[17:19:51] zakg|away, RoanKattouw_away are we using skype or mumble? [17:20:25] I suggest somke signals. [17:20:29] smoke, even [17:21:36] guillom, Happy Birthday! [17:21:43] Thanks =) [17:21:53] guillom, that might work for Roan and I.. Maybe not for Zak ;) [17:22:19] You just have to document it properly! [17:42:33] Right, meeting [17:42:45] That's not for another hour and 45 mins, right? [17:43:06] And happy birthday guillom [17:43:13] Thank you. [17:50:22] RoanKattouw, yeah, seems to be 19.30 [17:51:23] (my time ;)) [17:57:37] Yeah so 20:30 my time [18:08:40] hi RoanKattouw [18:10:44] Morning [18:10:55] *RoanKattouw is taking a school day today [18:11:16] i.e. won't be working much [18:11:22] no worries [18:11:24] It's the evening ;) [18:11:33] what work do you plan to do? [18:11:40] None planned today [18:11:45] We've got an API documenation discussion :P [18:11:48] Did CR for Kaldari earlier [18:11:51] Oh right and there's that [18:11:53] API docs call [18:12:04] Which I'd like to do on VOIP if possible [18:12:13] *RoanKattouw looks around for zakg [18:12:25] Hmph he's not on IRC [18:12:30] he timed out a bit ago [18:12:34] Ah [18:12:43] Well I guess he doesn't have a VOIP ext [18:12:51] So I'd prefer Skype, cause I actually have that installed [18:12:54] heh [18:12:59] As opposed to Mumble or what the heck ever [18:13:08] :D [18:13:13] apt-get install mumble [18:13:13] :P [18:13:17] Hmm maybe [18:13:19] *RoanKattouw is lazy [18:13:24] Remember, I'm not supposed to be working ;) [18:13:42] yeah, it's in the ubuntu repos :P [18:13:52] *^demon|away is doing the opposite today [18:14:01] ? [18:14:03] <^demon|away> I don't feel like doing school :) [18:14:06] hah\ [18:14:14] I would prefer working today [18:14:15] I don't think most people do ever [18:14:21] But I have to submit this thing by Friay [18:14:24] +d [18:22:31] I should probably eat before the talk with Zak.. [18:23:32] Yay, american TV has started again [18:28:16] I need to get in touch with this guy - http://jimhigson.blogspot.com/ [18:28:24] he's the author of Wikizzle [18:28:35] anyone good at internet stalking? [18:33:32] TrevorParscal, http://www.facebook.com/jim.higson [18:33:52] "Jim Higson" in facebook [18:33:54] Second result [18:34:15] I don't use facebook [18:34:19] and refuse to sign up [18:34:29] it's evil [18:37:51] http://www.ox.ac.uk/applications/contact_search/index.rm?lastname=Higson&initial=J&exact=false&site=staffgateway&find_email.x=33&find_email.y=15 [18:37:55] Higson, Jim [18:37:55] Kellogg College [18:37:55] jim.higson@kellogg.ox.ac.uk [18:38:30] TrevorParscal, done and done. [18:38:48] Reedy: you are a talented stalker [18:38:49] thank you [18:38:53] Ahah [18:38:56] That was easy :P [18:42:20] As opposed to the time you... ? [18:44:47] Facebook find you most people [19:02:08] reedy: hi! [19:02:15] hey alolita [19:05:57] *robla guesses from the php errors on http://prototype.wikimedia.org/flaggedrevs that ^demon|away is doing the 1.16wmf4 deploy there now [19:06:14] <^demon|away> robla: Me is trying to work on that ;-) [19:06:44] <^demon|away> s/Me is/I'm/ [19:07:17] me understood :) [19:10:27] [[Newspeak]] [19:15:23] RoanKattouw_away, zak just emailed [19:16:24] OK [19:16:26] Thanks [19:17:31] Explains why he quit and bounced ;) [19:22:49] <^demon|away> robla: Now with less failures :D [19:23:36] <^demon|away> This is 1.16wmf4 + flaggedrevs merged in [19:23:37] *robla cheers [19:24:34] <^demon|away> Ah, another core change I forgot. [19:24:48] ^demon|away: were there unanticipated merge issues, or just weird prototype specific stuff? [19:24:59] <^demon|away> Weird prototype specific stuff [19:25:10] <^demon|away> I had already done the merges. [19:26:41] protected function recentChangesBlockLine( $rcObj ) { [19:26:45] is full of errors.. [19:26:51] oh wait [19:26:53] extract weirdness [19:31:45] foreach ( $res as $row ) { [19:31:45] return true; [19:31:45] } [19:32:24] Bloody postgres [19:33:39] Hm, no Zak [19:34:19] was expecting 15... [19:34:27] 35 after what? the start? :P [19:36:32] 35 after the hour [19:40:21] <^demon|away> robla: I'm actually going afk for a bit. I started a list http://prototype.wikimedia.org/flaggedrevs/User:%5Edemon of things I found that broke [19:50:36] thanks ^demon|away ....been doing some testing and things seem to be in pretty good shape [19:51:25] pdhanda: I think I figured out the issue that I thought was a cross-browser issue. I believe the difference has nothing to do with browser, and more to do with logged-in/not logged in [19:51:54] oh [19:52:05] yeah logged in takes longer [19:52:57] any idea why? [19:56:50] pdhanda: seems like we it should theoretically be equivalent to diffonly=1 [19:57:04] robla: not sure, let me look at that now [19:59:56] and by now i mean after lunch :) [20:00:24] and by after lunch you mean tomorrow lunch? [20:01:20] heh [20:02:27] no prob...I need to get lunch too :) [20:03:21] "and by tomorrow I mean..." [20:34:06] errors on prototype again [20:34:13] welcome back ^demon|away :) [20:34:44] *robla really goes to grab lunch like he should have 30 min ago [20:50:34] oh weird...it's a specific diff that's causing it to blow up: http://prototype.wikimedia.org/flaggedrevs-w/index.php?title=Pakistan&diff=prev&oldid=7516 [20:50:58] er...http://prototype.wikimedia.org/flaggedrevs-w/index.php?title=Pakistan&diff=prev&oldid=3494 (which is "previous edit" from the previous link) [21:22:32] nimish_g: Would you like me to sync that i18n rev while I'm at it? [21:22:58] just did [21:23:04] RoanKattouw, thanks tho =) [21:23:14] oh, wait you mean the whole rev? [21:23:19] Ah there I see it [21:23:21] No, no [21:23:54] robla: did you just use "blow up" in reference to a page about Pakistan? [21:24:12] we really need to find a better test article [21:24:21] THE page about Pakistan, no less [21:25:34] hah [21:27:29] I see the actual problem with the page now, which is that for some reason, the diff viewer is letting us run off beginning of the rev history [21:29:47] guillom, flipzagging: hi [21:29:55] hi [21:30:02] are you feeling better alolita ? [21:30:19] guillom: heavy cold; voice outage [21:30:34] guillom: i thought i could beat it for the last 2 days [21:30:48] guillom: no such luck :-( [21:30:53] aww [21:31:07] Hot tea + honey =) [21:31:19] And chicken soup! [21:31:24] guillom: excellent suggestion, my friend :-) [21:32:07] guillom: so are you ok with the blog post/communications for uw launch [21:32:19] well, yes [21:32:24] I'm not working on that yet [21:32:31] but the schedule seems appropriate [21:32:36] guillom: should we go for the nov 30th post only [21:32:37] I'll coordinate with Jay [21:32:50] yes, I think so [21:33:00] guillom: should we also do an update on nov 24 on tech blog or wikitech [21:33:08] I'm not sure [21:33:25] I don't really see the point [21:34:10] well - our development community should be aware of what's in test instead of considering the sw available right? [21:34:54] flipzagging: are you around? [21:35:02] I think that someone, somewhere, will find out about it [21:35:27] alolita: sorry lost track of time, talking w/pdhanda [21:35:28] ok [21:35:29] yes, obviously, either from IRC logs or SAL [21:35:40] I'm ok with a post to wikitech-l [21:35:45] not so sure about a blog post [21:35:53] What about commons-l? [21:35:53] RoanKattouw: as they should :-) so it is always nicer to inform everyone [21:35:59] flipzagging: sure it's my fault [21:36:01] :P [21:36:10] RoanKattouw, commons-l for the launch, not the deployment [21:36:13] OK [21:36:16] guillom: i think wikitech-l works [21:36:20] otherwise there's no point at all in launching later =) [21:36:23] So wikitech-l for launch, everything else for deployment [21:36:26] hah ,true [21:36:30] Don't wanna attract the droves early [21:36:53] wikitech-l for heads-up about deployment, everything else for official launch [21:36:59] RoanKattouw: wikitech-l for nov 24 pre-launch [21:37:02] AH, sorry yeah [21:37:04] Got them backwards [21:37:09] :-) [21:37:14] we can even send the email to wikitech now [21:37:42] well it depends [21:37:43] yes as soon as RoanKattouw and flipzagging are ready and say ok [21:37:50] if you want to dampen the impact, send it now [21:38:07] i dont think anything should be sent now until nov 24 [21:38:08] if you actually want them to check it out when it gets pre-launched, better to send at the time [21:38:10] or nov 23 [21:38:19] flipzagging: agreed [21:38:27] i prefer the latter [21:38:30] I think we should send it on the day or the day before [21:38:31] might as well give it a pounding [21:38:36] Exactly [21:38:43] RoanKattouw: yes [21:38:45] As long as the users pounding it are aware of its beta status [21:38:47] well, I disagree, but I'm not going to fight for it [21:38:50] And Commons admins should be aware, I guess [21:38:56] oh yeah [21:38:59] hm [21:39:12] how do we reach them? [21:39:12] the main goal of giving a heads-up to wikitech is to have them help us make it ready, so the earlier the better [21:39:15] anyway [21:39:35] guillom: giving an heads-up is good when flipzagging and RoanKattouw are ready [21:39:58] Are we trying to solve an actual problem here? [21:40:14] just syncing up [21:40:20] I don't understand what we're talking about [21:40:28] me neither; so lets move on :) [21:40:34] Yeah [21:40:41] :-) [21:41:09] flipzagging: any blockers on your side [21:41:15] blockers, no [21:41:23] just a lot of stuff to do in the next 20 days [21:41:35] flipzagging: what about testing [21:41:55] ok, with Calcey, they said we could start with them tomorrow [21:42:04] there are a couple of issues with that [21:42:09] 1) need to write a better test plan for them. [21:42:15] Tomorrow is tonight, with their timezone [21:42:17] guillom: do we have user scenarios that we could provide to calcey for testing [21:42:28] Roan: it's unclear to me whether they were adjusting for my time zone [21:42:30] that we could reuse from requirements analysis [21:42:42] Oh OK [21:42:48] alolita, can you be more specific? [21:43:47] ..? [21:43:55] guillom: if we want to utilize calcey's help for testing, we need to provide them a test plan; so just wondering if you had defined use cases / scenarios for users using the upload wizard when doing requirements [21:44:16] we could reuse those in the test plan then for testing [21:44:19] "user has picture; user wants to upload picture; user manages to upload picture" [21:44:36] that's rather simplistic [21:44:44] ok, alolita, we don't really have very elaborate user scenarios, so let's move on [21:45:02] unfortunately, I'm the only person who probably knows all the twists and turns at this point [21:45:11] like, what's supposed to happen if you upload a corrupted file [21:45:14] Right [21:45:19] :-) [21:45:20] we have http://usability.wikimedia.org/wiki/Multimedia:Upload_wizard [21:45:29] Or a too large file, or a file with a blacklisted extension, or ... [21:45:31] so, it may fall to me to write the test case descriptions [21:45:36] RoanKattouw: exactly [21:45:46] You could just get Calcey started on generic smoke testing I guess [21:46:01] flipzagging: anyway there are a couple of other things [21:46:11] Talking to yourself now? :P [21:46:29] I am going to refer to myself in the 3rd person [21:46:31] anyway [21:46:40] :-) [21:46:45] 2) need to prep prototype for external testers [21:46:48] hey, do we remember something about getting design school students as interns or something? [21:46:51] a) templates? [21:46:54] so let's figure out what we need to do to outline tests [21:47:06] jorm: hey! [21:47:15] Something happened to the templates on commons.prototype, and Template:Information and such don't work, so they can't see if their upload info worked [21:47:23] or maybe it never worked [21:47:26] jorm, talk to Daniel about interns and you'll understand why it never happened. [21:47:31] and I just viewed wiki source [21:47:36] but it needs to work for Calcey [21:47:46] guillom: do you have any time to help with this process and keep the test plan in sync [21:47:48] is that going to be something depressing and infuriating? [21:48:03] flipzagging: can i help define some of these test cases with you [21:48:04] jorm, yes [21:48:10] alolita: sure [21:48:11] looking forward to it, then. [21:48:29] alolita, I can help if I'm told specifically what is needed from me [21:49:03] guillom: we need to try to test the main paths through the code, as well as define test cases more exhaustively for the less-used paths [21:49:15] guillom: so, "smoke" tests would be the standard thing like upload a jpeg [21:49:33] guillom: the others would be "try to upload a file that already exists" [21:49:35] guillom: so we need to capture these paths thru the code in a test plan [21:49:40] "try to use a title that exists" [21:49:48] "try to advance in the wizard without giving a description" [21:49:50] etc etc [21:50:04] guillom: so that we can get calcey to test thru those paths in various browser environments at the least [21:50:22] so we define a short list of tests that define main functionality and touch just about everything important, and a longer list of tests that are more exhaustive [21:50:30] and what alolita said [21:50:30] :) [21:50:40] guillom: usually this detail is defined during requirements analysis and then refined during design and then testing [21:51:11] and what flipzagging said well :-) [21:51:36] ok yeah well, we didn't quite do that [21:51:50] so should we do a test case sprint on friday? [21:52:05] we had a good picture from Guillaume about what was needed, but the more twisty stuff has been usually made up on the fly, perhaps after quick consultation [21:52:08] I'm taking Friday off. [21:52:13] oh [21:52:16] (I sent an email about this 2 days ago) [21:52:24] np [21:52:26] flipzagging, was going to send you an update today, sorry [21:52:42] guillom: do you have any availability to work on this next week? [21:52:48] yes [21:52:49] well, I'm not planning on taking the 11th off, no offense to our veterans [21:53:07] and Calcey wanted to start tomorrow... are we really saying we can't give them SOMETHING? [21:53:16] this isn't that hard, we can at least start now and refine later [21:53:22] flipzagging: that's what it looks like [21:53:27] we can certainly reuse http://usability.wikimedia.org/wiki/Multimedia:Upload_wizard [21:53:30] guillom: when next week? [21:53:35] and edit it to reflect the current state [21:53:47] whenever [21:54:11] guillom: when are you in next week? [21:54:29] I'll work all week, but Wednesday & Thursday from home [21:54:46] guillom: ok [21:55:09] let me try to do some of these tests thingies this afternoon [21:55:14] flipzagging: let's work on this - starting this friday [21:55:17] there's other stuff about the commons.prototype wiki (tech / templating stuff) [21:55:18] I can push what I was doing to next week [21:55:21] i can help [21:55:28] how about I prep a list of things to do and put it on the wiki [21:55:35] guillom can join in when he can [21:55:40] then Roan or Guillaume can pitch in when they feel like it [21:55:48] flipzagging: sounds like a good plan [21:55:57] right well, that's all I've got [21:56:05] except, oh, I updated prototype [21:56:13] flipzagging, are you doing that now? (the list of things you need) [21:56:14] so now it shows the Tutorial, internationalized even [21:56:20] flipzagging: great! thanks [21:56:20] so I can work on them this afternoon? [21:56:41] guillom: sure. It is typed into a text file on my computer. I will enlist experts to translate it to wiki form :) [21:56:48] ok [21:56:55] guillom: any thing to report on your stuff [21:57:12] flipzagging: You mean something other than the hella ugly and huge SPECIMEN one? [21:57:27] RoanKattouw: it's the real thing [21:57:34] guillom: are you happy with the translation work happening so far [21:57:34] RoanKattouw: loaded via instantCommons [21:57:40] no, today I was reviewing the translations & localizations, and trying to catch up on the small stuff I've put on backburner for a while. So I'll just put it back on backburner and work on this test stuff now. [21:57:53] the "no" was in response to "anything to report" [21:57:58] RoanKattouw: is there a way to change langs with a CGI arg? [21:58:06] flipzagging: The InstantCommons thing I knew, I reported a bug about the error message being strange when IC is disabled [21:58:10] flipzagging: ?uselang [21:58:30] Right, so http://commons.prototype.wikimedia.org/wiki/Special:UploadWizard?uselang=ar [21:58:33] guillom: :-) thanks for helping us with the test plan - it will help with deployment [21:58:46] which also shows us that the RTL doesn't yet work, but whatever [21:59:07] flipzagging, guillom, RoanKattouw: i think we're done - so thanks! [21:59:11] ok [21:59:19] flipzagging: I'll help you with the RTL support [21:59:20] just for today ;-) [21:59:20] thanks all, thanks for guest starring RoanKattouw [21:59:32] RoanKattouw: yes, I know how you do it, just need to run the Janus script [21:59:38] guest star RoanKattouw indeed :-) [21:59:41] Yeah, then sort out any remaining issues [21:59:46] Maybe add some @noflip's [21:59:53] @noflip? [21:59:55] ok [22:00:01] /* @noflip */ [22:00:17] On the line above a single setting or an entire rule [22:00:23] RoanKattouw: how about you talk me through it right now, if it's easy [22:00:25] Does what it says it does [22:00:31] I get it [22:00:41] so essentially, I assume, Janus does this [22:00:46] Yes [22:00:50] Janus obeys @noflip [22:01:00] create rules exactly like your css rules, except inside .rtl, and then flip direction on all of them. [22:01:06] except for @noflip. [22:01:15] No, that's not what it does [22:01:17] ok [22:01:24] You'd have a uw-ltr.css and a uw-rtl.css [22:01:42] Another thing it (optionally) does is exchange 'left' with 'right' and 'ltr' with 'rtl' in URLs [22:01:53] So you can have background-image: url(foo-ltr.png); [22:02:05] right, so that's just convention [22:02:25] Yes, and Janus will flip those if you set --flip-ltr-rtl-in-url or something like that [22:03:59] hm [22:04:23] let me read the docs and get back to you [22:04:39] btw, are people ok with this messaging? [22:04:40] http://commons.prototype.wikimedia.org/wiki/Special:UploadWizard?uselang=zh [22:04:58] this is what it says if it can't find a tutorial in your language [22:05:05] (imagine the red box is itself localized) [22:05:30] Looks good [22:05:31] Guillom? (see above) [22:06:04] no, it says I'm not logged in :) [22:06:05] RoanKattouw: you guys ported CSS janus to PHP, did you add any features? [22:06:08] gimme a sec [22:06:47] I added one bugfix [22:06:52] Well a few actually [22:07:03] Ported them to .py locally but never committed to MW [22:07:06] flipzagging, are you using the usual language fallback for each language? [22:07:25] i.e. if the Breton tutorial isn't available, you should fallback to French, not English, for example [22:07:27] flipzagging: You can use CSSJanus::transform() . There's a nice MW command line with php maintenance/eval.php [22:07:37] RoanKattouw: excellent [22:08:00] For docs, see includes/libs/CSSJanus.php [22:08:02] guillom: no, I didn't use standard language fallback in that case. Just exists or not [22:08:10] ok [22:08:24] guillom: I don't think there's a method for that, you have to hack it every time [22:08:34] really? hmm [22:08:51] well the fallbacks just chop off dialects, successively [22:08:52] anyway, flipzagging, I'd prefer a more subtle note after the tutorial, inviting to translate it, but we can go with this for now and improve it later [22:09:10] guillom: I totally agree, and thought hard about how to do that [22:09:27] guillom: but then, I thought, what is their purpose coming here -- to upload a file, or to help us fix the interface [22:09:40] right [22:09:42] guillom: so I wasn't sure how to phrase that [22:09:52] not sure they need a big red box saying we failed them, though :) [22:09:56] well [22:10:02] we could just show nothing at all [22:10:15] but then there's a big button that says next [22:10:19] for no reason [22:11:00] guillom: wow, I had no idea br should fallback to fr [22:11:00] my point is that if we show the English version, for example, they see we're showing the English version. No need to tell them :) [22:11:09] I'm guessing [22:11:13] guillom: fair point :) [22:11:40] well, give me an idea of how to handle it better [22:11:46] no apology at all? [22:12:06] flipzagging: Feature req: another next button above the image, scrolling down is a pain [22:12:07] or, if we have the apology, also say "hey, here's the page to translate it" [22:12:13] RoanKattouw: hella yes [22:12:24] RoanKattouw: also, the tiny next button looks really lost now [22:12:33] That's how I like my feature reqs to be responded to :D [22:13:01] oh btw the "Help Desk" is clickable, as requested [22:13:15] although, I spun for a while on how to make it more obvious that it was clickable, and failed [22:14:02] shading and stuff doesn't fit with the style of the tutorial [22:20:55] flipzagging: do you plan to make navpanel links blue so user can move back a few steps if he needs to? [22:21:14] vvv: well [22:21:46] vvv: it clearly is a good idea, but we'd have to change a lot. We don't have time before our first launch [22:22:02] I see [22:22:29] vvv: also, for a long time we weren't sure if moving backwards made any sense [22:22:41] vvv: since then I now see how to do it, but I just didn't impl it yet [22:23:17] vvv: basically we have to store the wizard state on the server rather than client, which isn't as hard as it sounds. [22:23:27] vvv: and then use popStates [22:23:41] Why would you need to? [22:23:41] flipzagging: it sounds pretty hard for me [22:23:52] If the links to go back are done client-side as well... [22:24:02] RoanKattouw: imagine, I upload an image, then before finalizing it, I go back to upload another [22:24:32] RoanKattouw: or, I upload an image, then close browser accidentally [22:24:41] RoanKattouw: before finalizing it [22:24:41] You'd want to resume your session [22:24:45] That's a general problem [22:24:51] RoanKattouw: yes, which we can fix now [22:25:30] RoanKattouw: so we can fix part of the going back issue with popstates, but the more general issue is better done with that and also obtaining state from server [22:25:43] whee! reboot time. [22:25:54] maybe I'm making it too complicated ;) [22:26:15] You're trying to kill two unrelated birds with one stone [22:33:31] flipzagging, can you tell me when your list of things you need from me is ready? [22:33:40] sorry [22:34:02]