[14:44:13] RoanKattouw: are you going to fosdem? [14:44:45] Haven't thought about that yet [14:45:08] I think I will be there, and probably go to nl for a bit afterwards [14:45:26] uni doesn't go back until the 28th of feb, and fosdem is 5+6 [14:45:35] so I think I'll cram in a few weeks of europe before I go back :) [14:45:44] :) [14:45:53] That's like your summer break, right? [14:45:57] yeah [14:46:10] so naturally I'm running away to the northern hemisphere to freeze for a while [14:46:46] That's what I'd do if I lived in Australia [14:46:58] in other news, how does one tell a ParserOutput object to call $wgOut->addJQuery()? [14:47:23] this is what I have now, but it doesn't play nice with UsabilityInitiative because the JQueries conflict. [14:47:26] $output->addHeadItem( biblioLinkedScript( "$dir/jquery.js" ) ); [14:47:51] Meh, you don't that's a flaw [14:48:05] damn [14:48:10] OTOH trunk now includes jQuery unconditionally through RL [14:48:18] And the cluster has a hack unconditionally including jQuery [14:48:32] So basically the answer is "just assume it's there" [14:48:37] so the installation I'm working on has UsabilityInitiative/Vector CollapsibleNav on [14:48:45] so presumably I can just comment it out [14:48:46] but ugh [14:49:42] Both trunk and the cluster include jQuery on every single page imaginable [14:50:02] So don't worry about ensuring jQuery's presence, it'll always be there [14:50:35] nod [14:50:38] still, awkward. [17:15:54] adam_miller: we're in 2002 if you can make it. [17:16:11] jorm: lies [17:16:30] i'm in 2002 and it's just me and the hold music [17:24:07] Hm I'd call in if my sound wasn't borked again [17:24:19] I'd have to restart Firefox to do a few minutes-long call [17:37:13] *RoanKattouw teaches jorm about copy-paste, which is essential when doing bulk things in Bugzilla [17:38:54] what did i do now? [17:40:13] You used all kinds of varied summaries for bulk-assigning LQT design bugs [17:40:34] Most people open them in adjacent tabs and do a Ctrl+V Submit spree [17:42:46] i'm not most people! [17:48:32] roan, those questions i just sent you may be being revisited this afternoon. [17:48:48] OK [17:51:44] OK [17:52:00] So what could change? Just rephrasing? [17:52:22] Is the branching structure (i.e. Q1 has an ifno branch, Q2 has an ifyes, Q3 has no branches) likely to change? [17:52:50] *TrevorParscal drops in on conversation that sounds exciting [17:53:47] there may be a branching question that has a series of checkboxes. [17:53:55] OK [17:54:05] That's easily doable [18:51:54] RoanKattouw: how's it going? [18:52:05] I'm good [18:52:25] Needed to take some rest last weekend and process my pictures [18:53:01] Work-wise, I'm now reviewing ArticleAssessment code [18:53:02] right on [18:53:10] The PP folks gave us another week, which we'll need [18:53:10] we have some bugs I need your help with [18:53:17] I noticed [18:53:23] I'm finally current on wikibugs-l [18:53:32] I've been writing a lot of code, but possibliy making some new issues in my wake [18:53:32] And I have a few dozen bugzilla tabs open [18:53:36] :) [18:53:49] I noticed stuff like &user=83.119.nnn.nnn [18:58:09] If there's any bugs in particular you need me to do, assign them to me if I haven't done so already [18:58:18] Otherwise ask questions on BZ or here :) [19:08:23] TrevorParscal: Oh also: we should do something with {{PLURAL in JS messages. Is Neil on this? I guess we should at least have a bug for it [19:08:46] RoanKattouw: I am not "on" it in any meaningful sense ATM [19:08:51] but maybe next week [19:09:12] in any case I would just be integrating mdale's JS2 library maybe with optimizations as we discussed earlier [19:09:17] Right [19:09:25] I meant "on it" in a loose sense [19:09:27] have the article assessment messages gone to translatewiki? [19:09:28] As in supposed to fix this [19:09:38] jorm: Looks like, I'm seeing a zillion translations [19:09:39] RoanKattouw: I have to do it for me, so I might as well do it for us ;) [19:09:45] okay. [19:09:53] I say a rev by Raymond adding it to the TWN file [19:10:04] flipzagging: Fair :) [19:10:08] yeah. i can fix this. [19:10:08] I have to do it for me too :) .. although i don't have too many PLURAL references in mwEmbed stuff so far [19:10:15] I have lots and lots [19:10:18] I trusted mdale [19:10:22] ;) [19:10:36] well should be strait forward to add in [19:10:39] yup [19:11:05] okay. i've got a fix on it. [19:11:16] $1 (Total $2 ratings) [19:11:18] and the coverage is pretty good.. might need some minor updates since the last time i touched it: [19:11:26] jorm: If you can somehow pull off a non-awkward way to avoid plural there, that'd be nice. Otherwise, we may or may not be able to have this not be buggy for n != 1 or userlang != en (rare case) depending on how fast flipzagging is [19:11:34] http://prototype.wikimedia.org/s-9/extensions/JS2Support/tests/testLang.html [19:11:39] jorm: That still isn't right for $2 == 1 [19:12:22] we're just going to have to look stupid for the 1 or 2 seconds that there aren't ratings. [19:12:33] All right [19:12:43] flipzagging: if you port it in.. would be good to add that page integrated into the test suite [19:12:45] I need to move the database files on prototype to a new partition. I need to turn the mysql database off to do so [19:12:46] Then I'll just remove PLURAL from the message saying it's a temporary thing [19:12:55] If your platform does not have plural substitution, it is always best to use the plural form [19:12:55] Until Neil integrates Michael's JS PLURAL code [19:12:59] any good time to do this? [19:13:13] flipzagging: Except there is no "the" plural form in most languages [19:13:19] Actually, you know what, I'm not gonna touch anything [19:13:22] e.g. Users don't actually freak out if they see "1 file(s) found" [19:13:29] or even "1 files found" [19:13:35] The way it's done right now {{PLURAL}} expands for the 0 case anyway because we're doing it wrong [19:13:44] really? [19:13:44] flipzagging: they freak out. believe me [19:13:56] I'm just gonna put a little note in qqq saying PLURAL does not currently work [19:13:58] really. [19:14:07] k. [19:14:12] flipzagging: Languages such as Polish inflect plurals the way English inflects 1st, 2nd, 3rd, 4th, 5th, ... [19:14:15] flipzagging: let me know if you need any help with that since there are some bits that are no longer relevant in the new resource mode. [19:14:16] Raymond_: I have charts and graphs, or at least I did, in a former job. Our issue, as usual, is that the WM community are perfectionists [19:14:23] mdale: ok [19:15:04] The other way to avoid using plural is to do stuff like this [19:15:10] rather than say "1 files found" [19:15:13] you say "Files found: 1" [19:15:16] jorm you are bharris? [19:15:21] also the most recent version is in /mwEmbedStandAlone not js2 [19:15:22] i am. [19:15:24] few users complain about that [19:15:27] fine :) [19:15:29] jorm: I was just saying I can keep the message as it is (reverting you) and just adding a note that PLURAL doesn't work yet [19:15:37] flipzagging: yea people complain doing that way [19:15:49] people complain about EVERYTHING [19:15:52] ( which is why it was written originally ) [19:15:54] this is about minimizing damage [19:15:58] if you don't have plural [19:16:12] Yes, people complain, that's why MediaWiki's i18n system is so feature-rich [19:16:22] fine? of course i'm fine. my girlfriend tells me that all the time. [19:17:21] jorm: So does that sound OK to you? [19:17:38] One of the issues I understand niklas has been studying is plural .... there are loads of problems with it [19:17:47] Oh yeah [19:17:55] even the standard ie CLDR has its issues [19:18:05] oh. go ahead. [19:18:12] The other reason MW's i18n system is so versatile is that it was primarily developed by a linguistics student :) [19:18:23] yea there is leakage / difficult to cover everything [19:18:30] I learned this from Siebrand, I want to learn more from Niklas and then I will blog about it [19:19:04] mdale plural is one of the linguistic things that can be expressed mathematically [19:19:36] so it is something that can be solved [19:19:54] the problem we're hitting is that it's all done in javascript, afaik. [19:19:57] or needs to be. [19:20:25] jorm it just means maintaining two transform files [19:20:35] generality the transforms are simple this is not a big deal . [19:21:02] and if we have a page that tracks regressions ie something like: http://prototype.wikimedia.org/s-9/extensions/JS2Support/tests/testLang.html its pretty easy to fix [19:21:07] if things break [19:21:26] ( should be integrated into the normal code review selenium testing ) [19:21:57] Preferably this wouldn't be done with Selenium, but with a unit test framework [19:22:01] Because it's really unit testing [19:22:11] do we have a javascript unit-test thing? [19:22:25] ( have not been tracking this stuff very closely .. been ~busy ~ [19:22:36] Not yet, no [19:22:42] But flipzagging has been looking at Jasmine [19:22:56] oh.. oky well yea if we get a javascript unit test thing in place then yes it would be more ideal to use that [19:23:02] mdale: I highly recommend jasmine [19:23:04] ... [19:23:11] that said I haven't checked in my test suites yet [19:23:55] adam_miller: Are you working on ArticleAssessment.js right now? If not, I'd like to change those [[#|Show]] links to something better [19:23:58] hmm but that runs in javascript in the browser? [19:24:07] yeah i am [19:24:13] i can commit my changes in a minute though [19:24:22] mdale: jasmine can run in the browser or 'headless' [19:24:31] mdale: if you have a spidermonkey or other such thing [19:25:06] cool.. so just need to integrate the headless bit ;) [19:26:02] adam_miller: OK I'll hold back till you commit then [19:26:11] flipzagging: you looked at http://docs.jquery.com/Qunit as well no? [19:26:24] It's not like I don't have 1948 mediawiki-cvs e-mails to wade through [19:26:42] mdale: no [19:26:58] being the 'official' jquery unit testing framework might guarantee a bit more of the projects longevity / momentum / support tools etc. [19:27:00] Syntax doesn't look as nice as Jasmine's though [19:27:11] mdale: jasmine is really a zillion times better than anything based on *Unit [19:27:31] It is, really. flipzagging , maybe you can pastebin a sample testcase in Jasmine [19:27:33] mdale: the syntax is not only nicer, it's more appropriate to JS. [19:27:39] sure [19:27:48] oky if its a lot better thats fine.. just has to be a lot better to not go with whats the most popular ;) [19:28:07] / not go with what our framework is using [19:28:32] If Jasmine development stalls it's not like our tests will suddenly break [19:28:40] (Provided Jasmine is indeed open source) [19:28:54] I agree that it's not nice to have your test suite be abandoned on you but it's not the end of the world [19:28:59] http://pastebin.com/STwidLp5 [19:29:00] So, again. Anyone have any input on what would be an appropriate time to stop the mysql server on prototype? [19:29:06] RoanKattouw is more about saving integration time .. [19:29:07] Jasmine is open source, by Pivotal labs [19:29:11] I need to move the database files. [19:29:19] Jasmine is still pretty new though, so that's a reason not to use it [19:29:29] also, most of the people in the Jasmine community are Railsy [19:29:50] (If I don't get an answer, I'm just going to do it. no blaming me ;) ) [19:29:53] that said, it's just about JS testing, so we don't care about the .rb infrastructure. [19:30:26] anyway the awesomeness of jasmine is that the test spec I posted turns into a very pretty HTML page, OR, a series of tests you can run in the browser. [19:30:34] er, OUTSIDE the browser. [19:30:42] flipzagging: yea Its fine to use it.. just have to weigh in alternatives... should at least look at Qunit [19:30:49] ok [19:33:01] RoanKattouw: committed [19:33:24] i'm going to make another change to the js to go along with that stuff, but i dont think you'll be in there right? [19:33:42] Oh actually I'm not gonna do this today after all [19:33:54] It's 9:30pm already and I need to get up early [19:34:01] So I'll do it while you sleep [19:34:05] ok [21:38:51] what does the usability team call their edit interface? [22:10:34] nothing fills me with more confidence than "we don't know what caused the leak."