10:52:46] having lunch with the fam [11:08:35] brion: i see you are on the video again.. I'm sorry that I haven't been able to contribute much lately. I've just been so busy with work that I don't feel like doing wikimedia stuff at home, esp. with summer still on. [11:09:20] thedj: no worries! [11:09:40] And you put a huuuuge amount of good work in already. It's much appreciated :) [11:11:19] after oktober 1st, i'm winding down my weekends of sailing, so i might have more weekend time available after that. [11:12:03] in that regard, we can blame rachel, for encouraging to pick up my dream of learning to sail, to which i'm now hooked, costing me WMF time :) [11:12:28] Hehehe [11:13:54] I may or may not have things bashed into deployable state before I give my presentation at Demuxed in SF.... That's October 13. [11:14:13] Probably will end up cutting it a bit fine ;) [11:14:39] ha ! that's usually the best motivator to get stuff done though [11:14:46] Yep :) [11:19:42] oh, i thought i was in the multimedia channel.. sorry mobile folks ! [11:20:17] Well we're making it work on mobile too so that works eh hehe [11:20:32] Oh man it's 4am I should probably go to bed [11:20:49] Catch you later thedj ! [15:56:09] dbrant: niedzielski : are you aware that coming back from the search fragment or even when dismissing the keyboard there causes a weird bounce? [15:56:37] bearND: i see a flash. i think that's the main activity toolbar flipping back on [15:57:12] bearND: i think one way to fix it would be to make a searchactivity [15:57:20] T145834 [15:57:21] T145834: SearchFragment should be hosted in its own Activity - https://phabricator.wikimedia.org/T145834 [15:58:00] ...or at least its own ActionMode [15:58:40] yeah, there's some legacy code there we just didn't have time to get to really reformulating to the current app standards [15:59:08] i know i had written some action bar stuff a while back that was pretty convoluted. i think activities would be one way of eliminating it [16:09:29] hi - does someone know how difficult it would be to run the a/b ru and it tests only on anons (or can comment on https://phabricator.wikimedia.org/T136746)? I was under the impression that this was our original setup, and the change doesn't affect us from a data perspective, but the community liasons think that we should give a longer time period for [16:09:30] responses if we do it this way, which means we might need to push the a/b tests to oct 3 most likely. jhobs: perhaps you can comment? [16:11:08] olliv: we don't have a way to do that right now AFAIK. We'd have to build one, which would delay our timeline as well. Someone else might know better though, /cc jdlrobson phuedx|nom joakino bmansurov [16:13:05] olliv: (let's bring it up at standup) [16:16:23] (I'll also bring this up during standup) do you have a rough estimate on what the delay would look like/how long it would take? If fairly trivial, I'd say let's go with it and go live with the change by the end of next week. If longer, I'd say let's keep it as-is and wait on the deploy until Oct 3 (since week of sept 26 is still blocked off) [16:17:58] I'm not too familiar with the sampling code tbh. I think we use upstreamed (core) sampling code authored by phuedx|nom, so he might know, but if that is the case, trying to get anything merged into core takes time [16:21:19] niedzielski: whoops, rebasing [16:21:33] dbrant: thanks :) [17:30:13] dbrant: forgot to mention, these two tickets are part of the Nav MVP but are not estimated or in a sprint, though one has a patch for review. Should these be descoped? Pulled into the sprint? Something else? https://phabricator.wikimedia.org/T145683 https://phabricator.wikimedia.org/T141162 [17:40:38] olliv, jhobs: as bmansurov_ suggests, this //should// be a fairly minor change [17:41:19] i don't think it needs to be done at the mw.experiments layer [17:41:27] but it seems like an obvious addition to that library [17:41:43] so, let's make the change in the extension for now, and then upstream to core [17:42:14] phudx: I didn't even think of just doing it after the sampling [17:42:23] * jhobs facepalms hard [17:42:38] wait [17:43:00] sampling and choosing an experimental condition are different [17:43:13] sorry, yes [17:43:20] won't that mean the experiment is doing 10% of all users, some % of which will then be put back in the control bucket? [17:43:25] so it won't really be 10% of anons [17:44:06] don't enter the user into the experiment if they're logged in [17:44:28] right? (i'm in need of caffeine after a 6 hour car journey) [17:44:42] oh I thought the experiment bucketing happened before Popups got to it [17:44:52] I think my understanding of how this works is just all muddled [17:45:10] I'm even not seeing an e in your name anymore! ;) [17:45:15] s/name/nick [17:45:21] yup [17:46:43] For some reason I thought experimental bucketing was being handled by core and then Popups was looking at a hook or something [17:46:58] which doesn't make much sense, so I think I've just been overthinking it too much [17:58:52] * phuedx is the king of overthinking :) [18:00:21] dbrant: mdholloway|afk niedzielski bearND delayed by room squatters (ED pssh) [18:02:23] nzr: i like your ticket name [18:02:32] T145888 [18:02:32] T145888: Make top of the explore feed shine - https://phabricator.wikimedia.org/T145888 [18:14:36] niedzielski: :) [18:14:56] niedzielski: dbrant can anyone send me new beta :) [18:15:28] nzr: we're in meeting at the moment :) so soon [18:30:32] nzr: it's in the mail [19:25:23] niedzielski-afk: thanks! [19:26:48] niedzielski-afk: dbrant bearND hey, so i need to involve comms in the tagline "explore the world of free knowledge" just a note. as this is a blocker for next release. i think we can spend a day or two getting comms feedback [19:45:19] nzr: this is fine but please consider the turnaround time on translations. the longer we wait, the fewer translations we launch with [19:45:53] niedzielski: yes. I have a fallback version which is just a pattern background if things don't work out [19:46:12] concern around last minute comms consultations and changes :( [19:46:18] nzr: ok that's cool [19:46:41] nzr: if you have an idea of what the string will be, maybe we should put it in and just accept that it won't be a perfect translation for every language [21:14:55] dbrant bearND|brb: we're coming up on cut off time for pre-weekend beta i think. please review the search patch revisions if you haven't already [22:11:16] dbrant bearND|brb: release notes cooking in the usual place [22:44:05] dbrant bearND|brb: possible blocker: T145911 [22:44:05] T145911: [BUG] [API 16] Log in / out and settings overflow not available when search icon shown - https://phabricator.wikimedia.org/T145911 [22:59:57] niedzielski: I agree that could be a blocker for the release unless we bump the minSdkVersion for it [23:00:19] niedzielski: bearND: might it be more prudent to wait til Monday, then? [23:00:39] bearND: i think we wanted to hold off on that until after the overhaul [23:00:44] also: T145913 [23:00:44] T145913: [BUG] Changing page language to non-English and opening gallery shows duplicate images for single image articles - https://phabricator.wikimedia.org/T145913 [23:00:54] (not sure if this is a regression from the last release or not yet) [23:01:58] i think it's better to move the release to Monday. Hopefully we have at least the API 16 issue sorted then [23:02:36] bearND dbrant: sounds good [23:03:45] T145913 is not a regression from the last release [23:04:09] bearND dbrant: have a good weekend :) [23:04:15] later! [23:04:57] have a great weekend [23:05:20] thanks!