[00:37:23] I really wish I could make namespaces dynamically [09:17:19] good morning [09:32:16] morning [09:38:57] moin moin [10:48:37] I hope I didn't edit conflict you guillom [10:49:03] apergos, nope; I was afraid we might for a moment, but we didn't :) [10:49:09] yay*whew* [10:49:48] this is why we need etherpad like editing [10:49:56] then we can join the session in progress and see that [10:50:03] phooey. oh well... [10:50:09] :-D [10:51:57] dammit [10:51:59] I am getting so sick of this [10:52:56] ah [10:53:01] so I was saying: [10:53:26] with etherpad-style edit sessions, someone can join the session and see that someone else is working onit [10:53:32] and edit other pieces without a problem [10:54:55] Sorry for the ranting; this computer is relatively new and Linux support for it is limited. I'm anxiously waiting for the new version of openSUSE (due later today) since it comes with a newer version of the kernel that will hopefully prevent this laptop from randomly shutting down every time I ask him to work a little more than usual. [12:08:35] RoanKattouw: interesting thread at least [12:09:07] The chaining thread? [12:09:11] I was noticing that some of the XML stuff in 1.16 i was glimpsing over was chained in some spots and not in others [12:09:15] ya [12:09:21] seems like people cant make up their mind heh [12:09:35] Where do we use chaining in Xml code? [12:09:47] hi troubled [12:09:54] I don't think it has chaining in the strict sense [12:10:03] Although we do do concatenation and nesting a lot there [12:10:55] sumanah: heya \o [12:11:50] I guess it might be more nesting in the sense that the class is not being initialised in the same call [12:15:10] sumanah: interesting mailing list you got. things got a little crazy the other day I see:) [12:15:34] troubled: it is so wide-ranging that I honestly do not know what thread you mean :) [12:15:35] although I never have to worry about having nothing to read in my inbox anymore ;) [12:16:30] wikitech-l has gotten a bit weirder recently [12:17:20] was fine when I first joined (all quiet and slow), then some flame broke out and now its like 80 messages a week heh [12:17:52] mind you, 1.18 and all, so I can understand things picking up a bit. blenders lists are the same I guess in that sense [12:18:55] troubled: also, I figure, our community of developers is growing, and new people add more traffic [12:19:53] sumanah: makes sense. still in observe mode atm myself! [12:20:00] :-) [12:20:45] troubled: http://www.ohloh.net/p/mediawiki if you use the "committers" item on the code analysis dropdown menu, you see that last month we had more unique committers than ever before [12:20:49] my whole dev setup was in 1.16 though for the stuff I was hackin away on. 1.18 is gonna be fun :) actually, we were thinking of testing it soon [12:22:07] sumanah: I wonder what happens when you overlay bug count on that graph ;) [12:22:15] troubled: hold on [12:22:40] https://bugzilla.wikimedia.org/reports.cgi?product=MediaWiki&datasets=UNCONFIRMED&datasets=NEW&datasets=ASSIGNED&datasets=REOPENED&datasets=RESOLVED&datasets=VERIFIED&datasets=CLOSED&datasets=FIXED&datasets=INVALID&datasets=WONTFIX&datasets=LATER&datasets=DUPLICATE&datasets=WORKSFORME&datasets=REMIND [12:22:55] understandable not entirely fair comparison, but curious about possible corelation [12:23:06] gah! wall-o-text url heh [12:23:10] troubled: blergh, sorry [12:23:28] no worries :) [12:23:38] troubled: so, there are some metrics projects happening elsewhere in FLOSS like http://wiki.meego.com/Metrics/Dashboard and http://eaves.ca/2011/04/07/developing-community-management-metrics-and-tools-for-mozilla/ [12:24:22] troubled: I would like for us to get a unified statistics dashboard so I can answer questions like that, but no one has time to make one [12:24:48] sumanah: like your statistics I see :) [12:25:09] troubled: the basic steps would be: install Pentaho on a Wikimedia Labs instance, feed it our Bugzilla API and links to Subversion repo (later git repo) and mailing list stats and IRC stats etc. [12:27:31] sumanah: im guessing its all just csv/sql data? [12:28:08] speaking of git though (sry for 2 threads!), I seen you guys were migrating. Hope that goes well. I know other people that are interested in git as well [12:28:20] troubled: basically. http://opensourcebridge.org/2011/wiki/Online_Community_Metrics:_Tips_and_Techniques_for_Measuring_Participation#Contributed_notes has some behind-the-scenes on how the Meego solution works [12:29:17] troubled: thanks. ^demon|away is leading that effort. what are your acquaintances switching from? [12:30:41] sumanah: well, they use svn now. git has come up many times already, but seems to be more of a topic that people bring up for something to talk about than anything [12:31:01] *^demon|away grumbles about getting pinged at 7:30am :) [12:31:09] sumanah: heh neat. I did some irc stats years ago with similar setup. Sql->csv->openoffice chart [12:31:22] eek! another daemon! [12:31:22] troubled: well, irssistats now systematizes that. [12:31:32] yay. [12:31:46] what of the days of pisg! :) [12:31:56] re the git transition, troubled, https://lwn.net/Articles/409635/ is a case study of Postgres's transition, and when I asked Drupal people about their transition, they said it was similar [12:31:58] my setup was irssi actually too. such a nice client [12:32:34] sumanah: boy, you are just full of useful info! who needs google ;) [12:33:06] i have been sharing your git urls and such though. least nice to see how other people deal with weird setups (Svn externals etc) and stuff [12:33:56] troubled: yeah, we believe in sharing that kind of infrastructure info -- Ryan Lane has been especially disciplined about it [12:34:27] troubled: Yeah Sumana knows like everyone in open source :) [12:35:04] :) [12:35:29] <-- floss groupie :) [12:35:40] come on in, we have cake [12:35:44] woot! [12:36:25] speaking of FLOSS, its wednesday today too, woot! [12:36:48] if you actually believed that I knew lots of FLOSS people then instead of saying "oh our talks never get in to FOSDEM" you would have said "Sumana could you poke the right FOSDEM people and ask them to approve our talks?" [12:37:46] hehe [12:37:59] I guess I'm not in the bending-everything-to-the-positive spirit today (yet?) [12:38:01] knowing them is different than "can twist their arms to get them to do what we want"! [12:40:30] *sumanah looks at recent FOSDEM mailing list archives [12:40:47] the only name I recognize, and the only person who might recognize my name, is Peter Saint-Andre [12:41:45] troubled: remind me, how is the Sphinx search stuff going? [12:43:08] sumanah: atm, its just running as is until we all get more time to continue with changes [12:43:29] nod [12:44:35] sumanah: the problem is that the reference design works fine, but the live server keeps getting partial indexing and I cannot debug since I have no access to the server [12:44:53] Oh, right, you're the Sphinx guy [12:45:01] aye [12:45:02] *RoanKattouw had forgotten what troubled's pet project was for a bit there [12:45:35] id like to rewrite the whole addon. sphinx itself is great, but the addon feels like it needs some polish [12:45:40] <^demon|away> Cake? [12:45:44] <^demon|away> Now I'm hungry [12:46:06] Meh, I had cheesecake every day for the last three days [12:46:10] ^demon|away: I am happy to distribute virtual cake to those who push my pet projects forward [12:46:12] cake? [12:46:14] don't need more [12:46:21] RoanKattouw: for you, virtual anticake [12:46:25] haha [12:46:28] I'll take his share [12:46:46] troubled: what is special about Wednesday? [12:46:54] Per CERN, the anticake will decay faster [12:47:06] sumanah: FLOSS weekly today [12:47:18] troubled: link? I'm an LWN gal but am happy to read more [12:47:30] sumanah: twit.tv/floss [12:50:24] if she wants to generate anticake at no energy cost the best way is to generate an equal share of cake at the same time, which might as well go to me... just sayin' :-P [12:50:56] anticake heh [12:51:09] (where by no energy cost, I mean, other than running the high energy accelerator required to produce cakes and anticakes from collisions.. of... um... [12:51:19] what's the target material in this case? :-P) [12:51:49] heh [12:51:59] There is energy cost [12:52:12] But yeah if she's gonna produce anticake the traditional way, there will be cake too [12:53:42] sumanah: anyways, today at 12:30PM EST, it will be live is all i meant. should tune in, its good show. Randal Schwartz (perl guru) usually hosts, but linux outlaws Dan Lynch (last name?) hosts as well as others. so good OSS stuff [12:53:48] you'd like it [12:54:46] troubled: I tend not to watch video podcasts, and today I have to finish up some work and then prep to leave for the Mumbai hackathon, but I may give it a go [12:54:49] thanks for the link [12:54:55] it'snotyouit'sme [12:55:21] Save the video podcast and watch it on the plane? [12:55:32] It's not like you won't have time for it :D [12:55:55] sumanah: ah no worries. usually catch the audio in background myself :) [12:56:27] (also, few people realize that the Cheesecake Factory generates an anticake product stream as well, for sale in a separate area of the giftshop) [12:56:43] sumanah: Oh BTW, speaking of on-line videos, has anyone ever told you that you and Rachel Maddow have eerily similar speaking patterns? [12:56:52] WHAT [12:56:53] no [12:56:59] Seriously [12:57:04] First time I watched her show, a few weeks back [12:57:05] no one has told me this [12:57:10] I was like OMG she talks exactly like Sumana [12:57:16] *sumanah is reflexively embarrassed, does not know why [12:57:27] Except her accent (pronunciation of individual sounds) is closer to that of Kat Walsh [12:57:46] Wikipedia says she grew up in Castro Valley, maybe that explains part of it :) [12:59:08] I grew up in a bunch of places, spent too much time in Stockton, so yes, that might be a partial explanation. Here I was hoping that her charisma, intelligence, compassion, etc. were the linking factors [12:59:41] hahaha [13:00:23] I tend to tie language things to regions, but you're right those factors exist too [13:01:06] ("troubleshooting cakes": http://danni.dreamwidth.org/62812.html ) [13:01:40] no shoot me! [13:22:30] http://etherpad.wikimedia.org/1180rc1ReleaseNotes [13:23:35] guillom: release notes ^^ in case you want to help wordsmith [13:24:25] I rewrote half of [[MediaWiki 1.18]] already. [13:24:36] Is my help really needed for the release notes? [13:24:37] Reedy: so that's not really the release notes in that Etherpad, it's the summary. [13:24:43] Ya [13:25:42] /what goes on the email [13:26:36] Reedy: in that case, shouldn't this email include mention of the file metadata support improvement? [13:26:57] *sumanah looks at https://www.mediawiki.org/wiki/MediaWiki_1.18 -- is there reason for that & this to be so substantially different in what they focus on? [13:27:01] Probably [13:27:30] Feel free to bulk it out some [13:28:01] Most of whats on that page is missing from the email [13:28:12] guillom: on the release notes: "really needed"? probably not. for this summary, I bet you are the best person to improve it by copying in, appropriately, what you wrote for https://www.mediawiki.org/wiki/MediaWiki_1.18 [14:10:43] hexmode, are we using the "Target milestone" field now? [14:11:59] guillom: I believe yes https://bugzilla.wikimedia.org/query.cgi?query_format=advanced&resolution=---&target_milestone=1.18.0%20release&known_name=1.18%20tarball [14:12:36] But I guess it's ok not to use it for other (smaller) projects like the blog skin? [14:13:25] *sumanah does not know, defers to hexmode [14:31:23] bawolff: also here ;) [14:31:31] oh hi :) [14:32:03] On the images you just listed on the bug as thumbnails not updating [14:32:21] I think rotatebot might just be not rotating things by 270 degrees correctly [14:33:20] bawolff: not possible. Reason: I wrote parts of rotatebot :D [14:33:26] The one's with upload summary of "(Bot: Image rotated by 270?? (EXIF-Orientation set from 8 to 1, rotated 180??)" seem to have been rotated incorrectly [14:33:28] ;) [14:33:52] bawolff: compare the small thumb inside the rotate tempate and the 800px thumb [14:33:57] differing [14:34:24] bawolff: can you give me a the link to the "wrong" image? [14:34:40] https://upload.wikimedia.org/wikipedia/commons/d/dc/Manastirea_Neamtului_-_July_2008_-_7.jpg for example [14:34:58] ( https://commons.wikimedia.org/wiki/File:Manastirea_Neamtului_-_July_2008_-_7.jpg is file desc page) [14:35:20] alreay have guessed the URL ;) [14:36:57] compare https://upload.wikimedia.org/wikipedia/commons/thumb/d/dc/Manastirea_Neamtului_-_July_2008_-_7.jpg/159px-Manastirea_Neamtului_-_July_2008_-_7.jpg vs https://upload.wikimedia.org/wikipedia/commons/thumb/d/dc/Manastirea_Neamtului_-_July_2008_-_7.jpg/800px-Manastirea_Neamtului_-_July_2008_-_7.jpg [14:37:20] the 800px is old [14:37:37] They look identical to me (both look sideways) [14:37:42] wah? [14:37:50] sure - both are sideways [14:37:56] but 180?? different [14:38:04] aren't they for you? [14:38:29] nope, let me try pretending in europe just to double check though [14:38:45] gnah.. [14:38:52] *Saibo goes to dl via wget... [14:39:25] yep, if I pretend to be in europe i get the old thumb [14:39:46] just dled via wget: same [14:39:51] 180?? difference [14:40:13] I was served by upload.wikimedia.org|91.198.174.234|:443 [14:40:38] (yes, I am in Germany) [14:40:41] but that issue aside, shouldn't rotate bot be orienting it right sides up, not 180 degree difference [14:41:00] lol - the bot is just as clever as the operator... [14:41:10] user gave the wrong order [14:41:17] lol [14:41:32] explanation: he gave the order based on the physical image [14:41:41] but the bot assumes the order is relative to the thumbs [14:41:55] as stated in the rotate template [14:42:33] since the user hasn't used our rotate gadget to place the template he maybe didn't look at the thumbs [14:42:40] or didn't purge before tagging.. [14:42:44] hmm, thats odd. mangaling the thumb url still gives the old thumb [14:42:55] that is an old upload: so he could have seen an non-exif-rotated thumb [14:43:23] the rotate gadget purges the file AND uses a unusual thumb size to be sure to get a fresh thumb [14:43:40] but the user didn't use it ;) [14:43:42] Platonides: is https://bugzilla.wikimedia.org/show_bug.cgi?id=25619 reviewed, then? [14:43:58] bawolff: anyway: back to the old thumb bug.. [14:45:35] yeah, i don't know whats causing that. What's especially odd is its appearently serving wrong thumb even on a squid cache miss (of course I could be mis-interpreting something since i'm not intemently familar with wmf's caching set up [14:47:56] bawolff: just curious: who are you? ;) A volunteer MediaWiki hacker? A WMF employed hacker? Server admin? The first one, right? [14:48:10] yep, I'm just a volunteer [14:48:30] thanks :D [14:49:16] well, I think the server admins should be able to trace the bug.. if they would look at it ;) [14:49:31] maybe it is not a MW bug but a server config bug [14:49:35] hopefully [14:49:52] oh, I'm 99% sure its a server config bug of some sort [14:50:02] hehe [14:50:54] bawolff: but you figured out that it only occurs if you access from europe? [14:51:16] that would explain why the server admins cannot reproduce it and don't care about :D [14:52:03] Well I think there might be multiple issues with thumbs not updating, the other one that was reported several months ago that seems to mostly not be happening any more happened for everyone [14:59:07] yes, it was more severe in the past [14:59:24] but a week ago (or so) it didn't occur at all - to my knowledge [15:12:23] bawolff: thanks for putting it in the bug report [15:12:47] no problem. Hopefully someone will read the bug report who knows how to solve [15:12:52] ??? be back later [15:13:02] bawolff: we could spam the tech channel ;) [15:13:07] or ping individuals :D [15:13:54] or convince hexmode (ping) to annoy the appropriate people about that bug [15:14:01] yeay, that's good .D [15:14:02] :D [15:14:07] /away [15:51:48] Saibo, where is the bug report for that cache problem? [18:39:56] got my kindle fire! [18:39:58] there *is* a "Allow installation of applications from unknown sources" switch in prefs. this makes me happy [18:42:06] but i can't find a switch to enable usb debugging, which makes me sad [18:46:53] http://bug-attachment.wikimedia.org/attachment.cgi?id=9472 [19:50:42] brion: got addicted to "adb lolcat" on Android? [19:54:00] heh [21:32:12] anyone want to help review patches? [21:33:06] the fun, if anyone is looking for it ;) is here: http://etherpad.wikimedia.org/PatchReview-2011-11 [22:26:15] I've have an odd issue with ResourceLoader... [22:26:21] for some reason a RL module I defined isn't loading the jquery UI dependancies that are in the module definition... [22:26:32] lemme stick it in a pastie for you.... [22:27:08] I noticed you typed 'dependancies' with an 'a', can you double check that ? [22:27:27] sorry, it's spelled with an e in the code [22:27:29] (it should be an e) [22:27:30] OK [22:27:42] Platonides: hey! helping out? nice! [22:27:46] http://pastie.org/2874746 [22:28:06] the script and css parts load fine, but the 2 jquery UI dependancies aren't loading [22:28:29] Where is this happening ? [22:28:41] Any errors in the console? [22:28:43] Got a url or svn place ? [22:28:49] Also, look at the URLs Firefox is hitting in the Net panel [22:29:06] (or Network in the Web Inspector if you're using Chrome)_ [22:29:11] no errors [22:29:33] it's trying to load them judging by the Net panel.... [22:29:36] You should see jquery.ui in one of the URLs [22:30:23] http://experimental.localhost.wikimedia.org/load.php?debug=true&lang=en&modules=jquery!ui!dialog%7Cjquery!ui!draggable%7Cjquery!ui!mouse%7Cjquery!ui!position%7Cjquery!ui!resizable%7Cpfp!form!rapidhtml!lightbox&skin=vector&version=20111116T052406Z [22:30:45] 1.17? [22:30:51] yes, 1.17wmf [22:31:20] Huh where's button [22:31:32] hmm [22:31:33] It's not in the URL [22:31:43] Maybe loaded already [22:31:52] kaldari: mw.loader.getState('jquery.ui.button') [22:31:56] what does that say ? [22:32:18] mw.loader.getState is not a function [22:32:26] right [22:32:31] why 1.17wmf ? [22:32:36] fundraising [22:32:37] (just curious, it oughta work) [22:32:43] Is there any URL in the net panel that includes jquery!ui!button ? [22:32:50] (I'm assuming button is missing but dialog is working fine?) [22:32:59] no dialog doesn't work either [22:33:05] Huh, that's weird [22:33:14] lemme look for button... [22:33:17] Do you get 'dialog is not a function' ? [22:33:37] kaldari: Can you paste what that load.php url gave you ? [22:33:44] that should be fairly short (as it's debug with just urls) [22:34:01] yeah dialog is not a function [22:34:12] the contents of that request [22:34:37] sure, one sec... [22:35:40] ok, it's including button in the load URL now. I had removed it for debugging, but it still doesn't load... [22:35:41] http://experimental.localhost.wikimedia.org/load.php?debug=true&lang=en&modules=jquery!ui!button%7Cjquery!ui!core%7Cjquery!ui!dialog%7Cjquery!ui!draggable%7Cjquery!ui!mouse%7Cjquery!ui!position%7Cjquery!ui!resizable%7Cjquery!ui!widget%7Cpfp!form!rapidhtml!lightbox&skin=vector&version=20111116T221836Z [22:36:02] the first part of the contents is: [22:36:04] mediaWiki.loader.implement("jquery.ui.button", function( $, mw ) { [22:36:04] mediaWiki.loader.load("/resources/jquery.ui/jquery.ui.button.js"); [22:36:04] ; [22:36:32] kaldari: there's your bug [22:36:39] load doesn't take relative urls [22:36:39] ? [22:36:53] never did never will [22:37:20] so why is mediawiki.js generating relative URLs for it then? [22:37:22] it'll consider that a module name [22:37:26] not mediawiki.js [22:37:38] I can blame the backend for that :P [22:37:52] can I change my module definition to fix this? [22:37:53] seriously though, this was a regression at some point [22:38:00] not related to your extension, this is core [22:38:09] (assuming you don't have ui.button defined yourself) [22:38:16] are you up to date on 1.17wmf1 ? [22:38:22] I guess I could just use the full url for the jquiry.ui dependencies [22:38:26] ? [22:38:38] yes, up to date on 1.17wmf [22:39:08] just using standard jquery UI components, nothing custom defined [22:39:56] strangely I'm pretty sure this was working a couple weeks ago. Did anyone update 1.17wmf? [22:40:07] kaldari: turn off debug mode [22:40:43] that fixed it [22:40:48] thanks! [22:40:58] it's a debug mode bug, because it's loading a separate url [22:41:18] which is the point of debug mode, but that's broken and I remember fixing that (or at least reviewing the commit that fixed it) [22:41:26] I must have tested it earlier without debug mode on [22:41:39] in production mode everything is embedded by php as one long string so paths don't matter [22:42:14] anyway, thanks for the help! Hopefully what I'm working on will make us a few more million ;) [22:42:47] It's a new lightbox-style donation interface [22:43:18] RoanKattouw_away: You remember the bug I describe here ? [22:43:34] It was either you or me who fixed that. Looking in history now [22:44:21] There are multiple such bugs [22:44:26] But yeah that's what you get in 1.17w1 [22:46:49] RoanKattouw_away: could you look at http://bug-attachment.wikimedia.org/attachment.cgi?id=9380 ?? [22:46:59] it looks largely useless, to me [22:47:07] what am I missing? [22:47:31] hexmode: Link to bug? [22:47:41] This makes no sense out of context [22:47:45] https://bugzilla.wikimedia.org/show_bug.cgi?id=12837 [22:48:27] RoanKattouw_away: by the way, something completely different. Perhaps you have a clue. I'm on the Toolserver running a bot that I'm developing/maintaining as of early 2011 (previous maintainer left at the Countervandalism network). Unfortunately it's written in C#/Mono; I run it as "mono /path/to/foo.exe > /home/krinkle/bots/foo.log 2>&1", that works fine and the bots works perfect. Now I want it to run in such a way that I can safely close ssh, [22:48:27] add ' &' to the command, and it's not blocking my shell, but then for some reason the bot doesn't work. No errors, no output. I do get a PID, but it never shows up in 'ps' (or it stops before I can look). [22:49:01] Just run it in screen? [22:49:11] (Have you ever used screen before?) [22:49:37] I have not. Never had to. Are you suggesting ' &' runs it differently other than just being in the background ? [22:49:55] It'll still be a descendant of the ssh process [22:50:04] And it'll die if the ssh process dies [22:50:11] screens can be detached [22:50:25] huh ? I start stuff with ' &' all the time and they last with no problem [22:50:31] Even when you log out? [22:50:36] yep [22:50:48] Oh I might know. [22:51:01] I usually add the ' &' in a sh file (so I can call it with a shortcut instead of the full command). [22:51:25] I guess that makes it a subprocess of the sh command, which is really in the background ? [22:51:32] I don't know [22:51:37] I just know that stuff lasts in a screen [22:51:42] okay, cool. [22:52:11] Run screen, it'll drop you back into a shell after a welcome screen [22:52:19] Then run your command normally, in the foreground [22:52:32] It'll occupy your shell, of course, so you just create a new screen with Ctrl+A C [22:53:12] then "Ctl+a d" to "detach" from screen so you can log out [22:53:16] Yeah [22:53:33] You can switch back to the first screen with Ctrl+A 0 , they're numbered [22:53:35] and "screen -r" to re-attach next time you log in [22:53:41] so I just plain 'screen' to get into it ? [22:53:45] Yes [22:53:51] And screen -r to reattach later [22:54:00] k, didn't get a welcome screen but I'm getting a normal '$' now [22:54:12] Ctrl+A Ctrl+A goes to the previous screen, Ctrl+A " (double quote) gives you a list of screens [22:54:27] RoanKattouw_away: any change you had in test branch is now in production [22:54:34] Yay thanks [22:54:43] When I go to connect with ssh, am I in a screen as well by default ? [22:54:45] Did you merge the branches? [22:54:49] yep [22:54:51] Krinkle: No, you have to reattach with screen -r [22:55:02] was actually really easy this time [22:55:04] So I tend to put screen -DR in my .bash_profile [22:55:23] -DR? [22:55:32] -R = reattach, or create if no existing screen [22:55:32] [detached] [22:55:35] cool [22:55:47] -D = hard detach the screen I'm reattaching to [22:55:47] -D -RR Do whatever is needed to get a screen session. [22:55:53] ahhh. ok [22:55:54] Yeah :P [22:55:56] I do -d -r [22:56:04] I'm not sure how DRR differs from DR [22:56:10] -DR does the same, but will create one if it doesn't exist [22:56:20] Krinkle: I like my man page better: -D -RR Attach here and now. Whatever that means, just do it. [22:56:34] this one came from --hel [22:56:36] help [22:56:52] Oh, RR is to choose the first screen if multiple screens are available [22:56:55] can I name them instead of just numbers ? [22:57:10] yes [22:57:13] 21257.pts-43.willow (Detached) [22:57:18] that's not descriptive to me :D [22:57:23] c-a n [22:57:25] I always have only one screen [22:57:28] well, not that [22:57:32] And then use windows inside that screen [22:58:47] *hexmode is googleing to see if there is a reason to use class_exists AND require_once [22:59:03] Redefined class error? [22:59:07] Wait [22:59:09] _once [22:59:46] Autoload ? [23:00:04] this isn't in an autoload [23:00:39] require_once used to be much slower [23:00:49] but seems to be a non-issue now [23:01:00] *hexmode amends the patch [23:11:09] Krinkle, you may also like nohup [23:12:55] Platonides: yeah, counter vandalism network uses nohup for all mono bots [23:25:51] hmmm are there fundraising banners running on commons? i dont' see any when i go over there logged out [23:28:37] RoanKattouw_away: Do you know if the extension build for MoodBar to get efficient access to account creation date and account age is scalable right now to be used in CentralNotice / globally ? [23:29:48] *DarTar is interested too ;) [23:30:04] They wanna use it to serve some kind of notice for an external survey. [23:31:03] Doing this from JS is possible but that would I doubt it can be done from the API in an efficient way. Probably best to do it in PHP and server it though a variable somewhere. [23:31:20] DarTar: I noticed a mention earlier in the thread about edit count [23:31:29] more specially number of edits between two timestamps [23:31:59] is that a required filter ability ? The latter is very much impossible afaik (it's been done on the Toolserver, but take 1+ seconds to load) [23:32:16] agreed, let me bring up the latest specs I received from them [23:37:53] It's not generic, it's only for 1mo, 3mo, 6mo and lifetime [23:38:10] But yeah, it does scale, the data is all in a counting table [23:38:30] So you just SELECT one row and that's it [23:39:54] Krinkle: mail for you, IRC is fine [23:41:54] DarTar: yeah, I'd prefer not to conf call. If anyone else prefers not to use irc, I have problem adapting, but due to my limited time I'd like to multi task and be able to keep track of other things (every minute counts :) ). [23:42:04] I have no problem adapting* [23:42:13] np [23:43:23] DarTar: okay, so it's only on en.wp [23:43:33] yes [23:44:07] RoanKattouw_away: alrighty, so no need to deploy that extension globally. I was afraid that might not work out as-is, but we don't had to worry for now. [23:46:47] DarTar: Can I quote parts of 'User metrics for the eligibility check' in this channel ? [23:46:58] sure [23:47:10] there's nothing private about that [23:47:12] on en:wp, display a total number of edits of at least 300 prior to the launching date of [23:47:13] the study + at least 20 edits within the last 180 days prior to the launch of the study; or [23:47:20] /quote [23:48:16] lemme check out what data is available