[00:00:19] i'm getting ops to roll it back [00:00:35] tfinc: ? rolling back? [00:00:51] I'm not sure what to do about it, awjr_away gave it -1 [00:01:13] tfinc, roll back what? [00:01:27] yeah I'll fire off another email jgonera thanks for looking [00:01:35] jgonera: their rolling back their SSL change [00:01:41] as far as i can tell none of knew it was happening [00:01:46] it's not that big, can we include it tomorrow if need be? [00:02:12] tfinc, oh, ok [00:02:46] alrighty folx I need to head to Dimi's play tfinc jcmish marktraceur brion anything you need [00:02:50] ? [00:03:23] ha Maryana, jgonera ^ autocorrected the wrong names [00:04:03] Project MobileFrontend-mac-ipad build #332: UNSTABLE in 10 min: https://wmf.ci.cloudbees.com/job/MobileFrontend-mac-ipad/332/ [00:04:03] * gerrit: Revert "Maintenance: Load minimum possible css / reorg css" [00:04:04] * jgonera: Rotate escape overlay arrow for RTL languages [00:04:04] * jgonera: PNGCrush all the PNGs [00:04:08] I think we're good [00:04:36] Project MobileFrontend-mac-iphone build #335: UNSTABLE in 9 min 56 sec: https://wmf.ci.cloudbees.com/job/MobileFrontend-mac-iphone/335/ [00:04:37] * gerrit: Revert "Maintenance: Load minimum possible css / reorg css" [00:04:38] * jgonera: Rotate escape overlay arrow for RTL languages [00:04:38] * jgonera: PNGCrush all the PNGs [00:06:42] tfinc: what are the cert problems you see? [00:06:52] en.m.wikipedia.org is fine in chrome on android for me [00:06:53] brion: this new cert doesn't include .m [00:07:05] i got ssl warnings on my phone and chome [00:07:08] in what way doesn't it include .m? [00:07:09] desktop [00:07:35] the hell [00:07:49] • Server's certificate does not match the URL. [00:08:06] so how on earth does that work on my mobile devices o_O [00:08:07] the cert is only for *.wikipedia.org [00:08:17] brion: no clue. it doesn't work on mine [00:08:22] The certificate is only valid for the following names: [00:08:23] *.wikipedia.org , wikipedia.org , wikibooks.org , m.wikibooks.org , *.wikibooks.org , *.m.wikibooks.org , wikidata.org , m.wikidata.org , *.wikidata.org , *.m.wikidata.org , wikimedia.org , m.wikimedia.org , *.wikimedia.org , *.m.wikimedia.org , wikimediafoundation.org , m.wikimediafoundation.org , *.wikimediafoundation.org , *.m.wikimediafoundation.org , wikinews.org , m.wikinews.org , *.wikinews.org , *.m.wikinews.org , [00:08:24] wikiquote.org , m.wikiquote.org , *.wikiquote.org , *.m.wikiquote.org , wikisource.org , m.wikisource.org , *.wikisource.org , *.m.wikisource.org , wikiversity.org , m.wikiversity.org , *.wikiversity.org , *.m.wikiversity.org , wikivoyage.org , m.wikivoyage.org , *.wikivoyage.org , *.m.wikivoyage.org , wiktionary.org , m.wiktionary.org , *.wiktionary.org , *.m.wiktionary.org [00:08:31] no m.wikipedia.org in there, indeed [00:08:33] the hell [00:08:36] brion: when does yours expire ? [00:08:39] it used to be only m.wikipedia.org [00:08:42] so hey [00:09:03] ah there we go, wasn't on https :P [00:09:06] that's damn odd [00:10:56] this is just one of those days eh [00:10:58] sighhhhh [00:11:04] New patchset: JGonera; "Remove unused images" [mediawiki/extensions/MobileFrontend] (master) - https://gerrit.wikimedia.org/r/53493 [00:22:53] New patchset: JGonera; "Remove unused images" [mediawiki/extensions/MobileFrontend] (master) - https://gerrit.wikimedia.org/r/53493 [00:24:35] hey tfinc, how close are we to having account creation on the apps? [00:24:44] Maryana, should we do the mingle magic and make the next iteration the current one? [00:25:02] jgonera - err, i think i broke something last time i tried that [00:25:34] Maryana: were doing the server side work this sprint and user next [00:25:46] awesome, thanks :) [00:27:34] brion knows that were foaming at the mouth to get it [00:27:53] hehe. yeah, we're just meeting w/jay & matthew to discuss the marketing plan [00:28:10] doesn't make sense to market it widely to people who don't already know what commons is/have an accout without account creation [00:28:24] brion: nexus4 used sparingly for email but no calls today still 80% batter charged [00:28:28] i'm impressed [00:28:40] not bad [03:30:26] New review: JGonera; "(2 comments)" [mediawiki/extensions/MobileFrontend] (master) C: -1; - https://gerrit.wikimedia.org/r/52317 [04:08:37] New patchset: JGonera; "Fix jagged inset shadow in menu" [mediawiki/extensions/MobileFrontend] (master) - https://gerrit.wikimedia.org/r/53514 [04:24:04] New patchset: JGonera; "Increase left padding in search input" [mediawiki/extensions/MobileFrontend] (master) - https://gerrit.wikimedia.org/r/53515 [07:24:08] New review: Nikerabbit; "Just fyi there are tools like optipng that produce even better results than pngcrush." [mediawiki/extensions/MobileFrontend] (master) - https://gerrit.wikimedia.org/r/53486 [08:45:28] [android-commons] yuvipanda pushed 1 new commit to master: http://git.io/rlwHnw [08:45:28] android-commons/master 10b7a97 YuviPanda: Do not crash if Android doesn't know the mimetype of the file [08:45:58] [android-commons] yuvipanda force-pushed beta from 29abc87 to 13e2349: http://git.io/tUPHIg [08:45:58] android-commons/beta 13e2349 YuviPanda: Single squashed commit for converting master into beta [08:46:24] Project Android-Commons (mobile) - Nightly builds build #131: SUCCESS in 40 sec: https://integration.mediawiki.org/ci/job/Android-Commons%20(mobile)%20-%20Nightly%20builds/131/ [08:46:25] * yuvipanda: Do not crash if Android doesn't know the mimetype of the file [08:46:25] * yuvipanda: Single squashed commit for converting master into beta [10:39:32] [android-commons] yuvipanda force-pushed beta from 13e2349 to 7a3b388: http://git.io/tUPHIg [10:39:32] android-commons/beta b9f6c65 YuviPanda: Refactored StartUploadActivity to it's own separate class [10:39:32] android-commons/beta 7a3b388 YuviPanda: Single squashed commit for converting master into beta [10:39:53] Project Android-Commons (mobile) - Nightly builds build #132: SUCCESS in 36 sec: https://integration.mediawiki.org/ci/job/Android-Commons%20(mobile)%20-%20Nightly%20builds/132/ [10:39:54] * yuvipanda: Refactored StartUploadActivity to it's own separate class [10:39:54] * yuvipanda: Single squashed commit for converting master into beta [10:43:37] [android-commons] yuvipanda fast-forwarded master from 10b7a97 to b9f6c65: http://git.io/_BjSYg [15:10:10] New review: Jdlrobson; "Not sure if I'm allowed to merge due to deployment sigh.. :-(" [mediawiki/extensions/MobileFrontend] (master); V: 2 C: 1; - https://gerrit.wikimedia.org/r/53514 [15:13:12] New patchset: Jdlrobson; "Increase left padding in search input" [mediawiki/extensions/MobileFrontend] (master) - https://gerrit.wikimedia.org/r/53515 [15:14:36] New review: Jdlrobson; "Now the added empty file is gone this is good. merge at will." [mediawiki/extensions/MobileFrontend] (master); V: 1 C: 1; - https://gerrit.wikimedia.org/r/53515 [15:33:41] New patchset: Jdlrobson; "Bug 45516: Rewrite getToken function" [mediawiki/extensions/MobileFrontend] (master) - https://gerrit.wikimedia.org/r/52317 [15:42:33] New review: Jdlrobson; "(1 comment)" [mediawiki/extensions/MobileFrontend] (master) - https://gerrit.wikimedia.org/r/52054 [15:48:02] New review: Jdlrobson; "(1 comment)" [mediawiki/extensions/MobileFrontend] (master) - https://gerrit.wikimedia.org/r/52054 [15:48:05] awjr_away: still away? [16:13:41] jdlrobson: yo! yt? [16:13:45] awjr: yo [16:13:47] yes [16:13:52] how's sxsw? [16:13:56] so the device stuff is making me very anxious! [16:14:03] yeh sxsw is great. The interactive portion has ended now [16:14:03] device stuff? [16:14:03] s [16:14:24] https://gerrit.wikimedia.org/r/#/c/52054/ < awjr [16:14:29] without that patch desktop banners can get into mobile [16:15:00] jdlrobson: how so? sorry, i just don't really have any context [16:15:14] so CentralNotice code review has been quicker [16:15:24] and they pushed a patch that enabled CentralNotice on mobile [16:15:36] there is no way to disbale CN on mobile? [16:15:37] and added a test against that device property to check if the user is on mobile or desktop [16:15:49] in the case of mobile they don't add a banner currently [16:16:03] but since we are not setting that deviceName in the config it thinks it is desktop [16:16:30] does CN only need to know if the device is mobile, or does it need to know the specific device? [16:16:38] i'm not sure if CN have deployed yet though [16:16:52] the existence of that config variable is all that is currently needed [16:16:54] if you can send me the link to a CN patchset i can test [16:16:58] s/test/check [16:17:13] $vars[ 'wgMobileDeviceName' ] = true would suffice [16:17:23] awjr: just use latest master of CentralNoticer [16:17:27] i've already tested [16:17:33] i have a desktop banner in mobile on beta [16:17:37] jdlrobson: i want to check if it has been deployed [16:18:08] awjr: 5f364806cac1f0c711316d5905429250a2853185 [16:18:33] jdlrobson: thanks [16:18:48] awjr: is your only grievance about the name of the config variable? [16:18:54] jdlrobson: no [16:19:09] jdlrobson: if CN only needs to know if we are in the mobile view, why are we returning device id? [16:19:19] because we want to target android/iphones [16:19:22] that was part of the story [16:19:30] so it also serves for that purpose [16:19:48] so we can target banners for mobile/android or mobile/iphone [16:20:04] ok - for a little more context, part of my concern is that the way it is being done right now is going to complicate implementing esi [16:20:14] how so? [16:22:18] because it depends on back-end detection and relies on core features for loading the appropriate script for a given device [16:22:54] im not sure how we'd wrap what is currently there in an esi chunk without having to touch core [16:22:59] how does device id effect that?! [16:23:17] which bit of code specifically? [16:23:28] device is dependent on the specific device (determined by X-Device) [16:24:11] awjr: i don't understand - how does adding a new property to device detection cause this issue? [16:24:23] *device properties [16:24:50] adding a new device property does not inherently cause the issue, but the way in which the device-specific scripts get loaded does [16:25:01] device specific scripts? [16:25:04] which are..? [16:25:35] er rather device-specific variables to be passed to javascript [16:26:00] awjr: i don't understand why that would be the case [16:26:22] the wgTitle variable varies per page for example [16:27:07] jdlrobson: if we remove the dependency on the X-Device header and stop caching resources separately for each device, how will we ensure that the appropriate device id is being sent to javascript when cached resources are served? [16:28:13] i know that we can rewrite this when we get around to implementing ESI in a couple weeks, but it would be nice to avoid having to if we can right now [16:28:21] awjr: I'd hope RL takes care of that… I mean I'd be worried to why the onMakeGlobalVariablesScript exists if it cannot handle that [16:28:49] how would RL handle that if we're using PHP to set the device id? [16:29:20] if a cached resource is being served, that code will not run [16:30:04] well the device id is served via javascript… via the onMakeGlobalVariablesScript [16:30:46] right but that variable will be cached along with the javascript [16:32:16] i don't understand why it would be - i mean mw.config already contains various things that can change across pages (wgImagesDisabled for example, wgTitle for another, Geo for a third...) [16:32:56] page specific != device specific [16:33:13] awjr: i guess i don't know enough about how this thing works [16:33:22] but i don't see how else i'd do it [16:33:33] the variable needs to be there from the start [16:34:08] could do lightweight device detection in js [16:34:14] too late [16:34:21] banner code is designed to run straight away [16:34:37] (to avoid flashes) [16:35:44] ok jdlrobson lemme poke around for a bit; worst case scenario we can deploy this and fix it later [16:36:32] ok anyway brb just doing some washing [16:37:27] btw we should really have a way to configurably enable/disable CN banners in MF [16:49:52] New patchset: awjrichards; "Make site notice display in mobile configurable" [mediawiki/extensions/MobileFrontend] (master) - https://gerrit.wikimedia.org/r/53601 [16:54:30] awjr: it will still load the js though - i can't remember if siteNotice gets created if absent.. [16:55:11] jdlrobson: also, js/css resources do not vary on X-Device, so i am concerned that that patchset will not work reliably. the first time that the resource is requested, that variable will be set for the device that is requesting it, and that value will get cached along with the resource, so it may not be accurate the next time it is requested and loaded from cache [16:55:43] mm [16:55:53] awjr: what's the point of html( 'notice' ) ?> [16:55:58] is that zero? [16:56:02] where? [16:56:08] https://gerrit.wikimedia.org/r/#/c/53601/1/includes/skins/SkinMobileTemplate.php [16:56:11] above the green [16:56:18] hey notnarayanq [16:56:20] err [16:56:22] hey notnarayan [16:56:24] hmm [16:56:24] hey brion [16:56:41] hey [16:56:41] how's the eye? [16:56:58] brion: much better. it wouldn't open for about 15 minutes when I woke up, was terrifying. [16:57:22] but after a shower + some more time, it got much better than yesterday [16:57:34] jdlrobson: it might be old cruft; it gets set by the hook 'GetMobileNotice' [16:57:41] mmm [16:57:47] YuviPanda: did you see a doctor? [16:57:56] which is not invoked anywhere (at least in production [16:58:06] no. if it had persisted like that, or as bad as it was yesterday night, i would've [16:58:11] awjr: are you able to fix mingle? [16:58:15] but it is only a small swelling now. Blinking does not hurt [16:58:17] the stuff from last iteration should be moved over.. [16:58:24] and as usual, I am ignoring my body for the most point and ignoring it :) [16:58:28] yikes [16:59:04] jdlrobson: i did know it was broken [16:59:14] what is the problem? [16:59:39] current iteration story wall awjr just shows the current iteration - none of the stuff still in code review / not done from last iteration [17:00:24] and previous iteration story wall doesn't show me the last story wall [17:00:24] so i don't know how to get to it [17:01:13] oh, i sent an email how to update that a few weeks ago [17:02:01] jdlrobson: standup [17:03:12] jdlrobson: no horse boss? [17:03:19] awjr: did you just say 'bible thumpers at the door'? [17:04:04] YuviPanda: yes :p [17:13:51] New review: Jdlrobson; "(2 comments)" [mediawiki/extensions/MobileFrontend] (master) C: -1; - https://gerrit.wikimedia.org/r/53601 [17:14:08] awjr: sent in notes. [17:14:29] tfinc: sent in notes anyway [17:14:32] notnarayan: I like the look of the campaign designs but i think it may need better affordances; it's a bit unclear how one navigates it [17:14:34] tfinc: can you grab siebrand ? [17:14:34] YuviPanda: thanks [17:14:35] thanks YuviPanda :) [17:14:40] i want him to check https://gerrit.wikimedia.org/r/#/c/51205/ :) [17:14:44] ragesoss: can you try *current* nightly? it should fix the upload issue [17:14:47] and the crashes [17:14:52] tfinc give me a few minutes before we sync up - i need to fix mingle [17:14:55] also awjr am i allowed to merge code right now? [17:14:56] brion: and grab siebrand for iOS TWN:) [17:14:59] jdlrobson: nope. i need to check in with arthur. get brion or maryana to wrangle him [17:15:16] i'm scheduled to grab siebrand on friday :) [17:15:27] jdlrobson: there's a Language ENgineering team office hours going on right now, so doubt you can catch him [17:15:28] jdlrobson: no, please wait until we get the sitenotice issue fixed [17:15:32] brion: yes, ill fix that. [17:15:41] k [17:15:49] YuviPanda: I think the certificates are still hosed. [17:16:17] I can't log in for test.wiki (and so, can [17:16:24] things appear to have gone a little sideways in mingle [17:16:26] 't easily download the new nightly) [17:16:39] ragesoss: bit.ly/commons-beta :) [17:16:39] New review: Jdlrobson; "This needs to be rebased but can then get merged." [mediawiki/extensions/MobileFrontend] (master) C: 1; - https://gerrit.wikimedia.org/r/48209 [17:16:46] let me try [17:16:54] New patchset: Jdlrobson; "WIP: Avoid SkinMobile-specific functions and properties" [mediawiki/extensions/MobileFrontend] (master) - https://gerrit.wikimedia.org/r/48209 [17:16:58] YuviPanda: I still won't be able unless the certicates are fixed, right? [17:17:23] ragesoss: i'm able to upload [17:17:23] so the 'current iteration' variable is set correctly in mingle, however the story wall is being filteerd on 'last iteration' [17:17:25] to testwiki [17:17:37] brion: YuviPanda why would the bad cert affect apps that are driven by the api ? [17:17:51] awjr: msg me when your ready [17:17:55] tfinc: it affects all HTTPS, doesn't matter if it is API or not [17:18:03] okay. I still get a crash when I try to log in, with the previous nightly. will fetch the new one. [17:18:08] ragesoss: but it seems okay now. try? [17:18:10] tfinc: if we're using the commons.wikimedia.org endpoint it shouldn't affect us, but if we used commons.m.wikimedia.org then it would [17:18:13] yeah, try the new one [17:18:16] it hink [17:18:18] k [17:18:28] this is to testwiki, but we are using commons. [17:18:37] this was a problem a few hours ago, should be fixed now [17:19:50] urgh, i think i broke mingle [17:19:52] ok the story wall should be fixed now jdlrobson Maryana [17:19:53] sorry, team :( [17:19:57] oh good [17:19:58] New review: Jdlrobson; "Brion is it worth still considering this patch after:" [mediawiki/extensions/MobileFrontend] (master) - https://gerrit.wikimedia.org/r/51287 [17:19:59] :) [17:20:01] mostly at least [17:20:03] tfinc ok ready [17:20:03] thanks awjr [17:20:14] hm, it now says "filter is invalid" for me [17:20:20] refresh, Maryana [17:20:29] ahh, there we go [17:20:31] thank you! [17:20:34] np [17:20:40] Change abandoned: Brion VIBBER; "Mozilla is actively working on a fix, so that should be out there before there's much use of FFOS. A..." [mediawiki/extensions/MobileFrontend] (master) - https://gerrit.wikimedia.org/r/51287 [17:21:03] awjr: i'm thinking of moving the review to forward a week or two. i'd like to time it before we deep dive into planning for the perf sprint. whats the sweet spot ? [17:21:22] tfinc you are specifically talking about Mobile Caching Architectural Review [TENTATIVE] ? [17:21:29] this will give us more time to prep and max can actually attend. which is critical as i expect both you and max to lead it [17:21:31] awjr: yes [17:21:59] New patchset: Jdlrobson; "Cleanup: Introduce mobilediff.css" [mediawiki/extensions/MobileFrontend] (master) - https://gerrit.wikimedia.org/r/52934 [17:22:33] tfinc ok - i am still a little unclear on what the purpose/agenda is for that [17:23:05] * jdlrobson needs lunch - anything before i go? [17:23:30] awjr: we'd get the architects together: mark, tim, etc and yourself + max to do a deeper dive into the performance issues ops is seeing [17:23:30] tfinc: but timing-wise we should shoot for the morning of 3/18 or 3/19 at the latest [17:23:50] i expect to flush out the agenda as the ops/mobile sync up @ 11 today w/ mark and ct [17:23:58] tfinc ok cool [17:24:10] YuviPanda: still crashes on login [17:24:20] ragesoss: sent crash report? [17:24:26] yeah [17:24:27] got it [17:24:28] yep [17:25:09] awjr: it can't be super morning as thats far too in the AM for Tim. i'm trying to find a time with Tim but he hasn't responded to my mails. but those dates help [17:25:54] tfinc ok; well it needs to happen with lead time for our estimation/review meeting next week. max is usually toast by 4pm pacific, and we have the quarterly review as well as deployment on monday/tuesday respectively [17:26:04] right [17:26:28] 3/19 looks pretty open right now [17:26:32] brion: fixed :) http://upload.wikimedia.org/wikipedia/commons/c/c3/Campaign-workflows.png [17:26:42] tfinc yup and if we need to, we can bump deployment to weds [17:26:54] or i should say, we can probably bump deployment to wds [17:26:59] wooop [17:27:17] ragesoss: interesting. the problem seems to be SSL for bits [17:27:22] YuviPanda: http://upload.wikimedia.org/wikipedia/commons/c/c3/Campaign-workflows.png [17:28:02] awjr: you were also asking about the needs section on http://etherpad.wikimedia.org/MobileArchitectureReview as well [17:28:04] ragesoss: does https://test.m.wikipedia.org still error for you? [17:28:07] (on the phone) [17:28:08] ? [17:28:14] those were pointers from Erik and RobLA about how to organize the talk better [17:28:26] notnarayan: can oyu make them pdfs next time? [17:28:28] the pngs are *HUUUGE* [17:28:30] so we'd need to prep the diagrams beforehand [17:28:36] and on my connection take about 3 minutes to load [17:28:36] brion: ping [17:28:42] tfinc: yeah i was wondering what the mobile team should be prepared with as well as what the ops/platform team should be prepared with [17:28:52] Just wanted to check to make sure you weren't waiting on me for anything for that review [17:28:53] YuviPanda: ok [17:28:53] YuviPanda: yes, test.m is still borked [17:28:57] given our conversations with mark thus far i'm seeing two focus areas for the review [17:28:58] ragesoss: clear cache? [17:29:04] general caching and dynamic sesctions [17:29:05] ragesoss: it isn't for me [17:29:07] I suspect it's in queue, but just wanted to check [17:29:42] so i'd need you and max to create those. if we go with 3/19 will you have enough time to create them ? [17:29:57] notnarayan: can you upload a pdf version now? it has been stuck about 20% of the way in now [17:30:33] awjr: --^ [17:30:40] tfinc: yeah one sec, thinking [17:32:14] i should tfinc. and fwiw we've already discussed not pushing dynamic sections (as currently implemented) to produciton, but i guess it would still be worth deciding whether or not it makes sense to approach another way or at all [17:32:47] awjr: i think its extra important to teach them about why we are using them in order to make them comfortable with it. as we'll look at them later [17:32:50] the earlier the better there [17:33:02] so i'm thinking of a general caching review and dynamic sections review [17:33:03] yeah [17:33:08] bbondy: howdy [17:33:11] notnarayan: can you ping me after the pdf upload? I tried refresh but my connection just can't handle that [17:33:13] thats what i'll pitch to mark @ 11 and we'll see what he thinks [17:33:24] k [17:33:25] YuviPanda: uploading right now [17:33:29] hey [17:33:42] (messages below that ping btw) [17:33:48] bbondy: ah yes you've got more patches on the win8 yes? i just updated my win8 machine yesterday, i'll poke those in a bit [17:33:52] jgonera: how did the sync up with ops on monday go? [17:34:02] YuviPanda: test.m seems to work on other browsers (easier than clearing cache) [17:34:17] ya I'll be doing more just wanted that share html bug to land first [17:34:34] did the initial review comments [17:34:46] awjr, pretty quick, what they would like to know is some data about our watchlist usage (related to moving pageimages to stable) but I wasn't able to give it to them [17:34:46] ragesoss: sounds like a cache issue. [17:34:47] :( [17:34:53] ragesoss: it perhaps cacched the cert? [17:34:55] YuviPanda: the png is 2.6 mb while the pdf seems to be 5.2mb [17:35:01] ... [17:35:06] that's weird. [17:35:08] jgonera: Maryana should be able to hook that up [17:35:17] shouldn't it be smaller since it is not rasterizing them all? [17:35:31] what am i hooking up? [17:35:38] awjr: what big changes do we have going out next week ? [17:36:00] tfinc do you mean next week or today? for either, i don't know - i'm still catching up [17:36:52] YuviPanda: no idea. why thats happening. any other ideas? [17:37:04] tfinc, we're probably going to deploy calls to action from the upload feature for logged out users next week [17:37:11] which should get a lot more attention on uploads [17:37:27] and also a centralauth fix/workaround [17:38:48] New review: awjrichards; "The other problem with this is that resources loaded via ResourceLoader do not vary the cache by dev..." [mediawiki/extensions/MobileFrontend] (master) - https://gerrit.wikimedia.org/r/52054 [17:39:52] YuviPanda: here is a jpg version of the same. its 548kb in size http://upload.wikimedia.org/wikipedia/commons/d/d5/Campaign-workflows-jpeg.jpg [17:41:11] New patchset: awjrichards; "Make site notice display in mobile configurable" [mediawiki/extensions/MobileFrontend] (master) - https://gerrit.wikimedia.org/r/53601 [17:41:39] awjr: next week. i'm looking at https://docs.google.com/a/wikimedia.org/spreadsheet/ccc?key=0Aoizbfxc5g6KdEkza0xkQnJlM0o0TXlwQXhDOUFvYnc and seeing if it needs an update [17:41:44] Engineering Projects Roadmap [17:42:03] awjr: any other questions about the review ? [17:42:22] i want to make sure you feel comfortable to lead the discussion [17:42:31] especially since its coming up fast [17:42:42] YuviPanda: I cleared data from the app and it still crashes. [17:43:10] tfinc i think im set for the architectural review for now - i really don't have time to focus on it right now though. i will tomorrow though and let you know if i have more questions [17:43:12] ok [17:43:51] tfinc also i'll let you know about the roadmap later this afternoon - i still need to catch up on what's slated for the current sprint and where we're at [17:43:57] k [17:44:09] awjr: let me know if you want a later week 1:1 to catch up [17:44:49] tfinc - can we plan to sync up later this afternoon? i think that would be good [17:45:07] tfinc maybe 5 or 530? [17:45:10] sure [17:45:15] cool thanks [17:47:56] New review: awjrichards; "(2 comments)" [mediawiki/extensions/MobileFrontend] (master) - https://gerrit.wikimedia.org/r/53601 [17:48:40] jgonera, brion: can you guys take a look at ^? i'd like to get that in as part of the dpeloyment today so we can make sure CN doesn't render banners on MobileFrontend until we have https://gerrit.wikimedia.org/r/#/c/52054/4 ironed out [17:49:20] Maryana, do you know how many people use the watchlist on stable or in total? or how many logged in people we have per some unit of time [17:49:54] jgonera, so i actually don't - we're only logging number of people who tap the star, not watchlist views [17:50:12] ok, that's what I thought [17:50:15] it's about 3,000 users tapping the star each month [17:50:16] if that helps [17:50:34] we are looking at whether users remove items from their watchlist from the watchlist view, and that number is tiny [17:50:35] New review: Brion VIBBER; "So this'll still load sitenotice-related code, but it won't actually do anything because it has no <..." [mediawiki/extensions/MobileFrontend] (master) - https://gerrit.wikimedia.org/r/53601 [17:50:37] jgonera: i wonder if diederik can help figure that out then? [17:51:05] i talked to diederik last week about giving us some prelim pageview stats broken out by article/special namespace [17:51:09] that should give us a rough estimate [17:51:10] awjr, probably, I already asked him but forgot to reply to his last e-mail, doing it now... [17:51:16] New review: awjrichards; "basically, yeah. it's probably not idea, but will at least prevent desktop notices being displayed o..." [mediawiki/extensions/MobileFrontend] (master) - https://gerrit.wikimedia.org/r/53601 [17:51:25] since currently there's only the watchlist special page in stable [17:51:32] YuviPanda: can we update http://mobile-reportcard.wmflabs.org/ by hand today? [17:52:01] i think diederik is still waiting on the analytics piece that awjr is working on, though [17:52:09] that's required for them to give us those stats [17:52:17] that mobile reportcard page looks terrible on mobile. :/ [17:52:23] haha [17:54:17] tfinc: yeah, updating now [17:55:30] New patchset: JGonera; "Make watchstar rotate when it changes state" [mediawiki/extensions/MobileFrontend] (master) - https://gerrit.wikimedia.org/r/53609 [18:03:21] tfinc: it is up to date now [18:03:34] tfinc: check in incognito if it doesn't feel like [18:03:44] YuviPanda, how do you test limn-mobile-data? do you just edit it on stat1? [18:04:03] jgonera: i test by deploying to mobile-reportcard-dev [18:04:21] and how do you deploy there? [18:04:23] jgonera: i've a git repo on stat1, and I've added it as a remote on my local. So I commit on local, push to stat1, run it from there. [18:04:40] jgonera: you need access to kripke on betalabs (ask on -analytics to add you) [18:04:55] jgonera: then you clone github.com/wikimedia/limn-deploy on your local machine [18:04:59] to add me where? I have a stat1 account [18:05:06] hm [18:05:14] jgonera: then you do 'fab mobile_dev deploy.only_data' [18:05:22] jgonera: to kripke, which is where the reportcards are hosted [18:05:34] heh. I think I'm responsible for that big error spike yesterday. [18:05:34] jgonera: stat1 generates stats, kripke is what hosts the limn instance [18:05:44] ragesoss: that counts only commons. not you :P [18:05:45] ok, I see [18:06:12] jgonera: *or* you could just setup limn locally [18:06:13] jgonera: gonna be much simpler. I had issues doing that and hence am using this [18:06:43] jgonera: and if you are in the office, you can also go grab milimetric (Dan) if you are running into probloems setting up limn [18:06:56] jgonera: it's probably going to be much faster for you than doing whatever I was doing [18:07:23] hehe, ok, I'll try that first, but I'll ask about access to kripke anyway [18:07:48] Maryana: re: the strategy document on wiki. apps already has 70 unique uploaders so far, so '100 a month' is going to be super easy to get :D I won't be surprised if we sail past that next week [18:08:02] nice! [18:08:03] jgonera: okay :) [18:08:15] i imagine, though, that the new uniques will taper off after awhile [18:08:39] maybe. But I doubt that's going to be at 100. [18:08:53] 212 files uploaded from Android! [18:09:01] 114 from iOS! [18:09:25] and a majority of the Android uploads seem non-english, which makes me very happy [18:09:52] YuviPanda, how's it possible that there's more uploaders than successful uploads? [18:09:59] (looking at http://mobile-reportcard.wmflabs.org/) [18:10:17] jgonera: uploaders is cumulative. [18:10:21]