[00:00] D'oh. All it took was rebuilding ES. [00:00] Oh well~ [00:00] Forgot I had that sitting around stale. [00:00] :) oops [00:13] hatch: ok [00:16] cool thanks [00:16] it's not 'quite' blocking me because i need to finish up some tests [00:16] but would like it to be next in line [00:17] * rick_h_ goes to look at what it was [00:30] hatch, you still on OSX? [00:45] Makyo [00:45] oops [00:45] Makyo yes, I'm waiting for the results of rick_h_'s bug before I go Trusty [00:46] hatch, have invites for atom.io if you want, for OSX. [00:46] Oh I saw all the kerflufl about this [00:46] too bad it's written in coffeescript :P [00:47] What's Sublime written in? [00:47] Python [00:47] but it's closed source [00:47] And do you like Python? [00:47] more than coffeescript [00:47] :P [00:47] Does it suddenly matter what your editor is written in? :D [00:47] haha no [00:48] Hehe [00:48] just commenting that it's unfortunate [00:48] it's going to hurt contributions [00:48] if they care... [00:48] It is? [00:49] yeah there are quite a few 'studies' that show that coffeescript dramatically reduces contributions vs the same thing in js [00:50] How old is CoffeeScript? [00:50] not entirely sure...my guess 4 years? [00:50] Javascript is 19. [00:51] So if it's almost five times as old, I'd guess 5 times as many people know it, and that's assuming only linear adoption. [00:51] right...the issue isn't with the language maturity it's that js people prefer to write js not coffee script [00:51] by a huge margin [00:52] since coffeescript isn't it's own language you need to know js to write coffeescript [00:52] so it's purely a style choice with syntax at that point [00:52] It isn't? The coffeescript interpreter is written in coffeescript. I'm not arguing for the syntax, I just think you have your reasons confused with your preferences. [00:53] every person I've talked to who writes coffeescript thinks in javascript [00:53] then writes the syntax for coffeescript [00:53] so until the browsers interpret coffeescript and give errors in coffeescript there won't be big addoption [00:53] imho [00:54] :) [00:54] you're better off writing Dart then compiling to js because then you get the performance gains associated with their compiler [00:54] Oh, sure, but there's gotta be force behind that. Saying JavaScript is the end game is pretty silly. Having someone like GH throw their weight behind CoffeeScript is going to add a lot of weight behind that. [00:55] it might, I'm not convinced that a whitespace significant language CAN work on the web [00:55] you can't compress it [00:56] well at least not without using some form of character substitution [00:56] Sure. [00:57] atm I think Dart holds the most promise as a js replacement - but js is closing in on that too :) [00:58] Cool. [00:58] Dart seemed alright. [00:58] I'm not totally convinced by it....but I think it's showing the most promise [00:59] the fact that their Dart2JS compiler can produce js which executes faster than you can write is a definite plus [00:59] of course thats on a case-by-case basis :) but even if it's == [01:00] the Dart ecosystem brings with it some things that other languages have had for a long time [01:00] module system, module repository etc [01:01] so many languages though....it's going to be interesting to see what shakes out server and client side [01:02] there is also multi threaded node http://jxcore.com/ [01:02] heh [01:07] Makyo you were saying that you were a fan of CS right? [01:07] CS? [01:07] crappyscript [01:07] er I mean coffeescript [01:07] :P [01:07] oh... [01:07] I've used it, it's okay. I just don't care what my editor is written in ?.? [01:08] Current project is a game in d3+CS. [01:08] oh I don't care either :) I was just commenting that it was unfortunate [01:09] You just spent 25 minutes arguing about it and ended by calling it 'crappyscript'. Sounds like you care [01:09] no we were discussing it [01:09] :) [01:10] Trying to, at least :) [01:12] haha [01:15] messy giuseppe for supper tonight! [01:16] Makyo is your game open source? [01:17] Will be. I have, like, one and a third levels so far, though. Writing it in Literate CS and including the source within the game as a mixed-media thing. [01:18] Well, I mean, it's on GH now, but not in a very playable state. [01:18] cool :) what's it about? [01:19] Goin' crazy :) Asks you to do orderly things, but they keep getting more and more corrupted the longer you do them. [01:20] haha that definitely sounds like it would get crazy [01:25] Yeah. Like, asks you to write the numbers 1-100 at a certain, randomized pace, but if you mess up, it starts deleting numbers. If you succeed, things still get corrupted, won't let you go on until you finish: http://imgur.com/S2hZ74c [01:31] lol that would drive me nuts [01:32] That's the point :) [01:32] Well, not you. [01:32] Well, maybe also you. [01:36] haha === rogpeppe1 is now known as rogpeppe [13:15] frankban: heh, so that test is a bit bonkers. The only thing I can think is that the .eql is doing some type conversion? charm.get('last_change').created == date is false [13:15] frankban: so that's a bad test and has been for some time. [13:16] frankban: that last_change only exists in this test so I'd just remove that test as not something supported any longer [13:16] rick_h_: +1 [13:16] * frankban lunches [13:32] hey rick_h_, i'm having a chicken/egg problem with charmworldlib regaring setuptools and ez_setup. you got a second to chat? [13:33] bac: sure thing [13:34] * bac launches hangout [13:34] rick_h_: https://plus.google.com/hangouts/_/calendar/Z2FyeS5wb3N0ZXJAY2Fub25pY2FsLmNvbQ.t3m5giuddiv9epub48d9skdaso [13:35] rick_h_: ^^ === marcoceppi_ is now known as marcoceppi [13:41] bac: https://github.com/bookieio/bookie_api/blob/master/setup.py [14:04] morning [14:06] morning hatch [14:06] guihelp: anyone available for a quick review of a critical card? https://github.com/juju/juju-gui/pull/147 [14:08] frankban: looking [14:08] rick_h_: thanks [14:12] frankban why is this regen of the shrinkwrap necessary? [14:12] how did we get along without it? [14:13] What was the change in our code that required this? :) [14:14] hatch: while time travelling I ended up in a time/place where zipstream-ctalkington/0.3.4 does not exist, and so the GUI build fails [14:15] lol [14:17] interesting it downgraded the zipstream module [14:18] yeah... interesting... [14:18] hatch: can you run another QA on it please? I want to make sure there's no side effects and hte more people that test/qa the better [14:18] the tarbal in npm is for 3.3 but it shows the most recent version is 3.4 [14:18] i think the module is broken [14:18] * rick_h_ grumbles about not having a offline folder for all these and not hittin the internet for them ever [14:19] wish we could do that === frankban_ is now known as frankban [14:32] rick_h_: for a $5 donation i secured bac.ngrok.com. whee. [14:33] sweet [14:33] * bac regrets not getting mitechie.ngrok.com [14:34] hah, all good [14:35] frankban: phantom crashed on a test run :( [14:36] hmm, but passed on ci and a second run passes so maybe my machine [14:36] rick_h_: on your local machine? [14:36] rick_h_: oh, ok [14:37] frankban: yea, just not happened that I can recall. [14:37] so bad timing to have it happen during this big dep update [14:37] yeah [14:37] bah, did it a second time [14:37] * rick_h_ goes to try it on the desktop to compare machines [14:37] rick_h_: how that looks like? [14:39] http://paste.ubuntu.com/7005330/ [14:39] the other crash happened in different tests [14:41] hatch: any issues trying this out? [14:41] It's running ok here, I'll try a new fresh one [14:41] maybe I didn't clear something properly [14:41] ok, make sure to clean-all to get the new deps [14:44] ok it's running again [14:46] npm could probably dramatically reduce it's server load if it allowed people to create rollups :) [14:47] then we wouldn't need to make 1BILLION http requests [14:47] lol [14:48] rick_h_ yeah runs good here [14:48] I'll try prod [14:50] rick_h_ looks like whatever you added puts phantom in a rocky state [14:51] are you running the most recent version? [14:52] hatch: since this failure seems to be intermittent, could you run make-debug/prod several times? [14:52] hatch: I am doing the same [14:53] sure [14:53] ok it's running 3x [14:53] will report back [14:54] hatch: cool thanks [14:57] passed all good [14:58] ok, updated my lxc and restarted everything and have hda 3 good test runs in a row [15:03] rick_h_, hatch: very well, landing it [15:37] man I hate it when I send an email from the wrong account [15:39] I need a 'are you sure you want to send from this account' dialogue [15:51] jujugui call in 10 [15:51] hatch of Makyo (whoever has good interwebs) can you run the call? I've got to stop in another call for a couple of min and will be late [15:51] on it [15:52] my interwebs are fixed, and I figured out the issue with my MBP and my network [15:52] the MBP tries to draw too much bandwidth from the wifi and then the wifi kicks it off momentarily...... [15:53] I mean...the first router was broken....but the issues yesterday were caused by ^ [15:53] the tech lady who came to replace the router told me about it...and she was right [15:53] :) [15:58] how does the MBP do that hatch? [15:58] bac not sure, my guess is that the router has a max transfer rate and when it starts to peg over that (apparently around 600KBps) it drops the connection [15:59] hatch: i'd think that would be negotiated and rate limited [15:59] rather than giving it the boot [15:59] you'd hope anyways [15:59] jujugui call in 1 [15:59] hatch: i think you need one of those cool airport towers [16:00] haha I have one of the fancy asus routers on hold for me [16:08] hmm [16:08] I wonder why I Don't have video [16:08] it was odd...it was black [16:08] how was the sound? did that come through ok? [16:09] audio was ok while I was on there [16:09] hmm ok that's so odd [16:09] maybe my camera broke [16:10] quit breaking stuff! [16:10] today has been an expensive week [16:10] wifi, dishwasher, (maybe) camera [16:10] hopefully I can fix the dishwasher cheap though [16:12] jujugui can we create 'fake' email addresses with our canonical emails? like you can with a typical gmail account? [16:12] oh I guess I could just test it myself [16:12] hatch: :) [16:13] Sent, hatch [16:13] hatch: let us know how it works out [16:13] yes...we...can! [16:13] ROCKON!!!!!! [16:14] rick_h_ my username is hatched [16:17] rick_h_: i think only moderators need to register on planning poker [16:17] bac: oh really? /me goes to look again [16:17] here's a game i created http://bac.planningpoker.com/play/games/215704 [16:17] see if you can join [16:19] wrong url. this one works http://bac.planningpoker.com/participants/new?token=au32sd [16:20] rick_h_: anyway, i'm bac on that site [16:20] bah, second. chrome is crashing on me since the update this morning :/ [16:21] ah cool thanks bac [16:21] clearly you underestimate the difficulty of making a good drink [16:21] hah [16:22] I assumed you can enjoy Big Lebowski even with a bad drink [16:22] haha [16:23] touché [16:24] * bac has never actually had a white russian as it sounds gross [16:24] yea, my wife likes them. Never felt like something I'd like [16:25] They're a cop-out drink. Laaazy. [16:25] Which I guess was part of the point in Big Lebowski :) [16:26] * rick_h_ is going to fetch lunch for me and sicko boy, biaf [16:29] haha [16:29] a white russian is a very hard drink to make properly [16:29] I've only ever had it done properly once in a bar [16:30] rick_h_, Makyo: machines calls ready for review: https://github.com/juju/juju-gui/pull/148 thanks! [16:30] Woo! [16:48] hatch: re bug 1285223 after a first local deployment, when dragging the same local charm again I get what it seems to be a "Local Charm Upgrade" broken inspector and no errors. this is in comingsoon [16:48] <_mup_> Bug #1285223: fakebackend error when deploying the same local charm twice [17:19] Gah, can't get core trunk to build. [17:24] Makyo: deps problems [17:24] ? [17:24] Makyo: godeps -u dependencies.tsv can help [17:25] frankban, I think so. fetching it again. everything undefined in utilts/ssh [17:25] Okay,w ill remember that in the future. [17:25] Hadn't used godeps before [17:25] Makyo: yeah at least it should warn you about missing deps/revisions [17:25] frankban, okay, cool [18:08] hey [18:08] how [18:08] frankban yes do it on jujucharms.com not comingsoon [18:10] hatch: ok I'll do tomorrow morning, and look for a way to dupe in trunk too [18:12] frankban ok thanks, the issue with comingsoon and trunk is that the UI stops the flow [18:12] if I can get my current branch landed the UI now has a button which allows you to resume the flow [18:12] hatch: IC, cool [18:12] but jujucharms.com (or an older build) will let you do it without the local charm upgrade UI [18:29] That machines stuff is awesome [18:29] Makyo: isn't it! Getting excited to do this machine view stuff now. It's going to be so cool [18:34] yeah once we have the machine stuff that's going to be awesome [18:45] hi marcoceppi, can you look at this charmworldlib branch when you get a chance? https://codereview.appspot.com/69430043/ [18:46] Makyo, rick_h_: so i learned all about flask last night. does look nice. i was wrong, though, as it was all in spanish. you can pick up a lot when it is spanish sprinkled with python and javascript. [18:46] bac lol [18:46] bac: cool still [18:46] at the coffee shop the other day it was funny because there was a spanish conversation going and they were talking about phones and such [18:46] fjoiewjfiewjfioew fjoiew facebook fjoiefjiweojfow gmail fjoiejfewjoifew motorola [18:47] it's interesting to hear tech talk in other languages as there's just enough to match that makes you think you're understanding a bit [18:47] rofl [18:48] hah, one dominican lady we know kept talking about her new phone and 'febo' (pronouned 'fay-boo'). turns out it was facebook [18:49] yeah, it was interesting what didn't get translated. stuff like 'GET request' [19:04] hey rick_h_ I can't edit/comment on the state url doc [19:05] hatch: oh, updating [19:05] I want to put a big red X over ALL of it [19:05] mohohahahaha [19:06] lol, everyone else but you can [19:06] haha u suck [19:07] you should now [19:07] hey, you weren't on the juju gui folder [19:07] so you didn't get it for the files [19:07] oh really? that's odd [19:07] I mean, it's only been over a year since I started here....it would make sense that all the 'when you start' things aren't done yet [19:07] lol [19:16] rick_h_ the machine view folder is empty now? [19:17] s'ok I'll just use the pdf [19:18] bac: I'm confused why this logic is showing up in the charm-tools itself, I though the point of having an "online" proof is that you guys could control this stuff in the API? I'm happy to accept it just trying to figure out what lives in charm-tools and what lives in the api (rick_h_?) [19:18] marcoceppi: not charm-tools but charmworldlib. [19:19] oh, wait [19:19] marcoceppi: we have several places that need to share this code [19:19] bac: gotchya [19:19] bac: okay, I was slightly confused [19:19] marcoceppi: that is understandable. [19:21] marcoceppi: if you could, review what is there but i'm in the process of moving the new stuff into a utils module. that removes the dependency on python-requests for users that just want the constraints stuff. [19:21] bac: ack [19:29] hatch: pdf [19:29] ? [19:29] hatch: oh the design [19:30] yeah, tis cool [19:42] hatch: http://responsiveimages.org/ [19:42] from http://blog.chromium.org/2014/02/chrome-34-responsive-images-and_9316.html [20:23] yay chrome fix released for dev channel [20:23] that was unusable [20:24] oh no, spoke to soon [20:26] rick_h_, trying to get to http://paste.mitechie.com/show/1048/ to see bundle proof output; your pastebin up? [20:26] Makyo: no :( [20:26] rick_h_, okay, just checking. Can run from cli [20:27] * Makyo lazy [20:27] it's not running is my laziness [21:18] jujugui: can the gui tests not be run on lxc? i get a deploy error trying to put the gui on machine 0 stating the series don't match. they won't match. host is trusty, charm is precise. if that is the case i'll update HACKING.md === BradCrittenden is now known as bac [21:20] bac: I run the gui tests in an lxc. The phantomjs tests? [21:20] or do you mean a charm tes? [21:20] test [21:20] soryy [21:20] s/soryy/sorry/ [21:20] yes, charm tests [21:21] bac: ah, yea I'm not sure about those [21:21] I think I've had issues in that past but not traced it down [21:22] rick_h_: ok. i'm running on ec2 now. super slow. [21:22] yes, they are [21:22] good time for a [21:22] * bac -> dogwalk [21:56] Okay, I have a good feeling for where/how existing charm checking is done and where it might go. rick_h_ want me to turn this into an investigation card and come up with a proposal? [21:58] Makyo: yes please [21:58] if bac comes back and you can do a pre-imp that'd be great [22:00] Sure thing. Dogwalk soon here, too, but will writeup first. [22:17] rick_h_, bac https://docs.google.com/a/canonical.com/document/d/1gp6v6_S-sPmYKI5NIxt0kI7mbhBtRI_JhXjG7tnkQQI/edit?usp=sharing Going to dogwalk quick before it gets super cold. Back in a bit. Will have a task ready if it's too far past EoD for folks, can pick this up in the AM [22:25] Makyo: i'll look at it later or in the morning [22:25] bac, Sounds good, will do cleanup stuff.