[14:48:27] RoanKattouw: i just did an svn up on my local wiki [14:48:37] and got all the resource loader code [14:48:51] Yes, RL is in trunk now, see wikitech-l [14:48:53] and it doesn't seem to "just work" [14:49:31] What's broken? [14:49:55] (I have to go right now, but bug reports against RL are welcome) [14:50:14] seems like main-ltr.css is not loading [14:50:20] and jquery is not loading [14:50:27] That's... strange [14:50:34] And didn't happen on our local wikis [14:50:38] Gotta go now but will investigate [15:00:54] RoanKattouw_away: running update.php seemed to fix it [15:40:33] Right [15:40:41] Yeah I guess it was serving you DB errors instead of CSS [15:58:01] RoanKattouw: it's still giving me problems in IE6 [15:58:08] lots of jquery undefined business [15:58:17] Hm [15:58:22] not sure if that's related to resource loader [15:58:26] Could be [15:58:31] but it was working before i updated [15:58:43] I'll have to create another VM, install IE6 and test with my local setup [15:59:06] ok...i was trying to get ArticleAssessment fixed in IE6 and now can't test it there anymore [15:59:41] also, do you have any idea how I could fix this error? [15:59:44] [Wed Sep 08 11:26:31 2010] [error] [client ::1] PHP Fatal error: Class 'ApiQueryArticleAssessment' not found in /Users/adam/Sites/wiki/prototype/includes/api/ApiQuery.php on line 312, referer: http://localhost/~adam/wiki/prototype/index.php/Automotive_design [16:00:06] There you go [16:00:09] Update AutoLoader.php [16:00:15] Or wait [16:00:35] waiting.... [16:00:38] Update ArticleAssessmentPilot.php to change 'ApiListArticleAssessment' to 'ApiQueryArticleAssessment' [16:01:09] Although it looks good to me, maybe you have an outdated version of that file [16:01:18] Oh, haha, there you go [16:01:29] The class name is wrong in ApiQueryArticleAssessment.php [16:01:36] yeah i fixed it [16:01:38] works now [16:07:10] RoanKattouw: do you know when reedy is usually online? [16:07:52] Well he's in LA this week [16:08:15] Supposed to be on vacation although he did commit stuff yesterday [16:08:38] oh...oops [16:08:40] Next week he'll be back in England (5 hours ahead of you) [16:08:46] i've been bombarding him with emails this morning [16:09:32] BTW The first error (of ~30) I'm seeing in IE6 is 'Object doesn't support this action' [16:09:47] I'll debug it late [16:09:49] r [16:09:54] k [16:10:00] i'll worry about it later [16:10:04] Are you working on addressing my review of ArticleAssessment.js or should I? [16:10:17] i am [16:10:29] but i've mostly just been trying to get it working again after updating this morning [16:10:40] i saw you already incorporated some of it [16:11:38] Yeah, the really really small things [16:12:01] Basically I deferred anything requiring more than 5 minutes of work or touching multiple files [17:17:42] RoanKattouw: when you say 'use defensive programming', are you talking about checking variables exist before i attempt to access them? [17:18:15] Yeah [20:40:57] why leave page dialog now triggers even when the page hasn't been edited at all? [22:08:55] adam_miller, list fixed [22:08:59] 72612 [22:09:07] ooooooohhhhhhhhh [22:09:29] awesome [22:09:36] Reedy: MVP. [22:09:44] MVP? [22:09:49] google it [22:10:04] Most valuable...? [22:10:09] http://en.wikipedia.org/wiki/Most_Valuable_Player [22:11:23] sorry i'll try to be more respectful of our cultural differences next time i compliment you [22:11:43] Reedy: MofM [22:11:50] er [22:11:52] MotM [22:11:56] does that work better? [22:12:18] k, so the survey extension. [22:12:19] Man of the Match? [22:12:37] yeah man [22:12:49] heh [22:13:01] heh. [22:13:05] heh [22:13:15] reedy: heh [22:13:23] adam_miller: hey [22:13:37] so, on the survey. do we have to repurpose a new extension, or can we use the current one? [22:13:37] Hi [22:13:43] alolita1: hey [22:13:51] actually, got both of you here. gang's all here! [22:14:02] look at that [22:14:09] cool! [22:14:17] i'll send an email later with these questions so we have them all. [22:14:20] first, a technical one: [22:14:29] I've not really had chance to poke... SimpleSurvey seems to do some kooky stuff with the usability initiative survey one [22:14:37] what is the upper limit of article traffic that we can handle before we start to fail? [22:14:48] like, how many hits or edits a day, etc. [22:14:58] i'm going to cc roan on it. [22:15:05] Roan was going to have a look, not sure how far he got.. Not touched base with him since he got home [22:15:36] I'm not sure how we go about working out traffic limits [22:15:55] k. [22:16:15] In theory, if we do our best to use optimised/indexed queries [22:16:35] there should be minimal overhead [22:17:22] second, let's discuss chopping the "This article has been revised 24 times" line and replacing with "This article has been revised more than X times". [22:17:33] I am totally happy with the latter. [22:17:49] Mhmm [22:17:55] I don't think it makes too much different [22:17:57] *difference [22:18:03] but it should allow for us to replace X at whim. [22:18:06] Maybe something to poke at in future [22:18:20] I think Roans point is, depending how the query works, depends on how many stuff it scans [22:18:45] If we want to just know it's > x, there is a way (i'm guessing), to make it stop when x > y [22:18:55] I'll add a thing for it anyway. [22:19:03] which is cheaper than finding out exactly how many things are > x [22:20:22] right. [22:20:23] I might go do some reading in a bit to confirm that [22:20:30] Or just poke domas ;) [22:21:40] I think the "be nice" rule comes in, unless we have a specific need for the exact value, we should just do per roan [22:22:03] yeah. [22:22:04] Ofc, this could vary based on some more advanced "stale" rules if howief comes up with some later on [22:22:11] so that's checked in now, a message for it. [22:22:22] r72614 [22:24:01] what's the category that we're using for this? [22:24:20] Do we still want "articleassessment-stalemessage-norevisioncoun" [22:24:22] Do we still want "articleassessment-stalemessage-norevisioncount" [22:24:31] On prototype, it's just Test [22:24:37] It's configurable via a wg [22:24:45] okay, so we need to get a category from them. [22:24:48] excellent! [22:25:06] yes, let's keep that anyways, so that it gets translated at least. [22:25:16] Might be worth just updating the comment/fixme [22:25:33] if we have to drop to full ambiguity - or variable ambiguity - wrt what it means to be "stale", it's helpful to have an ambiguous message. [22:26:14] like if we decide that stale = 5+ revisions *OR* +/` 20% byte change etc. [22:26:50] actually. [22:27:06] we should just use the "norevisioncount" message. let's just go with the ambiguous message over all. [22:27:40] i'll come up with better wording. [22:29:47] yay, ambiguity [22:30:06] i can just see us changing the definition earlier rather than later. [22:30:21] okay. [22:30:29] Mhmmm [22:30:50] so, the prototype. is that automatically refreshing with code, or do we have to push it? [22:31:05] push it [22:31:12] though, the gui is slightly broken atm [22:31:14] want me to update? [22:31:50] We could do with checking with siebrand about getting it onto translate wiki sometime soon [22:31:56] if we've got something worth testing, sure. [22:32:01] when the messages are "stable" [22:32:02] *Reedy grins [22:34:07] workin' on it, dude. [22:34:29] at some point, my life seems to have turned into one long meeting. [22:41:21] k. mail sent. [22:53:09] right