[11:44:03] hey yuvipanda - been quiet today... everyone in transit? [11:44:10] jdlrobson: yes [11:44:13] i just reached Pune [11:44:22] cool! [11:44:54] :) [11:51:45] jdlrobson, it's usually quiet here this time [11:51:59] fun begins when SF people wake up [11:52:09] yeh I've noticed [11:52:16] need to somehow rearrange my social life around that :) [11:52:57] I get up around 11-12 and sometimes work till 4am [11:53:32] MaxSem: sf people are all in-transit people [11:53:39] yup [11:53:44] I get up at 3PM and work until 5AM [11:53:59] usually - am on a more 'normal' schedule now for the hackathon [11:54:02] but we have philinje at least [11:54:09] ah yes [11:54:29] jdlrobson: works out great for me - I work for an hour or so (until 5PM), then head out until 11PM, and then head back in [11:55:15] Yeh I might have to try that. Problem is I tend to wake up early out of habit.. [11:55:33] ah yes [11:55:43] * MaxSem had no problem with switching from getting up at 5am [11:55:47] and I think timezone for you won't be as night-oriented as for me [11:56:32] you're just across the pond unlike us [11:56:37] yeah [11:56:39] Plenty of time to work these things out :) [11:56:45] true that [12:13:36] jdlrobson, so what is your current task? [12:13:49] im looking through the existing mobile site code [12:18:16] do you already have commit access? [12:18:41] not yet i'm in process of firing off an e-mail actually [12:18:54] at moment I've just been working of a git repository as I need to hook up with preilly first anyhow [12:19:58] yeah, I'm actually also groking at MF now - I'll need to add a real API to it [12:20:14] started with refactoring yesterday [12:22:04] the bit that's bugging me the most is actually the css stylesheets - many of them share the exact same definitions making it hard to debug [12:23:42] hmm #blackberry { test: 1; } [12:24:18] :) [12:25:08] there's a lot of javascript that css should be taking care of which is why the headaches [12:25:54] I need to get a blackberry simulator at some point which also means sorting out a virtual machine :S [16:02:45] hi yuvipanda! [16:15:46] * aude waves [16:16:07] Astagi: heya [16:16:09] aude: heya! [16:16:34] hi aude! [16:16:40] all ok guys? [16:16:50] how's the hackathon? when does it start? [16:17:13] aude: a day from now [16:17:14] 10th [16:17:17] today is only 8th [16:17:17] cool [16:17:21] :) [16:17:28] * aude jealous [16:17:35] :D [16:17:42] yuvipanda have a question [16:18:00] befor start doing something by myself :P [16:18:31] well I started working on the tablet, found some problem but fixed now [16:18:48] what about creating some animation there? tablet are very powerful.. [16:19:16] only for tablet I mean, for other device we let all "static" [16:20:59] Astagi: how do you plan on separating them? [16:21:14] also, first, we should do profiling and figure out where our slowness is :) [16:21:22] Astagi: have you run our app in a browser before? [16:21:33] I can use isTwoColumnView function ;) [16:22:05] nope [16:22:49] Astagi: have you done js profiling before? [16:23:04] if we can profile and find out where the heavy js parts are and remove them, then maybe we can have animations everywhere :D [16:23:47] you mean using firebug? [16:24:03] chrome's equivalent [16:24:03] yes [16:26:12] hm I see what you mean :) so what have I to do now? :D my idea was just see screen size and start animation ( large screen, big power :P ) [16:26:23] Astagi: that's not necessarily the case [16:26:32] a lot of people have cheap ~80$-100$ tablets [16:26:36] that are as underpowered :) [16:27:17] :D that's true yuvipanda, but are enough power to support little animations, aren't they? [16:27:54] doubt it. [16:27:57] i've used a few of those [16:28:05] and they're unusable without turning off all animations in settings [16:28:14] animation will take a second or two to slowly stutter through [16:28:19] and render tablet unusable till then [16:28:31] besides, content + features + speed first, no? [16:28:32] :) [16:28:59] oh I see :\ thanks for reporting this experience, my previous law has to be deleted :D [16:29:11] :) [16:29:14] yes , first them ^^ [16:29:29] on my prev. phone too - completely unusable without turning animations off in settings [16:29:37] and that's for java based ones. js based ones will be even slower [16:30:44] I see :\ at least Android allows us to buy cheap phones with a great os ;) [16:31:20] +1 [16:31:22] it's just a pity don't have graphical performances :( [16:31:38] much better than the s40 phones i'd be stuck with otherwise [16:31:46] well, i bought it for ~180$ [16:31:52] certainly money well spent [16:31:56] despite all the drawbacks [16:32:36] yep :) [16:33:21] :) [16:33:52] oh btw! [16:34:06] I saw you are using zepto now in network.js [16:34:33] I made a refactoring to use network.js during search [16:34:57] and cancel all the previous requests once you make a new one.. [16:35:10] that's ok? [16:37:47] what did I miss? [16:37:48] got dc [16:38:45] oh I see ^^ [16:39:00] was writing about network.js [16:39:09] I see you are using zepto now [16:39:21] I made a refactoring to use network.js during search [16:39:28] and cancel all the previous requests once you make a new one. [16:39:35] that's ok? [16:39:51] Astagi: so, if i use jquery ajax to make a request there [16:40:15] it crashes [16:40:17] the app [16:40:20] when loading saved pages! [16:40:23] hence using zepto there [16:41:07] yep I read about it [16:41:20] phantoms :\ [16:48:19] yes [16:48:23] sad and horrible phantoms :( [16:49:14] who you gonna call? :P [16:50:42] as someone who never grew up with Ghostbusters, nobody [16:50:43] :) [16:51:03] ahaha :D [16:51:39] well what to you think about refactoring yuvipanda? [16:51:52] I'm in doubt if delete any previous request :\ [16:52:10] Astagi: i'll look at it once i'm done with this css fix i'm doing [16:52:14] Astagi: been too much travelling [16:52:14] :( [16:52:37] yuvipanda - btw on the subject of that ajax bug - what url are you trying to get when it seg faults? [16:52:48] jdlrobson: file:/// [16:53:00] was it html, json, xml, text, other? [16:53:19] np yuvipanda, so let me know if there's something to change, will do that.. [16:53:33] jdlrobson: text [16:53:47] hmm, it is actually html but sent with type: 'text' [16:55:07] are you working on the graphical refactoring yuvipanda? [16:55:20] Astagi: yes, the 'de-spritify' things [16:55:52] nice :) [16:56:31] does that html have or jdlrobson: might, not that i know of [17:00:25] i can check [17:00:27] jdlrobson: why? [17:03:13] bit of a stab in the dark really - http://code.google.com/p/android/issues/detail?id=6593 [17:03:20] guys I have to go ;) have a good work!! [17:03:27] bye bye! [17:03:38] people reporting ajax requests with certain results crash the android browser, wondering if it might somehow be related to the problem here [17:07:35] jdlrobson: that's interesting. I should try reproducing it with the browser [17:07:50] jdlrobson: however, note that it seems to be only if i request it with jquery, and not by zepto [17:09:07] jdlrobson: view-source:http://en.m.wikipedia.org/ is basically same structure as what i'd be saving/retreiving [17:09:38] and it *does* contain those [17:10:48] sorry my laptops decided to start thrashing making doing anything very slow :) [17:11:06] so can you remind me of the bug? [17:11:29] i can't find the link [17:12:05] yuvipanda> [17:12:06] jdlrobson: that's interesting. I should try reproducing it with the browser [17:12:06] [5:07pm] [17:12:07] jdlrobson: however, note that it seems to be only if i request it with jquery, and not by zepto [17:12:07] [5:09pm] [17:12:07] jdlrobson: view-source:http://en.m.wikipedia.org/ is basically same structure as what i'd be saving/retreiving [17:12:07] [5:09pm] [17:12:07] and it *does* contain those [17:12:12] http://code.google.com/p/android/issues/detail?id=6593 [17:12:16] is the link you gave me [17:12:24] sorry the link to bugzilla bug :) [17:12:41] ah [17:12:43] wait [17:13:54] the other theory I have is had was that when you do jQuery ajax and request dataType html it will evaluate any included script tags [1] but if you are requesting text that's unlikely to be the problem unless it is inferring [1] http://api.jquery.com/jQuery.ajax/ [17:14:08] jdlrobson: https://bugzilla.wikimedia.org/show_bug.cgi?id=34207 [17:14:12] thx [17:54:33] Yuvipanda: did you see the log of WTmobile standups? [18:05:19] heyyyyyy [18:05:32] what about our standup? [18:11:07] good timezone sheaclare. [18:13:19] Amgine: You can see that my timezone is all messed up? [18:14:12] No... people in this channel are all over the globe, so saying "good timezone" ~= "goo time-of-day in your timezone" [18:14:28] good, even. [18:15:04] Amgine: heya [18:15:08] MaxSem: :D [18:15:11] what about it? :) [18:15:38] Amgine: we've switched to good $timeOfDay [18:15:50] Amgine: no i haven't [18:16:15] yuvipanda: https://www.mediawiki.org/wiki/UCOSP_Spring_2012/Standups/07_Feb [18:16:41] ah, no groggypanda on that one [18:17:00] That's why I thought you'd want the log. [18:17:14] :) [18:17:29] 13 more minutes before I turn 21, let me see if I can fix a bug and read logs before that [18:17:40] [18:23:24] yuvipanda, you should be runnin' to a liquourstore now, to buy some champaigned in 13 minutes;) [18:23:33] MaxSem: i tried, nothing nearby [18:24:07] * MaxSem waves at philinje [18:24:29] hi there, sorry to be late [18:24:57] philinje: heya [18:24:59] aude: around? [18:27:36] yuvipanda: wondering about the tiling server? [18:28:27] philinje: yes [18:28:27] aude had some ideas [18:28:47] philinje: i'd love to use the mapquest tiles, but tfinc was worried about legal issues. [18:29:00] i don't see any, but can you pick who'se the best person to talk legal issues to? [18:29:18] yes, it's Michelle [18:29:30] is there a URL I can send to her? [18:34:13] yuvipanda: it looks like Mapquest charges money to developers, did Tomasz know that? [18:34:28] philinje: no i don't think so [18:34:34] wait for it [18:35:16] ok, they have a community version [18:35:29] Is there a reason not to use osm? [18:36:43] philinje: http://developer.mapquest.com/web/products/open/map [18:36:47] just the OSM tiles [18:36:56] there's a better link available though [18:37:33] so why is Mapquest better? [18:39:06] philinje: http://wiki.openstreetmap.org/wiki/MapQuest [18:39:29] [18:39:40] philinje: OSM we can't use at scale [18:39:57] they have a license limit [18:41:02] ok [18:41:27] You can, however, copy and serve osm? [18:43:20] Amgine: yeah, i think [18:46:56] it looks like that is what Mapquest does [18:47:14] so I assume Mapquest has more infrastructure than OSM directly? [18:47:54] sorry was on a call [18:47:54] philinje: yes, way more infra [18:48:03] philinje: and they're giving it away for free [18:48:10] philinje: and we'd want to use it only till we get our own tileserves up [18:48:42] is this the right legal page: http://developer.mapquest.com/web/info/terms-of-use [18:50:26] philinje: yes [18:50:42] philinje: or possibly not [18:50:47] it says 'Open API' [18:50:55] not sure if same as OSM tiles [18:50:58] hi [18:50:58] philinje: find someone at Mapquest, give a call and talk? [18:51:11] philinje: aude said he knew someone [18:51:17] adn they'd be glad to help us out [18:51:21] hey aude [18:52:08] do you think mapquest would be acceptible, per privacy policy? [18:52:34] aude: we don't know which privacy policy applies [18:53:55] yuvipanda: ok [18:54:11] i've pinged them to find out more [18:54:32] aude: do you know someone at mapquest philinje could talk to? [18:54:48] check this out: http://developer.mapquest.com/web/info/terms-of-use-add [18:55:02] yuvipanda: i know a bunch of people but need them to point me to the right person [18:55:43] ah okay [18:55:43] the mapquest open stuff is okay to use, but we still need to give them a heads up, given our scale [18:56:02] what are the chances to get 1-2 servers for mirroring? [18:56:12] or trying stuff on labs? [18:56:26] philinje: ^ [18:56:44] is philinje the ops person for this? [18:57:10] I'm in the OSM dev channel, btw. [18:57:32] Amgine: ok [18:58:02] irc://irc.oftc.net/osm-dev [18:58:40] heya brion [18:58:50] hey dude [18:58:54] ShareKit on iOS - i see XCode project references them, but there actually seem to be no files [18:58:58] am I missing something? [18:58:59] brion!!!! [18:59:03] also how's your health? [18:59:20] not been at 100%, but i'll survive eh [18:59:37] yuvipanda: you need to do a 'git submodule init' and 'git submodule update' [18:59:45] and then again within the sharekit dir [19:00:02] yuvipanda & philinje, first stage of my work on GeoData is complete. Now I need to see the visualisation of results. Only this way we will be able to see if there are some hiccups or if there are more filtering options needed [19:00:07] survival of the brionest [19:00:09] that needs adding to https://www.mediawiki.org/wiki/Mobile/PhoneGap/iOS_build_instructions [19:02:01] added i tihink that should work [19:06:50] back in a bit [19:15:17] What is the current number of downloads, roughly? [19:16:15] Amgine: ~1.2M [19:16:28] Thanks. [19:18:30] wow! [19:19:26] Yah, I'm kinda... "how long has it been out?" [19:19:35] ~3 weeks [19:19:49] aude: that's just android. [19:19:57] for OSM, we'll roll out on both android *and* iOS [19:20:14] so expect that to be at least double that fast. [19:20:19] i doubt the toolserver can keep up [19:20:35] oooh.... iOS [19:21:01] we'd at minimum need a caching server [19:21:12] even with mapquest? [19:21:20] maybe just squid, even? [19:21:21] philinje: can you hook aude with someone in ops? [19:21:26] as a proxy [19:21:32] idk if that would wor [19:21:33] k [19:23:01] there is no one here yet [19:23:23] here? [19:24:19] in the SF office [19:24:32] strangely empty today [19:24:35] philinje: hehe [19:24:51] either way, you need to 1. get in touch with mapquest folks 2. find someone in ops for aude to talk with :) [19:24:53] CT is in a meeting and I will ask him soon [19:25:13] okay~! [19:26:15] MaxSem: ok, thanks [19:26:29] yuvipanda: what did you think about the usage restriction? [19:26:42] philinje: from my reading of it, i don't see how it affects us [19:26:51] licensing issues only if we derive data from them - not something we're doing [19:26:55] we're just using them for base tiles [19:26:57] so looks okay to me [19:26:58] but [19:27:00] IANAL [19:27:07] ha [19:27:21] you looked at this: http://developer.mapquest.com/web/info/terms-of-use-add [19:27:24] right? [19:27:43] 5000 geocodes, 5000 routes, 5000 searches per day [19:28:15] doe anyone remember how many searches we are eyeing right now? [19:28:18] seeing [19:30:00] * aude can connect with the mapquest folks [19:30:12] already asking them about wikimania sponsorships ;) [19:30:56] unfortunately, I don't have access to the GeoNames account [19:31:13] and expects mapquest people at wikimania hackathon [19:31:39] is 5000 searches per day a low number? [19:31:53] with 1.2 million installs, yes. [19:32:20] how are you using searching and geocoding? [19:32:46] i think we were getting 25,000 GeoNames accesses per day a week ago, but that was individual articles, I believe [19:32:46] philinje: those aren't the right terms of use i think [19:33:17] i went to the Community license, then there was this link to additional terms where the usage limits are spelled out [19:34:14] so I suspect we could pay for an Enterprise license [19:34:22] philinje: okay just to be clear - we're NOT using mapquest [19:34:27] we're using Mapquest's OSM tiles [19:34:29] If your application will get heavy usage (currently defined as more than 4,000 tiles per second), please let us know by [19:34:35] which is NOT mapquest per se [19:35:00] philinje, http://www.geonames.org/img/charts/ws-creditsld8163-NhBKEPcu6P86DGxC7SYivA.png [19:35:19] we should ask them anyway but i doubt it's a problem for them [19:35:31] more, is it a problem for us, per privacy issues? [19:35:31] thanks [19:35:47] ok, understood about privacy issues [19:35:56] aude: we don't know yet - need to figure out *which* of their privacy policies applies to us [19:35:58] which is why caching or better yet, rendering our own is best [19:36:03] yuvipanda: i don't know [19:36:09] I think they'll have a separate one for their OSM tiles vs their normal map stuff [19:36:10] we need to ask [19:36:12] so we need to talk to them [19:36:16] maybe [19:36:24] MaxSem: the credits and requests are equivalent to articles, right? [19:38:59] i will email open@mapquest.com and see what happens [19:39:19] does anyone remember the difference between credits and request in the GeoNames chart? [19:39:42] aude: Okay, I've worked out a hardware setup for rendering with my host. It should be fast enough, but you would need to actually do the software setup. [19:39:51] philinje, 1 request = 2 credits eaten [19:40:05] GeoNames hate us:P [19:40:18] and each request is an article? [19:40:47] GeoNames is a really easy local setup. Most of their contributors set it up on their home machines. [19:40:48] Amgine: that's nice, do we need to connect you with Ops? [19:40:54] each request would be one view of the 'nearby me' feature [19:41:00] on BOTH ios and android, iirc [19:41:04] no, each request is "find me some articles around (foo,bar)" [19:41:19] ok [19:41:23] philinje: I dunno, was just thinking this is a temporary until you set it up on a WMF system? [19:41:26] philinje: i've emailed them [19:41:35] i can put you in touch asap [19:41:43] aude: thanks [19:42:03] Amgine: what host? [19:42:12] ArxNet [19:42:17] ah [19:43:14] for the geonames stuff, isn't the geodata extension supposed to replace it? [19:43:18] Amgine: yes, i don't know the policies around servers, etc. [19:43:30] aude: yes [19:43:40] philinje: cool [19:44:02] Max wants us to start testing his new API [19:44:43] nice [19:44:50] is it on labs? [19:49:19] aude, http://mobile-geo.wmflabs.org/w/index.php/Main_Page [19:50:27] wa [19:50:27] h [19:50:32] i managed to fuck up my git repo? [19:50:33] :| [19:50:52] okay, this is too late now. need to be up tomorrow morning [19:50:56] gnite everyone! [19:51:01] nn yuvi [19:51:01] gnite philinje aude MaxSem :) [19:51:05] and Amgine [19:54:54] Project WiktionayMobile - Nightly builds build #6: FAILURE in 6.4 sec: http://integration.mediawiki.org/ci/job/WiktionayMobile%20-%20Nightly%20builds/6/ [19:54:55] * stagi.andrea: Ignore temporary files [19:54:55] * stagi.andrea: Fixed all the enable-disable problems [19:54:56] * yuvipanda: Saved pages work again. Using zepto just for the AJAX. UGH [19:54:56] * yuvipanda: Do 'page saved' notification only *after* page is saved [19:54:57] * yuvipanda: Creation of a phonegap plugin for the store of preferences in android preferences storage. [19:54:57] * pfhayes: First commit [19:54:58] * pfhayes: Adding phonegap [19:54:58] * pfhayes: Adding phonegap jar [19:54:59] * pfhayes: Adding build.xml [19:54:59] * pfhayes: Removing binary files [19:55:00] * pfhayes: Adding jquery [19:55:00] * pfhayes: Renaming Wikipedia -> Wiktionary [19:55:01] * pfhayes: Adding custom wiktionary styling [19:55:01] * pfhayes: Adding todo [19:55:02] * pfhayes: Use iScroll on iOS 3 (untested) or 4 (tested 4.3 sim) [19:55:02] * pfhayes: Renaming Wikipedia -> Wiktionary [19:55:03] * pfhayes: removing near-me...again [20:03:28] pfhayes: did you read the log from yesterday? [20:03:53] Amgine: no not yet, i wasn't aware it had been posted [20:04:21] Yah, I didn't e-mail it. [20:04:21] https://www.mediawiki.org/wiki/UCOSP_Spring_2012/Standups/07_Feb [20:04:43] Yippie, build fixed! [20:04:44] Project WiktionayMobile - Nightly builds build #7: FIXED in 8.6 sec: http://integration.mediawiki.org/ci/job/WiktionayMobile%20-%20Nightly%20builds/7/ [20:04:44] pfhayes: Fixing another Wikipedia -> Wiktionary bug [20:04:46] Brion is under the impression you have access to the wiktionary github [20:07:17] Amgine: that is true, but i think we need to get rid of the WiktionaryMobile branch currently under github.com/wikimedia, and replace it with a fork from github.com/WikipediaMobile [20:07:30] *github.com/wikimedia/WikipediaMobile [20:07:54] If you know how to do this, do it. [20:08:11] I can't [20:08:20] i am not an admin on the github.com/wikimedia account [20:08:31] what needs to be done is [20:08:45] a) delete the repository under github.com/wikimedia/WIktionaryMobile [20:09:06] b) fork the repository github.com/wikimedia/WikipediaMobile [20:09:19] and that will be the new WiktionaryMobile repository [20:09:36] but I don't have the authorization to create new repositories under github.com/wikimedia [20:10:55] Okay. Tfinc is (or very shortly will be) in IST, which is UTC + 5.5. So, 10.5 hours ahead of you. Can you start stopping into this channel in the mornings and try to trap^Wtalk to him? [20:11:34] does he have an email address, or some other way of contacting him? [20:11:43]