[11:39:08] I always forget about this channel [11:39:27] anyone happen to know about the two comments in test.php in the normal extension? [11:39:37] "\xc0" => null, # Head byte at end is known to fail [11:39:37] and [11:39:43] print "Quick checks ok; try the more thorough test suite in includes/normal\n"; [11:39:47] (there is no includes/normal) [11:40:00] if not I'll ask brion who wrote that like 5 years ago now :-P [11:45:50] I'll even retract the first question and leave just the second one [11:45:59] s/question/comment/ [12:05:58] found 'em [12:06:03] he meant *that* includes :-/ [13:51:53] morning guillom - hey, I think I added the correct new template to http://www.mediawiki.org/wiki/Mobile_site_rewrite but tell me if I did it wrong [14:39:31] sumanah, hi; it's not the correct template, but you can blame me -- I haven't done a howto page for these yet. I'll fix it. [14:39:51] guillom: I demand to take at least 24% of the blame [14:39:59] Granted. [14:40:47] In unrelated news, the weather is so weird these days. [14:41:15] We have rain storms alternating with blue skies and sun. [14:42:31] <^demon|away> After 4 or 5 days of absolutely gorgeous weather, the heat has returned here :( [14:42:33] guillom: have you checked to see whether you might accidentally be living in a best-of compilation? [14:43:28] guillom: or maybe the weather gods marked their changes as minor so you didn't see them [14:44:46] I hate it when dressing up is guesswork :P [14:46:41] <^demon> Wear shorts and a t-shirt, but take a parka just in case? [14:47:04] Fortunately, this is where working from home comes in handy. [14:47:40] <^demon> +1 for sure! [14:48:00] <^demon> Who cares what the weather is outside? It's a nice ~72 inside all year round :) [14:48:04] :D [14:48:29] <^demon> And seeing as my commute is ~10ft from my bed to my desk ;-) [15:01:42] meh [15:01:48] *Nikerabbit votes for ISO units :) [15:06:57] I +1 that. [15:21:06] yuvipanda: ping [16:28:03] Anyone who is able to poke or investigate: The Wikimedia irc feed is down. Counter-vandalism patrols are complaining. [16:28:32] !admin !help !op !ugala-bugala !does !this !do !anything ? [16:28:32] --elephant-- I don't know anything about "admin". [16:28:50] Krinkle: it happens [16:28:56] you want #wikimedia-tech [16:28:59] I know [16:29:08] and there is no stalkword, if they don't answer they aren't there [16:29:12] But some people here aren't there who I believe can phisically poke some pps [16:48:04] I'm at home... [16:48:21] Me too :P [16:48:31] kaldari: Are you at home or in the office ? [16:49:02] in the office [16:49:23] Can you poke someone at ops ? irc.wikimedia.org has been down for 3 hours, nobody seems to know about it or be in IRC [16:49:32] countervandalism movement is complaining big time. [16:49:41] as it depends on that 200% [16:50:03] Can you poke someone at ops ? irc.wikimedia.org has been down for 3 hours, nobody seems to know about it or be in IRC [16:50:22] sure... [16:51:01] kaldari: hey, re https://secure.wikimedia.org/wikipedia/meta/wiki/GLAMcamp_NYC#Assignments_at_end_of_GLAMCamp Upload tool & Data Ingestion, what's up? [16:51:39] I'm looking for a metadata working group lead [16:51:59] multichill: if you & Ryan are working and I just haven't heard about it, sorry [16:52:46] and if you want me to help find a tester or report back to the GLAM folks that you need more time, let me know [16:53:07] *sumanah attempting to follow up on her post-GLAMCamp reporting responsibilities [16:54:12] Krinkle: the only ops person here is Peter and he's never heard of irc.wikimedia.org, but I have informed him of the problem non-the-less. [16:54:38] Krinkle: I imagine the other ops people will arrive shortly [16:54:50] kaldari: The hostname is "ekrem", perhaps he knows that ? [16:55:44] Krinkle: cool, I have conveyed the info to him [16:56:09] sumanah: Looks like you've done a great job so far [16:57:23] sumanah: I'm stuck in a week-long, take-no-prisoners fundraising meeting, so I'm not available for working on GLAM stuff at all this week [16:57:46] We're not even allowed cell phones [16:58:04] I'm only able to type this because I'm hiding under the table right now [16:58:39] tapping out messages in Morse Code [16:58:52] :) [16:58:58] multichill: ^ [16:59:04] meetings starting... cya later [16:59:07] bye [17:03:38] roan, tparscal: can we do an aft standup on irc today [17:03:42] timo: ? [17:03:51] krinkle: you around? [17:04:02] Yes [17:04:27] dartar: you around? [17:04:40] hey alolita [17:04:41] sumanah: What's up? Nothing much. Wiki Loves Monuments was on my mind last weekend and now I'm in Barcelona [17:04:43] let's do an irc sync-up on aft since everyone is remote [17:04:52] np [17:04:55] Could you please send me the chatlog? Do you still have it? [17:05:04] dartar: awesome [17:05:19] multichill: the chatlog from just now, talking to kaldari? [17:05:39] krinkle: are you ready to give an update now [17:05:48] 30 seconds earlier [17:05:52] ;) [17:05:52] Yes. [17:05:53] So, [17:05:54] No, the one from last week (you said you would send something) [17:06:03] krinkle: go for it [17:06:21] The preference has been implemented yesterday after the meeting. This is ready for prototype as far as I'm concerned [17:06:31] multichill: ok, just a moment [17:07:05] krinkle: awesome - can we push this prototype later today [17:07:08] I'm currently working on the dashboard, I think the front-end is ready as it is (the same structure can be used for this table as for the highs-lows). However it appears the script that arthur/howie made is experiencing some difficulties in my environment. [17:07:27] Oh hey I just got here [17:07:28] krinkle: what's the difficulty [17:07:34] It maybe related to my localhost, but there's some problems in the script as well [17:07:34] hey RoanKattouw [17:07:38] roan: we're having our aft sync up now [17:07:39] undefiend variables and php notices. [17:07:40] on irc [17:07:44] undefined* [17:07:54] krinkle: can roan help you [17:08:03] I guess so, yes. [17:08:19] Later, time for beer with Alex in Barcelona ;-) [17:08:43] krinkle: great ; please ping roan then - it should not take much time to resolve environment issues :) [17:09:08] I was gonna review all outstanding AFT revs today [17:09:14] Because I want to push trunk tomorrow [17:09:33] Roan: thanks - krinkle will also push to prototype today [17:09:35] for testing [17:09:56] krinkle: you have access to push to prototype right? [17:10:04] alolita: I dont believe I do, not yet. [17:10:22] I can arrange that [17:10:24] krinkle: then we'll have to invoke our magician - Roan [17:10:35] I mean, I can give Krinkle access [17:10:45] Roan: its time to got krinkle to push to prototype [17:10:57] Roan : its time :D [17:11:05] multichill: I cannot find it at all. this is maddening. [17:11:31] IIRC it was Platonides & kaldari talking and Platonides suggesting a design or plan [17:11:41] are we not meeting on phone? [17:11:54] krinkle: once you're synced up with Roan on the dashboard environment issues; [17:11:54] Krinkle: Send me your SSH public key and your preferred username [17:11:57] jorm: hi [17:11:58] jorm: IRC only [17:12:03] no irc only today [17:12:07] since everyone is remote [17:12:22] me included (not dartar though) [17:12:48] *DarTar is keeping an eye on your desks [17:12:48] that dario. needs to learn to take a day off. [17:12:53] agreed! [17:12:55] err, day off from his commute. [17:13:16] krinkle: so can you send us all a ping once you've pushed to prototype today [17:13:19] multichill: aha! search http://prototype.wikimedia.org/logs/%23wikimedia-dev/20110603.txt for your own name and just scroll back a bit. [17:13:24] alolita: yep [17:13:27] jorm: day off our commute indeed :D [17:13:52] dartar: thanks for keeping an eye on things in the office ;D [17:14:19] alolita: can I be next? I have another meeting in 5' on 6 [17:14:32] hi [17:14:35] yes go for it [17:14:38] ok, [17:15:29] re: dumps, everything is in order, I sent two small requests to Roan to make the dumps more readable and they should be applied by next week if RoanKattouw has time to do so [17:16:28] I also talked to the WSoR folks and it's still premature for them to make any decision regarding data hosting, so I suggest we go for a /datasets directory on dumps which will accommodate any data we produce for the time being [17:17:04] unless we want to create a dedicated subdomain (data. or something) [17:17:14] but I don't think we need that at this stage [17:17:35] I also note that there has been a new wave of comments at mw.org [17:17:51] ok [17:17:58] jorm has been responding to some of them, many don't make any sense [17:18:07] like "move the AFT to the talk page" [17:18:30] or "Please stop" [17:18:36] for others (especially those linked to a bug posted to bugzilla), we may want to comment politely [17:19:11] yeah, the "Please stop" keep coming [17:19:46] that's it on my front, give me a shout if you need help with testing the prefs [17:19:50] dartar: that's unfortunate ; we really need to reach out and work with the community in helping them understand [17:20:33] agreed, it's difficult for me to react on design-related issues as I was not part of those decisions [17:20:59] but I can definitely answer on why we are not moving to talk pages or go for an opt-in solution as some suggested [17:21:15] I have to bow out now and eat some delicious shoarma [17:21:22] Roan: enjoy! [17:21:25] someone also posted a proposal for a more compact design [17:21:27] I'll read the backlog and respond to any pings when I get back [17:21:30] RoanKattouw: bon app' [17:21:42] DarTar: let's work on this further tomorrow with howie [17:21:51] sure [17:21:59] krinkle, roan, dartar, jorm: thanks much! [17:22:05] I have to leave as well [17:22:05] for the updates [17:22:08] later! [17:22:15] *DarTar waves [18:01:38] neilk_: hey [18:01:42] i have some fixes [18:03:40] Hey TrevorParscal [18:03:48] see pm [18:52:08] hi mmerz [18:52:12] hi denis [18:52:22] hi alolita [18:52:28] mmerz, denis: hope all is well [18:52:29] hi denis [18:52:53] all is well! :-) [18:53:01] all is well [18:53:06] awesome [18:53:09] hi mmerz , alolita [18:53:23] so you saw Roan's note on code review status [18:54:00] yes, we saw his review, I think it should be fine [18:54:42] jeroen is working on the changes as we speek [18:54:52] I invited him here, btw [18:54:55] AFAIK, Roan is planning to sync up with Jeroen [18:54:58] cool [18:55:17] mmerz: is he around on this channel [18:55:48] do you know his username? [18:55:49] Roan also has mentioned that he would like a CSS review done [18:56:02] could you arrange this? [18:56:20] yes; i will ask another team member to help [18:56:23] JeroenDeDauw is in #mediawiki if you need him [18:56:38] Roan: hi! thanks for the pointer [18:56:49] hi RoanKattouw - thank you [18:56:50] roan: thank you for the info! [18:57:20] Roan: you are the one who Jeroen should sync up with for any code updates based on your comments [18:57:30] so do ping him as needed [18:57:41] Well he knows what to do [18:57:46] If he has questions he can come to me [18:57:53] Roan: yup; ok that works [18:58:20] roan: I have questions about the event logging you mentioned [18:58:24] hi Jeroen [18:58:30] thanks for joining in [18:58:34] hi [18:58:37] hi JeroenDeDauw [18:58:50] we need it for tracking displays only... could you implement this using the event logger? [18:58:56] hi jeroen! [18:59:18] questions for both Roan and Jeroen [18:59:22] mmerz: What do you need exactly? [18:59:32] Track display events with timestamps and such? [18:59:55] no, just track the total number of displays [18:59:56] I just took care of the .click and .cookie stuff, code at https://github.com/JeroenDeDauw/wmdesurvey/blob/master/index.html [19:00:23] Jeroen: great, thanks! [19:01:06] I think we took the .click and .cookie stuff from the original surveys code, if I am not wrong [19:01:42] Yeah, it's copied from http://meta.wikimedia.org/w/index.php?title=Special:NoticeTemplate/view&template=2011EditorSurvey [19:01:44] the not-.click stuff I meant ;) [19:02:13] Seems like pretty much every banner there is using onclick for the close button [19:02:16] great that you fixed it already, jeroen! [19:02:29] mmerz: what's the plan for testing [19:02:41] mmerz: are you using a test site [19:02:48] like prototype [19:03:50] a test site is very difficult for centralnotice [19:04:17] true but have you tested the survey and data that you will be receiving from it [19:04:19] alolita: we are not, but copying the css and js to user/vector.css and /vector.js should be sufficient to test I think [19:04:30] Yeah that'll work [19:04:38] It's actually better cause you get to do it on the real wiki [19:04:45] Jeroen: cool [19:05:02] the survey itself is being tested right now [19:05:06] This is what we've been doing so far, only the last changes I made have not been tested this way yet [19:05:18] I'(ll go ahead and do that now [19:06:10] what about event-logging? [19:06:29] we would need to track the total number of displays only [19:06:51] mmerz: great that the survey is being tested [19:07:27] what would be the best solution for this? the event-logging infrastructure that roan mentioned? [19:07:36] Yeah that'll work [19:07:46] If you just need the total number of displays, clicktracking will do it [19:07:56] cool [19:08:05] could you set this up? [19:08:14] All you need is mw.loader.using( 'jquery.clicktracking', function() { $.trackAction( 'youreventname' ); } ); [19:08:29] ah ok, this is easy [19:08:30] Jeroen : could you add the clicktracking calls in [19:08:40] Please use an event name prefix like wmdesurvey- so we can easily separate your events from others' using a prefix search [19:09:06] perfect, thank you! [19:09:33] so we need to test the last changes and do a css review, right? [19:10:28] and will there be a second review for the JavaScript? [19:10:53] I'll just review the diff [19:11:09] ok, cool [19:11:18] JeroenDeDauw: Once you're done tweaking, could you send me a link to the diff (in-wiki or in github, I don't care) between the version I reviewed and the latest version? [19:11:32] yup just reviewing the diff is all we should target since Roan's pretty busy already [19:11:53] sounds good [19:12:19] mmerz: we should do the javascript/css review only once jeroen is done [19:12:24] so probably next week [19:12:39] when are you planning to run the survey live? [19:13:17] if possible, we would like to use the original schedule [19:14:09] June 23-24 [19:14:21] totally at this point depends on jeroen's completion of all functionality discussed today; more testing to be done for integrating clicktracking and then final review by roan of the diff and a css / js review [19:14:52] we should be able to support that provided jeroen is ok with completing code and testing by end of next week [19:16:15] I will assist with testing [19:17:19] mmerz: so can we enable this on the 23rd since we don't do any deployments on fridays (24th) [19:17:57] and enabling the survey means we switch it on via central notice on deutsch wikipedia [19:18:06] sure! 23-24 only as it will run for 24h ;) [19:18:19] all the survey infrastructure is outside the cluster - right? [19:18:31] right [19:18:36] only for a day? how do you ensure enough responses then [19:18:42] DE and EN [19:18:45] i thought you were planning to run for a week [19:18:48] aah en too [19:19:10] I would prefer a longer timespan for scientific reasons, but it was a political thing [19:19:14] just asking why only a day's run since you may not get enough responses [19:19:24] the editor survey from wmf was run for a week [19:19:34] until 5000 responses were received [19:20:06] and you think thursday - friday activity will get enough responses [19:20:08] :-) [19:20:11] if we could run for a week, this would be absolutely preferred [19:20:29] how many responses are you planning for [19:20:38] what if we get only a 100 responses in a whole day [19:20:55] thats a small sample [19:21:09] best would be 3000 for DE... and 1000 for EN [19:21:10] ok well - just asking to make sure we can support you :-) [19:21:16] yes of course :-) [19:21:38] yes, lets run for a week! :-) [19:21:43] so if we don't get anywhere close to 3000 in a day's run - what do you want us to do? [19:22:14] since i would like to plan out our availability for enabling/disabling the survey [19:22:56] mmerz: then you want to run the survey from thursday 6/23 to 6/30? [19:23:04] lets run for a week and with a lower volume, would that work for you? [19:23:41] well - its your call - i am providing you the data to get a fair sample for so much effort that all of you have put into this survey :-) [19:23:41] RoanKattouw: I got 3 diffs (first commit first): [19:23:42] * https://github.com/JeroenDeDauw/wmdesurvey/commit/069c42f7356eec17aa076e05ef95668d2bfc7dbd#index.html [19:23:42] * https://github.com/JeroenDeDauw/wmdesurvey/commit/95ee20204715089d16be5e237944064e00336d49#index.html [19:23:42] * https://secure.wikimedia.org/wikipedia/en/w/index.php?title=User%3AJeroen_De_Dauw%2Fvector.js&action=historysubmit&diff=433259720&oldid=433258250 [19:24:27] if we run for a week, the start date would not matter so much (as long as it is in the same week) [19:24:36] mmerz: yes [19:25:06] I just tested these changes on EN Wikipedia, and seems to work as it should [19:25:09] great alolita, thank you! my choice is a week! [19:25:28] thank you!!! [19:25:31] :-) [19:25:39] mmerz: ok let's run for a week then :-) [19:26:04] thats awesome, really [19:26:05] thank you alolita [19:26:38] so as soon as Jeroen says he's done ; we'll do a final code review next week (roan doing only a diff review) and a css/js review [19:26:39] absolutely great!! :-) [19:26:40] *Denis_WMDE is fredastairing for a second [19:26:56] and then enable it on the 23rd through 30th [19:27:02] :D [19:27:06] :D [19:27:21] alolita: I am done now [19:27:38] wow ... [19:27:40] Jeroen : you've added the calls :D [19:27:42] awesome [19:27:51] so we can do this real time then [19:28:21] Roan: can you do a quick diff review whenever you have time in the next 5 days [19:28:49] Yeah [19:28:54] I'll do it today [19:29:00] and i will get a css/js reviewer in the meantime [19:29:12] Roan: awesome; thanks much!!! [19:29:20] .... [19:29:24] alolita: The JS is fine [19:29:29] CSS is the part I can't review [19:29:36] and we should be in good shape to enable the survey on the 23rd [19:30:09] great work! [19:31:03] If it's stuff I didn't write I could review it [19:31:08] (aft?) [19:31:19] Roan: i will get a css review then - awesome krinkle [19:31:33] Of AFT ? [19:31:34] krinkle - you could probably speedread thru the CSS :-) [19:31:47] this is the editor survey for deutsch and en wiki [19:32:31] ok we're set then for now ; mmerz, denis : let's sync up next week at the same time then [19:32:44] to make sure we're on track [19:33:10] perfect! thank you very much! [19:33:21] ok thanks everyone then ; ttyl ; bye [19:33:37] me = very happy! :-) [19:34:32] mmerz: have a great day; ttyl [19:34:44] thank you everybody! [19:34:54] alolita: have a great day too! [19:35:07] JeroenDeDauw: Did you trigger the clicktracking thingy too? Then I can check if it came throguh [19:35:19] (When testing on enwiki) [19:35:47] RoanKattouw: yeah, I did [19:36:14] I confirmed the request to the API was successful, but can't test further then that obviously [19:36:56] It seems to have come through just fine [19:37:00] Sending you the data in PM [19:38:10] JeroenDeDauw: BTW you can't have console.log() in production code [19:38:17] It breaks if there is no console object [19:38:38] if ( console && console.log ) { console.log( 'foo' ); } works but it's best to have no logging in production code at all [19:39:01] RoanKattouw: I know, that's why I'm calling some function, where you just comment it out for production usage [19:39:22] Data looks fine [19:40:05] Oh [19:40:16] OK; one of your diffs uncommented it [19:40:23] As long as it's commented out in production it's fine [19:40:30] Right :) [19:41:02] Oh and if you're using $ you should wrap your code in (function($) { ... })(jQuery); [19:41:48] hellou [19:42:35] RoanKattouw: well, the 2011 editor survey is not doing that either, so I'd argue it's not needed here [19:42:59] JeroenDeDauw: Also, instead of $('#foo').click(baz); $('#bar').click(baz); use $('#foo, #bar').click(baz); [19:43:06] Yeah it's probably not needed in production, that's true [19:43:38] Also, why do you need the random number in the &b= parameter? [19:43:43] RoanKattouw: cool, did not know that [19:44:01] Although I don't see how it applies to this code, there are no 2 click things with the same function. [19:44:22] Oh, gah, you're right [19:44:24] Subtle differences [19:44:24] RoanKattouw: the idea is to be able to see how many people shared the link with other people [19:44:38] OK [19:45:00] Well for two digits of randomness, that's only 90 possible values in practice [19:45:12] Heh [19:45:26] You're making the same assumption as I did on seeing that code :d [19:45:34] Manuel actually wrote it [19:45:41] It's the start, not the length [19:46:06] OH [19:46:08] Right [19:46:12] hehe [19:46:15] .substring(2) means from position 2 to the end of the string [19:46:46] I don't see why you'd want to remove the first two digits, but sure [19:46:56] Oooooh I do see [19:46:59] It's always "0." [19:47:01] The first two characters will be '0.' [19:47:03] Yeah [19:47:08] OK I understand it now [19:47:12] Not that it matters much, but why not [19:47:32] Could probably use a comment though :D apparently it was so magical you and I both had trouble understanding it [19:48:21] True enough [20:04:05] ^demon: RoanKattouw: do either of you know what's blocking 1.17b2? [20:04:22] robla: Let me point you to a revision, sec [20:04:45] https://secure.wikimedia.org/wikipedia/mediawiki/wiki/Special:Code/MediaWiki/89720 [20:04:55] <^demon> You and your secure urls... [20:05:20] That's Tim changing the release notes file to say 1.17.0rc1 (as opposed to beta1 or beta2), release date June 9th (tomorrow) [20:06:21] I will merge the final 3 revs in the queue now [20:07:00] cool, thanks! [20:08:06] <^demon> Yeah, last night I had the impression that there wasn't a holdup, just Tim reviewing those final merges. [20:08:20] <^demon> And berating everyone for not adding REL-NOTES and him having to do that rev ^ there [20:09:03] Actually one of them was already merged by ^demon recently but wasn't untagged [20:10:17] ArticleFeedback was just pushed to prototype. [20:10:32] (and core 1.17wmf1 as well) [20:11:04] I am about to start doing AFT review [20:11:35] Alright, 1.17 merge queue is empty [20:11:44] 1.17wmf1 merge queue is a different story but I'll deal with that later [20:11:46] "Wikipedia uses cookies to log in users. You have cookies disabled. Please enable them and try again." [20:11:49] eh.. ? [20:12:05] I can't log into prototype it seems [20:12:07] <^demon> You hit that bug too? I thought I was the only one! [20:12:08] Try checking remember me [20:12:14] Yeah prototype has that bug for some reason [20:12:17] <^demon> I've been hitting that bug for ~3 months on trunk locally. [20:12:21] But it's running cluster code [20:12:26] And the cluster doesn't have this problem [20:12:31] right [20:12:55] Note that if I uncheck Global login, and keep remember me check out as well it also works [20:12:59] either both or neither. [20:13:39] <^demon> Well I don't have CentralAuth locally, I haven't been able to log into my wiki for 3 months :( [20:14:11] thx to all being helpful - and bye [21:35:20] Oh my gosh, Arthur's dashboard code is pretty broken [21:35:30] The maintenance script is, at least [21:36:35] alolita: Tell Arthur his dashboard code was pretty broken, and tell him to read the CR comments (he should've gotten a notification e-mail). I'll fix the issues because he's in fundraising land now, but he should read the comments, he'll learn stuff [21:57:42] So, there's this: http://www.mediawiki.org/wiki/MoodBar_0.1/Design [21:58:18] questions, comments, concerns? [21:59:13] <^demon> Possibilities for other moods? [22:00:01] that's actually on my "research for the future" tip. [22:00:05] "The user clicks outside of the dialog area; or" [22:00:08] *RoanKattouw grumbles [22:00:18] It makes sense but having to implement that it's kind of annoying [22:00:27] Fortunately that's probably not gonna be my job [22:00:40] probably krinkle's. [22:00:50] Yeah [22:00:52] *Krinkle wakes up [22:01:03] <^demon> Wikipedia made me Hungry. [22:01:11] heh. [22:01:16] <^demon> Wikipedia made me Vengeful. [22:01:24] Basically you have to attach a click handler to the document that catches all clicks in the entire document, then grab the click's X/Y coordinates and check they don't coincide with the dialog. Or something. [22:01:30]