[12:03] jujugui: anyone have time for a review: https://codereview.appspot.com/9449044 [12:03] bac I'll look at code [12:04] thanks gary_poster [12:04] huh, that's interesting [12:04] so when you focus on it you can see you have more room [12:04] but normally it is compressed [12:05] that seems nice on the face of it :-) [12:05] bac do you want qa, or is code check ok? [12:06] gary_poster: luca has looked at it and liked it [12:06] gary_poster: qa would be swell if you have time to spin it up [12:06] ok will check [12:06] gary_poster: it's pretty fun! :) [12:06] :-) [12:15] bac. LGTM, qa good [12:16] thanks gary_poster. just need one more jujugui review now... :) [12:16] :-) [12:16] bac: I can. Do you need QA too? [12:17] benji: nah. i'll put up a branch for luca to look at [12:17] benji: *finally* got that lp2kanban branch landed. [12:17] bac don't forget that he knows how to start up a branch himself now [12:17] k, I'll look at it now [12:18] he == Luca [12:18] oh really, what pushed it over the edge [12:18] gary_poster: oh, nice [12:18] he got it set up at sprint [12:18] luca, can you have a look at the branch in review at https://codereview.appspot.com/9449044 [12:18] the fact that we have the sandbox as default now means it is easier [12:19] it is the auto-resizing-textarea-on-focus work i showed you in oakland. [12:19] i guess london is lunching [12:23] benji: i didn't put FileSaver in there. i just made the Makefile aware [12:23] benji: so you're preaching to the wrong choir. :) [12:23] heh [12:23] i'm not sure how it ever landed, tbh [12:23] even if you did, I wouldn't blame you, it's the /system/ man! [12:24] it was a mess and beautify threw up on it. isn't 'make prep' run as part of lbox? [12:25] * bac tunes .procmailrc...again [12:41] bac, benji, app/assets/javascripts/ is where we are supposed to put our vendor imports that don't have npm sources, as established from before we joined the project. It also has our code that we want to open source. Are you advocating separating them, or something else? FWIW, .lbox.check runs make check, which runs "lint test-prod test-debug test-misc" so if lint is ok then it will land. beautify is only suppos [12:41] ed to fix what the closure linter would complain about...ah, and the linter honors LINT_IGNORE [12:41] where the beautifier ignores it [12:41] so that's the diff [12:42] not sure why we have that diff [12:42] but anyway...please make a card for something you think we should improve :-) [12:42] right, my contention is that putting dependencies somewhere other than under app/ would be an improvment [12:42] I [12:43] will make a card now [12:43] quite impressive: http://thomasstreet.net/blog/spreadsheet.html [12:43] so many cards in the "Discuss" slot [12:44] that's because we haven't discussed for three weeks :-/ [12:44] and we are missing two people today, both of whom should be a part of some of these discussions [12:44] frankban, yeah, angular looks pretty cool to me on the face of it [12:45] * benji adds a "Discuss why we haven't discussed things in the Discuss lane" card. [12:45] :-P [12:45] we had a sprint! [12:47] benji: I see you have two very similar discussion cards: either my board is broken or you really really want to reduce noise emails [12:47] heh [12:47] heh [12:48] that might be from a board hiccup; I'll fix it [13:04] bac: just back from lunch, I'll check it now [13:04] thanks [13:07] eek [13:07] * gary_poster realizes we have no licensing whatsoever [13:10] who has no licensing? [13:10] juju gui [13:10] you mean in the code? we've got GPL3 listed on LP [13:10] yeah, the code :-) [13:10] I think we are should be using AGPL [13:31] hatch, Makyo are event-tracker.js and unscaled-pack-layout.js in the assets/javascripts folder ours, or taken from somewhere? [13:34] * gary_poster wonders about putting licensing in handlebars templates, and really doesn't want to. [13:39] successfully branched my first branch. I'm gonna bzr the pentagon now. Eat it! [13:46] bac: thanks for the review! [13:47] guihelp: anyone want to take another look at a quick Python branch? https://codereview.appspot.com/9464043 [13:48] luca, lol [13:48] gary_poster: :) [13:49] luca: \o/ [14:00] luca: the guy who "owns" the nick "luca" has been away for 7 weeks, when he is gone for 10-15 weeks the ops will give it to you if you ask nicely (then you won't have to be luca__________________ any more) [14:02] benji: interesting, it's annoying to have the password thing pop up hehe [14:02] frankban: looking [14:03] sadly, the guy who owns "gary" logs on frequently :-P [14:03] frankban: doh, that's your same branch. nm. [14:03] hatch: hi, do you have a minute for a quick call? [14:03] sure one minute [14:03] benji: i was able to get twitter to give me 'bac' since some dude in japan hadn't used it in a few years. i doubt they do that anymore. [14:04] it's a bit of a curse, though [14:04] frankban, after that, please ping me for possibly a slightly longer call :-) [14:04] heh [14:04] frankban: guichat? [14:04] hatch: sure [14:04] gary_poster: ack [14:04] thx [14:07] huh, teknico landed branch rev 666 yesterday and no one noticed [14:08] gary_poster, unscaled-pack-layout is D3 modified for our purposes. [14:08] ok thanks Makyo [14:09] gary_poster, Also, I tried to land the remove annotations branch last night, but every time I merge with trunk, it wipes out all of my changes, like it's resolving every merge with --take-other [14:09] weird! [14:09] Wondering if that's because the MP has already been marked as merged? [14:10] maybe but I doubt it. You can change the MP status manually yourself to check though, Makyo [14:10] gary_poster, alright, that's what I was going to try this morning. If not, I'll try a new branch. [14:11] Makyo, yeah, cool, that was my next suggestion as well. You could try a new branch and manually merging in the old one to see if it works [14:12] yay no more appcache! [14:12] :-) [14:13] I'm surprised that google didn't release any new hardware at io [14:14] gary_poster: ready in the usual place [14:20] the new hangouts shows me that gary_poster and frankban are currently in a hangout and that I can join :) [14:37] gary_poster: just curious why you didn't do a single line like (C) 2012-2013 for the headers [14:38] bac, because of https://sites.google.com/a/canonical.com/legal-services/software-ip : see the end of the "Source Code" section of "Notices for software and documentation" [14:39] IOW, I have no idea, just trying to follow their instructions [14:41] gary_poster: their instructions don't give a multi-year example. on LP we used a range, which i think is acceptable. [14:41] bac, oh, I was misreading "When modifying an existing file, if your changes are substantial enough, add a copyright line at the top of the file, after any existing copyright notice:" [14:42] they were talking about files from other sources [14:42] ok, I will switch to 2012-2013 [14:46] Gah, finally. Had to recreate changes by hand, since trying to merge old branch into a clean branch of trunk gave me "Nothing to do." [14:51] looks like CI had a canonistack error, anyone logged in to refire? [14:54] * hatch fired off another [14:58] jujugui call in 2. please update kanban asap [15:25] CI timed out.....firing again :/ [15:41] they are runnin gnow [16:19] installing ubuntu on google glass is so cool [16:24] * hatch wonders if it also runs steam [16:24] ;) [16:44] gary_poster: doing this work I have noticed that we have a :gui:/charms view - is this supposed to be here? I haven't even heard of this section until now :) [16:45] hatch, not sure on call will ping [16:45] alrighty [17:20] hatch I think the service charm tab links to that, yeah? [17:20] we want it for now, but not for long [17:21] negative - it looks like it's a place to search for charms [17:21] I think the inspector will have a :gui:[service name]/charm tab. oh, really?! hah. [17:21] um, hatch...yeah, let's rip it out. [17:21] :-) [17:22] consider it ripped! [17:22] :-) thanks [17:22] I'll do that in a separate branch [17:22] * gary_poster runs to go get lunch before our call [17:22] good idea [18:10] * bac has head-ache. lies down. bbiab. [18:19] gary_poster, got the N10 today, do I need to let anyone know? [18:19] Maybe reply to the person who sent and confirm? [19:03] benji: sent some notes about the mem leak you're fighting, I think we have a test case already that can help show the issue [19:05] at first I though I sent notes, then I though I was being asked to send notes, and then I relized: I have been sent notes [19:08] suddenly I feel less helpful [19:08] and more confusing [19:11] communication is hard [19:47] gary_poster: do you have any additional info on the temporarily missing endpoints card? [19:47] bac on call will be with you soonish and can talk then [19:48] ok [19:55] bac guichat? [19:55] sure [19:57] gary_poster: i see your frozen head [20:23] bac, could you give me your appspot review of https://codereview.appspot.com/9499043 (you already gave me the MP +1, thank you; the appspot change has the 2012-2013 change) [20:23] jujugui, could I please have one other review of https://codereview.appspot.com/9499043 ? It adds AGPL to everything, so it is gigantic and mechanical [20:24] lol wow [20:24] yeah I can do it [20:24] thank you :-) [20:30] done! [20:33] gary_poster, AGPL? [20:34] gary_poster, its distributed by usage.. so why not GPL? [20:35] as I understand it the reason prime reason is because juju-core is AGPL [20:37] plus, we may have non-web-app bits soon, so those aren't distributed in order to use them [20:37] thanks hatch [20:39] bcsaller: question: I am seeing a model in more than one model list (as returned by the model's .lists() method), all of the lists are "serviceList"s but as far as I can tell only one serviceList is created and it lives in the Database object (app/models/models.js:626). Can you confirm or deny my reading of the code? [20:42] benji: is this with fakebackend or no? [20:42] bcsaller: improv [20:42] benji: fakebackend would have its own database, but sharing there would be an issue, ok reading [20:43] well, I just half-answered my own question; I added an intitialzer method to ServiceList with a console.log() call and I am seeing many being created. [20:43] any idea why? [20:43] bcsaller: io over? [20:44] hatch: there are some coding labs now [20:44] hazmat, my thinking was what benji said, plus we effectively serve these from a server to actually be useful. that's maybe the weakest argument. I also didn't feel that GPL was any advantage of choosing GPL over AGPL. WDYT? [20:44] benji: console.trace() :) [20:44] oh, bcsaller: you're not really here; don't feel compelled to listen to me ;) [20:44] bcsaller: ahh, not partaking in any? [20:45] hatch: I did some but taking a break now, days of presentations is draining for me [20:45] yeah you can really only absorb so much [20:45] brain work is exhausting :) [20:46] benji, gary_poster, i guess given the tie in to juju-core its reasonable.. just that its always been unclear to me where the boundary of AGPL breaksdown.. its code run by the browser, so does the browser now need to be opensource AGPL compatibile to be accessed legally. ie where does the chain end. [20:47] thats like saying an open source kernel makes all code it runs open source [20:48] I like that idea [20:49] to the courts! [20:49] hazmat I'm with you in spirit that I don't find the AGPL's distinctions clear with JS apps in particular, though I don't think people would bother to try to argue that AGPL went as far as your example. From the perspective of the company, though, I thought going for the more restrictive license was a good starting point. I'm asking our lawyers about whether I need to worry about collecting user data, so I can ment [20:49] ion that on the way [20:52] bcsaller: did you happen to catch any talks on Dart? It's always interested me... [20:53] hatch: I didn't but I did see some good web dev talks in general [20:53] cool - a lot of the talks are online already so this weekend might be a remote io weekend :) [21:41] Makyo: is there a reason why you are hanging the flags off of the window object without any namespacing? [21:41] seems like it should be (at least) jujuFlags [21:42] or are we not concerned about clashing? [21:42] in this instance [21:52] is this an issue? /:gui:/service/memcached/:flags:/foo becomes /:flags:/foo/:gui:/service/memcached/ when using the built in url parser [21:52] should we put a special case in there to keep the flags namespace at the end instead of alphabetical? [21:57] also when navigating to / the flags namespace is removed - the properties are still saved so it should still work but it may not be clear to users [22:17] hatch, sorry, was dogwalking, should have said. [22:19] hatch, flags is on window for some reason I don't really remember that we glossed over at the sprint. I'll try to think back to it and ask around, and we can change it if we need down the road. [22:20] hatch, as for namespace order, so long as :flags: only gets /foo, I don't see a problem, personally. [22:21] hatch, the last might be an issue, but should be relatively easy to change if so. [22:26] alrighty - I'm just implementing stuff with flags right now and those are the things I came across so far [22:28] da na nu na nu inspector widget da na nu na nu u uh unnnnn === Makyo is now known as Makyo|out