[02:26:15] * tomaszf sits down to the more budget work [09:51:41] MaxSem: That setup.py branch is merged. :-) [09:54:19] whee [10:05:34] MaxSem: Would you like commit access to the main vumi-wikipedia repo? [10:05:59] It would probably be easier than working in your own fork. [10:10:00] jerith, I'm currenty not feeling comfortable about my Python skills [10:10:28] though if you're using gated trunk, I guess I can't inflict much damage:} [10:10:45] MaxSem: That's fine. Our workflow is based heavily on code reviews. [10:11:04] We use "git flow" for our repos. [10:11:49] All work happens in feature branches, we submit pull requests when we want reviews and then it gets merged to "develop" (which is basically "trunk") when everyone's happy. [10:12:21] okay [10:14:05] MaxSem: There are often people around in #vumi who can help you out if I'm not available. [10:14:26] And you have commit access to the repo now. [10:14:31] thanks [16:16:22] * ajerith contemplates SMS-only vumi-wikipedia. [16:58:33] greetings all [16:58:44] Hey tfinc. [16:59:30] hey ajerith [16:59:45] jdlrobson: do you have a fix for: https://gerrit.wikimedia.org/r/#change,3693 [16:59:48] looks like we have everyone but yuvi on [17:00:16] preilly: yes see latest comment [17:00:31] all right .. lets do our standup [17:05:39] * YuviPanda waves [17:05:43] standup started? [17:05:55] * YuviPanda pokes tfinc [17:06:20] * YuviPanda pokes MaxSem awjr [17:06:36] yup, call tfinc [17:08:08] hmm, did I get disconnected? [17:08:24] FAIL [17:08:48] just me or everyone? [17:10:39] * YuviPanda looks around [17:10:56] Just you. [17:12:34] sigh [17:12:49] * YuviPanda pokes MaxSem awjr tfinc again [17:13:12] * ajerith heads to band practice for the evening. [17:13:25] doctor, help - everyone keeps ignoring me! [17:13:32] next patient! [17:13:51] MaxSem: me too, i'm not even on the call [17:13:56] i don't even know if there's a call [17:13:59] wait [17:14:00] MaxSem: I may be online sporadically from band. I'll be home about three and a half hours from now and at your disposal then. [17:14:04] that's a joke, i didn't get it. [17:14:06] gah [17:14:07] well, there used to be call [17:14:08] * YuviPanda facepalms [17:14:27] did it end prematurely? [17:14:34] ajerith, okay [17:15:13] MaxSem: +1 on splitting the API code out into its own module, by the way. :-) [17:15:23] :) [17:16:13] * YuviPanda pokes maplebed [17:16:15] preilly sanity check this RegEx for me? /AppleWebKit\/(53[456789]|5[456789]|[6789])/) should match any version over 534 [17:16:21] teehee! [17:16:33] YuviPanda: ! [17:16:43] lets do you standup over irc [17:16:46] your* [17:17:01] tfinc: okay. [17:17:10] i dialed you a number of times but never got a response [17:17:14] Friday was spent poking people and getting lots of feedback, most of it which has been fine [17:17:22] how is RC1 looking ? [17:17:23] tfinc: i got it a bit late, and was in the call for about a minute [17:17:25] then got disconnected [17:17:35] eh it happens [17:17:37] no [17:17:38] sorry about that [17:17:58] heard something about brion and windows marketplace :) [17:17:58] anyway [17:18:05] yes [17:18:19] jdlrobson: looking at that regex; is it the case that the only 3 digit version numbers always start 53? [17:18:20] there are some ui/ux people coming from microsoft on ewd [17:18:23] wed* [17:18:24] tfinc: i think there's one thing we should fix, which is the facebook share behavior. [17:18:30] what do you mean? [17:18:35] that all highe numbers (54, 6, etc.) are 1 or two digits? [17:18:38] also. i'm worried about the amount of change happening today [17:18:40] both beta+apps [17:18:44] i think it may be too much [17:18:57] tfinc: first time you hit share, you see only login. you need to hit share again tos hare [17:18:59] *to share [17:19:09] jdlrobson: if what you're looking for is 3 digit numbers larger than 534, that regex isn't what you want. [17:19:19] jdlrobson, use ranges insted of lists, i.e. /AppleWebKit\/(53[4-9]|5[4-9]\d|[6789]\d\d)/ [17:19:27] jdlrobson: /AppleWebKit\/(534| 5[456789][\d])/ [17:19:33] YuviPanda: you mean that the first time you hit share, it only lets you log in. then you have to tap it again to really share something ? [17:19:34] tfinc: i talked with jdlrobson, beta -> main should not kill app. i'll do more vigorous testing now. [17:19:39] tfinc: yes. [17:19:49] tfinc: that's unless you're already logged in. [17:19:53] first time ever, that is. [17:20:01] YuviPanda: yes. and remember you need to test it with both 1.1 AND 1.03 [17:20:04] maplebed: it depends how webkit version numbers work but I think it should also match AppleWebKit/54 [17:20:04] +1 MaxSem [17:20:13] it takes up to a week for our users to upgrade to a newer version [17:20:14] it just shouldn't match anything under 534.. [17:20:15] tfinc: yes! [17:20:24] that's our final thing we do before we 'release' [17:20:27] yes ranges make much more sense [17:20:36] I'm writing a test framework and would like to use your wiki mobile as an example on how we can run the same test cases on Android and iOS. Would that be ok? [17:20:58] YuviPanda: so. are you still on track to get the app out today? [17:21:07] YuviPanda: we also need to draft a blog post for it [17:21:11] i can help with that [17:21:15] tfinc: yes. [17:21:51] tfinc: awesome. [17:22:14] maturana: awesome! [17:22:25] maturana, it's open source:P [17:22:27] maturana: that sounds interesting. lets us know more. [17:22:28] MaxSem: maplebed /AppleWebKit\/(53[4-9]|5[4-9]|[6-9])/ is what i want I think [17:23:00] jdlrobson: that won't match 546. [17:23:09] you want what MaxSem wrote but with a ? after the final \d. [17:23:29] yes it will.. [17:23:37] maplebed, we need unit tests! [17:23:41] /AppleWebKit\/(53[4-9]|5[4-9]\d?|[6789]\d?\d?)/ [17:23:51] preilly: awjr : do we have any non beta changes going out today that could affect the app push? [17:23:55] "AppleWebKit/546".match(/AppleWebKit\/(53[4-9]|5[4-9]|[6-9])/) does indeed match [17:24:02] i'm debating about doing the app push tomorrow to lessen the amount of moving parts today [17:24:05] jdlrobson: because the match is anchored with a / on the right, you're expressing a strict limit on the number of digits. [17:24:20] oh. [17:24:24] but im not doing a full reg expression test here [17:24:24] The frameworks are open source calabash-ios and calabash-android. We're just building support for hybrid apps now and would like to make a promotion video and that kind of things with a "real" app. We be using it for live demos at workshops and conferences as well. [17:24:26] the / is not part of your regex but the delimiter. [17:24:44] yeh sorry my bad - that was confusing [17:25:26] I would still add in the optional \ds to make it clear what you're going for rather than implicitly accepting them because the string isn't bounded on either side. [17:26:52] tfinc: hmmm [17:26:55] * jdlrobson nods [17:27:14] tfinc: we could do that too, if you want. [17:27:52] YuviPanda: do we have any launch blockers ? [17:28:43] for the app? [17:28:52] no [17:29:13] * YuviPanda starts more testing with the beta [17:29:31] YuviPanda: yes [17:30:07] heads up that yet again i'll be in budget meetings for the whole day [17:30:14] this should be the last one for a bit [17:32:28] Got disconnected :( [17:32:31] The frameworks are open source calabash-ios and calabash-android. We're just building support for hybrid apps now and would like to make a promotion video and that kind of things with a "real" app. We be using it for live demos at workshops and conferences as well. [17:32:49] jdlrobson: do you have any changes for the blog post you'd like to get in before i stake it on techblog ? [17:33:15] nope [17:33:23] it's all good to go as far as i'm concerned tfinc [17:33:59] maturana: that'd be nice [17:34:54] MaxSem: what's our tracking bug for these new perf issues ? [17:35:09] * jdlrobson gets to work on footer changes [17:35:42] jdlrobson: i think we hit a decent compromise for the footer. good enough for the beta. [17:35:52] how will it look in under languages ? [17:36:10] tfinc, nowhere:) they are already resolved [17:36:14] other* [17:36:30] we haven't resolved the 15ms -> 150ms performance drop that patrick was talking about [17:36:38] just to be clear... im working to these designs: http://www.mediawiki.org/wiki/File:Mobile_footer_B.png http://www.mediawiki.org/wiki/File:Mobile_footer_B_expand.png [17:36:38] YuviPanda: If we get some nice tests written we could try to run your nightly build against our device cloud at LessPainful. I don't know how much automated ui test you have already. [17:36:57] maturana: none so far, but we'd like to change that [17:37:10] maturana: we NEED to change that :) [17:37:47] we need to and we also want to [17:37:58] Cool ;) [17:38:30] * tfinc tries to find the tracking bug for the nav bar [17:38:40] philinje: greetings [17:39:33] tfinc, it was caused by regression in profiling code, ie there's no performance regression [17:39:42] We might be able to help with that. Can't promise how much time we can put into writing the tests, but we could let you run testing at no charge. [17:40:08] maturana: that would be really helpful [17:40:25] were currently have an open RFP for someone to come and do these QA builds outs for us [17:41:13] maturana: http://hire.jobvite.com/Jobvite/Job.aspx?j=ogRaWfwX&c=qSa9VfwQ <-- do you know anyone who would be a good fit for this? [17:41:28] hmm .. i can't seem to find our nav bar bug [17:41:33] need to mark some bugs as blocked by it [17:42:05] here we go 32117 [17:44:17] awjr: two new otrs tickets in the queue about disabling not working [17:45:06] tfinc: Are you looking for someone who can write the test for *all* your supported platforms? [17:45:26] maturana: i'm looking for someone who will come in and create our qa pipeline [17:45:37] and creating tests will be part of their job [17:47:00] tfinc: can you forward them to me? [17:47:04] sure [17:47:45] awjr: sent [17:47:53] thnx [17:48:01] were getting too many reports about it for it to be isolated [17:49:18] tfinc: is there a way i can communicate with the people filing the otrs tickets? [17:49:34] tfinc: I'm not sure I know anyone with experience on all platforms but i'll think about it. [17:49:44] awjr: do you have an otrs account? [17:49:51] maturana: thanks [17:50:08] tfinc: possibly but i dont think so [17:50:20] if i do i don't think i've ever used it [17:50:28] tfinc: Does it make sense that we write test for Android and iOS if you're looking for someome to write it yourself? [17:50:40] maturana: yes [17:50:55] awjr: get an account then you can just respond to these directly [17:51:06] tfinc: is there documentation somewhere? [17:52:02] tfinc: or can you jsut tell me how to get an account? [17:52:26] awjr: hmm..i got mine ages ago [17:52:37] either from phillipe, casey, or james [17:52:47] tfinc: The calabash projects are under EPL would that be a problem. It explicitly says GPLv2 in the RFP [17:53:12] tfinc: from the language of both of those tickets i think the submitters might not have clicked 'Desktop view' but i'd like to confirm with them. i think they're missing the old, verbose 'permanantly disable the mobile version of this site' [17:53:16] maturana: i'm not familiar with the EPL so i'd have to send it to the lawyers [17:53:27] awjr: lets hope its just that [17:53:45] yeah, im going to try and confirm with them [17:55:22] tfinc: Alright. We'll write some tests and then get back to you. Then we can figure out what to do. [17:55:30] maturana: thanks! [17:57:12] YuviPanda: have you had a chance/ability to test going from the current iOS app -> 3.1 ? [17:57:58] jdlrobson: how goes crafting a list of test phones ? [17:58:10] greetings Jarry1250 [17:58:12] busy day here [17:58:46] Hey hey. [17:59:14] o/ [18:00:45] tfinc my list so far is basically all the blackberrys and symbians you own. :) these are the ones i have the most headaches teting [18:01:13] jdlrobson: k. if thats the case send me an email and i'll make you a care package :) [18:07:07] awjr: we are looking good in https://gerrit.wikimedia.org/r/#q,extensions/MobileFrontend+status:+open,n,z code review [18:07:20] awjr: so, I think we should be fine deploying from trunk today [18:07:24] preilly that is absolutely baller. [18:07:29] awjr: which is totally nice [18:07:38] hellz yeah that will be a lot easier [18:08:07] awjr: should be able to just update your repo and diff the git and svn directories to get all the changes [18:08:26] awjr: also, WURFL will be gone as well [18:08:38] awjr: so, we should become a bit faster as well [18:08:45] preilly yeah should be pretty straightforward. although i guess i won't be able to easily pick up all the revision ids [18:08:55] * preilly really seems to use the term, "as well"… [18:09:18] i do as well [18:09:27] awjr: well, you can based on Change Id [18:09:42] awjr: the older imported from SVN doesn't have them does it? [18:09:57] preilly not sure i'll have to poke [18:14:28] YuviPanda: i think we should block the app release on beta and wait till tomorrow. thoughts ? [18:14:42] fine by me. [18:14:52] YuviPanda: i don't think that changes anything for you [18:15:00] YuviPanda: just makes today a bit less insane for us [18:15:05] +1 [18:15:19] i'll just test until i'm confident then pick off 1.2 [18:15:21] for future weeks lets offset mobile web and app by a week [18:15:31] YuviPanda: start on the blog post [18:15:38] we want that ready to go [18:15:39] right. [18:15:50] YuviPanda: we have lots of other mobile blog posts on tech blog [18:15:54] awjr: it looks fairly straight forward — http://pastebin.mozilla.org/1535260 [18:15:56] test then blog post. [18:16:07] and i'll share the beta post were doing for beta with you [18:16:09] tfinc: we're doing a 'what's new + iOS!' blog post, right? [18:16:13] not a 'nuts and bolts' one [18:16:22] YuviPanda: yes [18:16:30] YuviPanda: also include android uptake stats [18:16:30] cool [18:16:33] i can export you a csv [18:16:36] super! [18:16:37] do that [18:17:02] YuviPanda: shared [18:17:58] cool [18:19:26] we should also make sure that the opt-in beta link in the blog post actually works correctly and use bit.ly for the stats collection [18:19:47] I think the old link for beta opt-in is broken [18:20:09] tfinc: I think you missed me saying, "we should also make sure that the opt-in beta link in the blog post actually works correctly and use bit.ly for the stats collection". [18:21:02] preilly: what broke with it ? [18:22:14] heatherw: footer changes -> http://jonrobson.me.uk/wikipedia/footer.html let me know what you think [18:23:22] tfinc: let me check again [18:23:26] jdlrobson: how many languages do we have the Wikipedia ® image for ? [18:23:42] also .. any chance we can do that with text rather then an image ? [18:23:51] tfinc: it points to http://en.wikipedia.org/wiki/?mobileaction=opt_in_mobile_site [18:23:52] ah... good question. Only 1 at the moment [18:23:56] jdlrobson: thats bad [18:23:59] lets do it as text [18:24:07] then we can pick up / re-use translations [18:25:34] jdlrobson: looks good to me! [18:26:01] how can we do it as text? not use the word mark? [18:26:04] well the alt text will get translated tfinc but the image at the current time won't [18:26:10] my local git repo for MobileFrontend seems to have … disappeared. [18:26:12] is that still R? tfinc [18:26:23] the 'MobileFrontend' directory is there, but it is completely empty [18:26:47] awjr: well, that's not good [18:26:51] no, it's not [18:27:02] i had done a bunch of contact us work in a local branch [18:27:02] awjr: did you do some sleep "rm -rf''ing [18:27:17] preilly it's possible i have a pultergeist [18:27:26] What does git status say? [18:27:50] jeroid the dir is completely empty so git doesnt even say it's a repository [18:28:06] Ah. ;_; [18:28:09] awjr: wrong folder? [18:28:11] these are dirs in a VM that i use for development [18:28:22] YuviPanda that's what i thought at first but it's not [18:28:29] it's right next to my clone of core [18:28:48] argh [18:29:31] awjr: heads that phillipe is flying back today [18:29:34] my inclination is to blame the vm or vm manager. it's really weird though, the MobileFrontend directory wouldn't even exist where it is now without a git clone [18:29:41] awjr: hopefully that email will catch him at the airport [18:29:49] tfinc: ok thanks [18:30:24] tfinc: that might be a reason to go back to legal? the word itself, is it registered? (i have no idea how that stuff works) [18:31:04] heatherw: do you mean if we have it registered in other languages ? [18:31:22] did we not render these in other languages ? [18:31:54] i mean, if legal wanted to switch to the marks, not just the word. [18:32:20] we do have "wikipedia" in other languages, i don't know if adding an ® to them is correct. [18:32:47] heatherw: right … hmm [18:32:58] tfinc: the wikipedia logo could point to logo_copyright_.png and use an alt text [18:33:10] although that would lead to a lot of 404s [18:33:17] jdlrobson: that would be full of fail :( [18:33:23] (i could also revert to text where it's not english) [18:33:26] thats a launch block then [18:33:34] awjr: There isn't done kind of strange overlay filesystem involved, is there? [18:33:43] :/ [18:34:14] jeroid well i use fuse to locally mount my dev directories from the VM to my local machine [18:34:25] i wonder if that caused some serious weirdness [18:34:34] but i'd never had issues in the past [18:34:56] Maybe something got unmounted somewhere? [18:35:16] * jeroid had that happen once. [18:35:28] jeroid im in the VM now directly and … well it's just not making sense [18:36:04] what's weird is i had one other extension checked out form git, and its dir is totally empty also [18:37:43] i wonder if something went horribly wrong with fuse/sshfs [18:38:03] but i dont know why it would happen to only those directories which were sub-sub directories of the mountpoint [18:38:19] jdlrobson: heatherw talking with kelly about non english text now [18:38:20] i had probably a day's worth of work only committed locally :( [18:38:25] awjr: :( [18:38:27] k tfinc keep me updated! [18:39:54] * awjr enables auto-snapshotting in vmware [18:40:54] That's one of the reasons I push regularly. [18:41:16] jeroid yeah but this stuff was not pushable - experimental and unfinished [18:42:45] awjr: http://zero-testing.wikipedia.wmflabs.org/xhprof_html/callgraph.php?run=4f70b887534df&source=mobile_frontend_extension&all=1 [18:44:16] jdlrobson: kelly is punting to our trademark lawyer michelle who's not in yet. i just txt'd michelle so hopefully well get a response [18:45:16] awjr: and http://zero-testing.wikipedia.wmflabs.org/xhprof_html/?run=4f70b887534df&source=mobile_frontend_extension&all=1 [18:46:55] jdlrobson: worse case though. lets think about our options [18:47:07] awjr: try apt-get install scalpel [18:47:23] awjr: then scalpel /dev/sda1 -o output [18:47:27] jdlrobson: we could use the image or (ideally) text with ® on en and just the localized text for Wikipedia on the others [18:48:01] preilly what's it do? [18:48:02] preilly: i never knew about scalpel .. awesome [18:48:18] awjr: essentially undelete [18:48:20] ! [18:48:47] but it seems to work at the device level [18:48:59] awjr: just do a sudo scalpel "device name/Directoryname/file name" -o "output directory" [18:49:19] where name/Directoryname/file name is the file you're looking for? [18:49:31] jdlrobson: i mean its not as if we show our marks anywhere now on the desktop :) [18:49:53] correct tfinc that would be the simplest option, alternatively we could throw the label altogether [18:49:54] awjr: yeah [18:50:16] the marks aren't even on the logos anywhere http://wikimediafoundation.org/wiki/Wikimedia_official_marks/About_the_official_Marks [18:50:17] we might also encode the url of the image in the localized string to allow contributors where applicable to provide an alternative logo for their given language [18:51:00] jdlrobson: do we have access to the localized string for wikipedia in the footer? or will we have to wait for a round of translate wiki ? [18:51:02] awjr: also, you'll need to configure the program to recover file types by uncommenting the respective file types you need to recover in the config file. For those who may not be familiar with commenting and uncommenting, here is a brief explanation: Commenting a line of code means placing a '#' at the beginning of the line so that the compiler/interpretor will ignore this particular line during program execution. In order for the recovery be [18:51:02] productive, you MUST uncomment the file types you wish to recover in the configuration file. The config file is located in the /etc directory, as shown below: [18:51:20] a round of translation tfinc [18:51:23] tfinc: you'll have to wait [18:51:42] tfinc: also, we need to insure that the translate folks translate it correctly ;) [18:52:09] tfinc: as, some of ours aren't actually correct — but, more agreed upon in nature [18:53:01] preilly: seriously ? [18:53:02] wow [18:53:05] thats scary [18:53:14] no MW:message we can pull from ? [18:53:47] tfinc: brion says NOPE [18:54:06] * tfinc signs with dismay [18:54:08] tfinc: because it comes from the config not the message cache [18:54:21] * preilly didn't know tfinc knew sign language [18:54:36] my experience in just working on indic languages, my computer couldn't read half of them so they all turned out wrong, svg should have been reliable but was not always. [18:54:41] * brion speaks fluent sigh language [18:54:44] *SIGHHHHHH* [18:54:48] :) [18:54:52] * tfinc didn't know that tfinc could both do a budget meeting and stay active in here [18:55:01] * heatherw is impressed [18:55:09] * preilly paralinguistic respiration FTW [18:55:20] :) [18:55:21] way to be nerdy brion [18:55:22] :D [18:55:30] haha [18:55:42] paralinguistic .. wow … nice [18:56:13] * preilly — thinks that it's 5 dollar word Monday... [18:56:38] yes! [18:56:55] any iterm2 users in here ? [18:57:01] i stumbled on to it over the weekend [18:57:09] Used it a while ago [18:57:13] but now on the default term [18:57:15] pretty nice it is [18:57:24] * preilly is glad that he didn't say, "pharyngeal fricative"... [18:57:31] tfinc: nope [18:57:50] Terminal.app for me [18:58:01] way to be old school preilly [18:58:31] tfinc: iterm2 ftw. [18:59:00] * tfinc shakes fist at kids these days spoiled with their color terminals. yells .. "back in my day we could count all of our terminal colors on one hand … and we were thankful for all the shades of green" [18:59:24] tfinc: Amber for me. [18:59:39] None of this fancy green stuff. [19:00:01] MaxSem: take a look at: http://zero-testing.wikipedia.wmflabs.org/xhprof_html/?run=4f70bc8f1cb81&source=mobile_frontend_extension&all=1&symbol=ExtMobileFrontend%3A%3ADOMParse [19:00:02] Apple ][C for me [19:00:10] tfinc: i've been using iTerm since i got OS X [19:00:15] *iTerm2 [19:00:55] YuviPanda is spoiled. he's growing up in the iterm2/git/github world [19:00:57] :D [19:01:15] indeed. my first open source 'patch' went into git [19:01:28] though I was using svn like git for a few years before that (local server :P) [19:01:36] MaxSem: and http://zero-testing.wikipedia.wmflabs.org/xhprof_html/?run=4f70bc8f1cb81&symbol=HtmlFormatter%3A%3AfilterContent&source=mobile_frontend_extension&all=1 [19:02:01] oh apple ][ [19:02:12] preilly, onwhich page was it recorded? [19:02:37] MaxSem: http://zero-testing.wikipedia.wmflabs.org/w/index.php/Quark [19:02:42] C64 for me. [19:02:53] MaxSem: click, "Profiler output" on the bottom of the page [19:02:58] awjr: ^^ [19:03:17] oo [19:03:24] awjr: you can now see profile information on any request to http://zero-testing.wikipedia.wmflabs.org/w/index.php/Main_Page [19:03:31] nice preilly [19:03:34] heatherw: i loved the old ][C :D [19:03:35] awjr: just click on, "Profiler output" on the bottom of the page [19:04:13] tfinc: <3 [19:04:39] I wrote my first code in the 80s. [19:04:58] MaxSem: I think http://zero-testing.wikipedia.wmflabs.org/xhprof_html/?run=4f70bc8f1cb81&symbol=DOMXPath%3A%3Aquery&source=mobile_frontend_extension&all=1 sucks! [19:04:59] <- 70s> [19:05:10] preilly, doesn't look like it's slow: this page is larger than average, I've seen up to 400ms on [[Barack Obama]] [19:05:10] Managed to miss CVS, at least. [19:05:15] this needs unicorns http://craig.is/making/rainbows [19:05:20] for all rainbows need unicorns [19:05:20] i was too busy not being born in the 80s to be able to write code [19:05:27] tfinc: kidbleach.com [19:05:38] it's not even nearly 500ms average reported on graphite [19:07:19] Amgine: I wasn't around back then. [19:07:26] MaxSem: we should import Barack_Obama to http://zero-testing.wikipedia.wmflabs.org/w/index.php/ [19:08:39] preilly, the high reported times are caused by mismatching profiler calls, see bottom of http://noc.wikimedia.org/cgi-bin/report.py?db=all&sort=real&limit=5000 "Profiling error: in(ExtMobileFrontend::DOMParse-filter), out(HtmlFormatter::filterContent)" [19:10:27] http://www.cpm.z80.de/source.html [19:11:10] MaxSem: I understand, but it's still slower than it was [19:11:49] preilly: can we see the same metric in graphite ? [19:12:58] preilly: do we have an m. for all of these http://meta.wikimedia.org/wiki/List_of_Wikipedias [19:12:59] ? [19:13:02] tfinc: vs, the "count cpu% cpu/c real% real/c" view in report.py? [19:13:51] tfinc: yes [19:14:23] hmm, can't connect to zero-testing anymore [19:14:34] MaxSem: why not? [19:14:55] preilly: just a perf graph that shows our decrease in performance post last push [19:15:04] MaxSem: do you mean by SSH "ssh -A mobile-feeds" [19:15:15] no, http [19:15:22] ssh works [19:16:20] jdlrobson: ® for EN ™for all others [19:17:10] arghh ok [19:17:16] are we gonna go for the image still for just english? [19:17:19] or text for all? [19:17:25] tfinc: https://graphite.wikimedia.org/dashboard/temporary-13 [19:17:42] preilly: can we have something better then temp-13 on these links ? [19:17:54] jdlrobson: do we have a good reason to use the image over just plain text ? [19:18:13] only sexiness as far as i'm aware :-) [19:18:31] tfinc, https://graphite.wikimedia.org/dashboard/MobileFrontend-DOMParse [19:18:39] Plaintext is sexier. [19:19:01] jeroid: +1 [19:19:22] tfinc: https://graphite.wikimedia.org/dashboard/PRETTY_STUFF [19:19:31] i must point out beauty is in the eye of the beholder jeroid and tfinc ;-) heatherw that ok with you? [19:19:43] awjr: phillippe just let me know that your otrs account is on the way [19:19:51] jdlrobson: lets mock it up and then we can make a judgement call [19:19:52] thanks for the udpate tfinc [19:19:59] sure doing so as we speak tfinc [19:20:42] * jeroid hugs MaxSem again for the plaintext APIs. [19:21:09] tfinc: just use the, "save as" option to name the share [19:21:24]