[12:32] d [12:33] thisfred, does this help with your branch? [12:33] https://code.edge.launchpad.net/~cmiller/desktopcouch/getport-at-call-time [12:33] CardinalFang: o wow, no rest for the wicked huh? ;) [12:33] thanks, I'll have a look! [12:33] Well, I sync'd it out at the last minute yesterday and didn't tell you. [12:34] hmm, syncdaemon crashes at startup, and then report a problem crashes.... [12:35] thisfred, Eh? [12:36] CardinalFang: sorry, nothing to do with your branch [12:48] CardinalFang: encouraging news: the branch only started couch once. Will now test with higher load on the machine... [12:49] CardinalFang: ehm [12:49] CardinalFang: it only *reports* starting it once [12:49] CardinalFang: there do seem to be two running after trial, and there were none before [12:50] I don't think there's a report-er outside the start-er. Hrm. [12:51] CardinalFang: maybe I'm just misreading ps [12:51] eric 8321 0.0 0.0 1836 544 pts/0 S 07:47 0:00 /bin/sh -e /usr/bin/couchdb -c \"/etc/couchdb/default.ini\" -c \"/home/eric/.config/desktop-couch/desktop-couchdb.ini\" -b -r 0 -p /home/eric/.cache/desktop-couch/desktop-couchdb.pid -o /home/eric/.cache/desktop-couch/desktop-couchdb.stdout -e /home/eric/.cache/desktop-couch/desktop-couchdb.stderr -R [12:51] eric 8348 0.0 0.0 1836 324 pts/0 S 07:47 0:00 /bin/sh -e /usr/bin/couchdb -c \"/etc/couchdb/default.ini\" -c \"/home/eric/.config/desktop-couch/desktop-couchdb.ini\" -b -r 0 -p /home/eric/.cache/desktop-couch/desktop-couchdb.pid -o /home/eric/.cache/desktop-couch/desktop-couchdb.stdout -e /home/eric/.cache/desktop-couch/desktop-couchdb.stderr -R [12:51] eric 8349 1.7 0.2 64964 9220 pts/0 Sl 07:47 0:00 /usr/lib/erlang/erts-5.7.2/bin/beam.smp -Bd -K true -- -root /usr/lib/erlang -progname erl -- -home /home/eric -noshell -noinput -smp auto -sasl errlog_type error -pa /usr/lib/couchdb/erlang/lib/couch-0.9.0/ebin /usr/lib/couchdb/erlang/lib/mochiweb-r97/ebin /usr/lib/couchdb/erlang/lib/ibrowse-1.4.1/ebin -eval application:load(ibrowse) -eval application:load(crypto) [12:51] -eval application:load(couch) -eval crypto:start() -eval ibrowse:start() -eval couch_server:start([ "/etc/couchdb/default.ini", "/home/eric/.config/desktop-couch/desktop-couchdb.ini"]), receive done -> done end. -pidfile /home/eric/.cache/desktop-couch/desktop-couchdb.pid -heart [12:51] (sry that was a pastebin candidate I guess :) [12:52] Even pastebin can not contain it. [12:52] First two lines seem to be two separate couches to me [12:54] How many listeners do you have? [12:54] thisfred, $ netstat -apn |grep beam [12:55] CardinalFang: just one [12:55] Then you have only one daemon. [12:55] CardinalFang: and from looking at the code, it makes no sense: it printed "done." only once [12:56] so it can't have started more than once [12:57] thisfred, my new code, replacing "sleep(2)" with waiting until a process appears with the same characteristics as we use to decide we need to start something -- this should solve slow-box problems. [12:59] CardinalFang: awesome. I'll just assume this works. I'll ask on #couchdb if two processes is worrying or normal, or something in between [13:00] Rgr. Erlang is weird that way. [13:00] thisfred, mention that only one is listening for TCP. [13:01] CardinalFang: erlang is weird in almost every way, (reading erlang for kangaroos now) but I kinda like it. It borrows from Prolog heavily, which is still one of my favorite languages. [13:01] I've never used either. :( [13:02] My old habit of learning a new language every year stopped 5 years ago. :( :( :( [13:02] I did try Clojure for a while this year. [13:02] AFK 30min. [13:15] What is the "Shared With Me" directory for? [13:16] xcdfgkjhgcv: for when people share things with you [13:16] xcdfgkjhgcv: somebody shares a folder with you, you accept it, it lands there [13:17] Chipaca: Roger that. [13:59] thisfred, any news? [14:01] CardinalFang: jan____ thought it was strange, but not impossible. I'll ask Noah Slater when he gets online, or ask the mailing list. One thing to test would be to see if it goes away when stopping couch. Which points to another issue: we don't have a stop method in desktopcouch right? would not be used often, but nice to have for testing, debugging, etc. [14:10] thisfred, Hrm, it should be easy. How do you want to call it? [14:11] CardinalFang: the tests could call it in the same way as start, and maybe it should be callable from the command line as well [14:19] thisfred, tests don't call start. Part of finding the port to talk to makes it spawn desktopcouch if it's not running. So, it's trickier. [14:21] CardinalFang: well ok, same way as the portfinder calls it then. My point is, maybe we should test start and stop in isolation, and also, maybe (one of) the tests should call stop before running, so we check the portfinder [14:22] CardinalFang: discussed this briefly with statik yesterday, and it may be better to test against a dedicated testdb. Or do we already do that? [14:22] thisfred, No idea. [14:22] so running the tests won't mess with a developer's couchdb [14:23] probably not [14:23] It would mean some refactoring, I don't think it's as urgent [14:23] it's not urgent [14:24] I think a good first step would be to have a python only stop method [14:24] and call that in the setup of one test that talks to the db [14:24] Ah, it should discover whether the tree is a dev tree, and if so, to use the ubunet tmp, iirc. [14:24] * CardinalFang checks. [14:25] jdo: regarding my issue with having an Active account but not being able to access /files (and being redirected to /plans)... [14:26] mattgriffin, due to the problem in the subscription_nightly job which was fixed yesterday, your subscription didn't get renewed [14:26] mattgriffin, I can renew it manually [14:26] jdo: cool. thanks. [14:27] mattgriffin, I have renewed it and you should work now [14:27] jdo: excellent. thanks! [15:01] * jblount looks around for someone to start the meeting [15:01] MEETING STARTS [15:01] if you're working on desktop+, say 'me' [15:02] then three lines: DONE, TODO, BLOCK [15:02] take one and pass it on [15:02] me [15:02] me [15:02] me === rodrigo_1 is now known as rodrigo_ [15:02] me [15:04] that's it? Okay, rodrigo_, you're up. [15:04] hmm, nobody else? [15:04] ok [15:04] statik, vds...? [15:04] • DONE: Got tomboy 1st sync to work. couchdb-glib 0.4.3 package fixes. Published new tomboy with sync patch in beta PPA [15:04] • TODO: Add more tests in couchdb-glib test suite. More openSUSE packaging. Change tomboy syncing prefs interface to show many servers. Add social services accounts config to about-me. Talk to Ara about writing mago tests for evo-couchdb [15:04] • BLOCKED: none [15:04] teknico: go [15:04] DONE: completed and landed the phone sync subscription check branch, reviews [15:04] TODO: more reviews, branch to remove relative imports, start work on disabling free phone sync after 30 days [15:04] BLOCKED: none [15:04] next: jblount [15:04] me [15:04] DONE: Decided I need help with my js truncating stuff, talk with statik / warthogs / anyone listening about using u1, fix file names not showing up in gecko browsers. [15:04] TODO: FACE, talk with urbanape about truncating stuff, talk with contacts people about ui for that, FIX ALL BUGS ON WEBUI (right now that primarily means display quirks in css stuff) [15:05] BLOCKED: truncating stuff. It plagues me because I know there should be a smart way to do it. [15:05] urbanape: YTMND [15:05] DONE: Pushed a new branch of bindwood with a whole lotta debugging (turned off by default, but enabled with a slight edit). Approved a few branches - new files UI is go [15:05] TODO: Start fixing the new files UI, making it pretty, reliable, and robust. Help out jblount on some of that. [15:05] BLOCK: Nope [15:05] vds, you're up. [15:05] DONE: pair programming with teknico, some discussion about FX deployment, some review [15:05] TODO: still deployment stuff [15:05] BLOCKED: no [15:05] I guess that's it [15:05] Thanks, all. [15:05] MEETING ENDS [15:05] (stragglers, feel free to paste in channel) [15:05] * jblount pats urbanape on the back [15:06] dang! [15:06] lollygagger! [15:07] You broke my keyword watcher with "MEET-NG STARTS" instead of "MEET-NG BEGINS". :( [15:08] d'oh [15:08] I'll try to be more consistent in the future [15:08] huh [15:09] I've spammed the channels with thisfr-d, so you know what I'm working on. :\ [15:09] ☭ DONE: 0.92.0 release, Fixed #412150, #409474 [15:09] ☭ TODO: Finish #386443 [15:09] ☭ BLCK: 1.0a on server side (pending oauth.py upstream 1.0a patch approval) [15:10] DONE: much desktopcouch polish [15:10] TODO: 30 more minutes of polish. then back to Gwibber. [15:10] BLOCKED: None [15:15] sorry i missed the meeting, i've been on the phone [15:15] dobey: who needs to approve the oauth.py patch? [15:17] statik: upstream... Leah Culver [15:17] statik: but i am like " this close to forking it [15:18] dobey: is it a backwards compatible kind of thing? [15:18] statik: 1.0a requires API changes for the server side stuff [15:19] so is it a change we can safely apply to the packaged version in karmic, or would we need to ship two versions of the library for people to be able to upgrade on their own schedule? [15:20] dobey: this is the patch, yes? http://code.google.com/p/oauth/issues/detail?id=110 [15:20] well, we can change our private copy for ubuntuone, and not touch the python-oauth package for now. she said she has collected several patches for the issue, so i don't know what these "other patches" are like, and if she will land one of them instead of mine, with different api changes than mine has [15:20] statik: yes [15:20] statik: and http://groups.google.com/group/oauth/browse_thread/thread/8f54d2779e71db13 is the discussion thread [15:21] thanks [15:21] i pinged her yesterday in irc, and she still hasn't gotten 'round to replying to my questions, or landing anything it seems :( [15:22] dobey: i'll send a note on that thread, to see if it wakes anyone up [15:22] good luck :) [15:28] brb [15:30] python (in javascript) in the browser: http://www.skulpt.org/ [15:31] thisfred, Okay, new changes. I found one way that start-couchdb can make two to run, but I don't think that's what caused your two-count earlier today. [15:31] CardinalFang: awesome, same branch? [15:31] Same branch. [15:32] * CardinalFang hopes aquarius doesn't freak out at the changes. [15:32] Though, I hope to be away next week, so he has a chance to change things back. [15:33] hehe [15:34] It's the risk of taking a holiday, people will touch your stuff [15:36] is this known? I've seen it recently: https://pastebin.canonical.com/21052/ [15:36] CardinalFang: that definitely did something,: two couches (two processes each) are started, and the port can't be found: [15:36] CardinalFang: http://pastebin.com/d5348bc0b [15:36] Yay! Oh, wait. [15:40] urbanape: I thought that was fixed, are your apt and sourcedeps up to date? [15:40] ah, probably not, duh. [15:41] how do I manage to keep forgetting that? [15:41] Time for the Lenny tattoos on my forehead. [15:43] thisfred, in desktopcouch/start_local_couchdb.py bump "100" up to "1000", please. [15:48] CardinalFang: hmm, there is only 1000 [15:48] should I bump to 10000? [15:48] thisfred, "100" [15:48] CardinalFang: nm [15:48] CardinalFang: looking at the wrong file [15:53] CardinalFang: ok, now only 2 errors: http://pastebin.com/d7f6829bd but still 2 full couches with 4 /usr/bin/couchdb processes started [15:57] thisfred, have you considered, perhaps, that your computer is haunted by goblins? [15:57] many a time [15:57] There should be an exception for that. [15:58] except GoblinInOpticalDrive [15:58] will be in python 3.0 I believe, but that doesn't help us now [15:59] It's 3.1, and you still have to import from __future__ :( [16:04] facundobatista, Hey. Did my subprocess EINTR patch ever make it anywhere? [16:05] facundobatista, I was trying to avoid the problem by using os.spawnv, but that has the same problem. :( [16:05] CardinalFang: that first error I get, and noone else, it seems, may be a clue that points to where the goblin lives [16:05] CardinalFang: what problem? [16:05] CardinalFang: os.spawn* is deprecated by subprocess [16:06] CardinalFang, what are you talking about? [16:06] CardinalFang, the issue in Python's subprocess? about closing the fd if error? [16:08] facundobatista, no, about read being interrupted. [16:08] * CardinalFang digs for email. [16:09] facundobatista, when trying to figure out what to do with os.spawn, I found your fingerprints. http://bugs.python.org/issue686667 [16:09] CardinalFang: is it when you try to read from STDERR? [16:09] dobey, Not me. [16:09] CardinalFang: well, can you explain the exact issue and point me at the code? i'll tell you what's wrong :) [16:16] jblount, statik: http://emberapp.com/urbanape/images/truncated-long-filenames [16:20] urbanape: holy geez, thats awesome [16:24] that's a pretty simple, straightforward approach. Where we generate the per-row info object, I just truncate the filename if it's longer than (say) 35 characters. [16:25] facundobatista, I can't find the email.. :( Okay, see if this sounds familiar. All system calls in Unix may be interrupted. In subprocess and os.spawn*, we use system calls but do not catch the OSError (e.errno=errno.EINTR) that may happen. [16:25] seem reasonable? I can propose it as is. [16:25] dobey, you too, if you care. [16:26] I've reproduced a few in just a few minutes. [16:26] urbanape: +1 (I'm on the phone with statik, and he likes it too) [16:26] k [16:26] facundobatista, os.spawnv: https://pastebin.canonical.com/21057/ [16:26] facundobatista, subprocess #1: https://pastebin.canonical.com/21058/ [16:26] facundobatista, subprocess #2: https://pastebin.canonical.com/21059/ [16:28] CardinalFang: is this code in trunk? [16:28] facundobatista, Back on June 12, I have you a patch for subprocess that caught these OSErrors and retried. [16:28] dobey, What? Python trunk? [16:29] CardinalFang: no, desktopcouch [16:29] ie, where you're seeing the SIGINT [16:29] huh, that's odd [16:29] dobey, not SIGINT. [16:30] dobey, You can reproduce it with trunk, I'm sure. [16:31] It may take a few minutes of hard running. [16:31] facundobatista, I originally complained that subprocess.Popen was leaking file descriptors. [16:34] grr [16:39] facundobatista, Okay, I tracked it down. These are separate problems. [16:39] The first still persists in my 2.6-karmic, though. [16:39] Today's is a new/separate problem [16:40] * CardinalFang thinks he scared facundobatista off. :( [16:40] * CardinalFang would flee too. [16:57] * jblount tries to sort out lunch [17:20] CardinalFang: after changing tries to 1000, things get weirder still: again everything errors on finding the port, and then a couch is started after the tests are finished running [17:21] CardinalFang: and now I have something like 5 running, with 2 processes each [17:21] CardinalFang: 7 to be exact [17:24] CardinalFang: I think this is the original problematic behavior aquarius tried to fix with one of his recent branches [17:25] * thisfred <- going to the super market in 5 mins === ink_away is now known as inkvizitor68sl [17:30] CardinalFang: I've switched over some tests to testtools.TestCase instead of twisted.trial.unittest.TestCase, since we don't need it, and it's breaking in interesting ways on goblin haunted old computers. [17:32] CardinalFang: all the tests in records, to be exact. [18:05] Is there a known issue atm with logging into the U1 website? I'm getting redirected back to the OpenID page every time I try on three different machines. [18:09] fader_: hrmm, it's working ok for me [18:09] fader_: but i have seen a few bug reports come in about that, this morning. [18:09] dobey: Weird. I wonder if it's something with my account :/ [18:09] I'm getting this on karmic, jaunty, and an OSX machine in Safari. [18:10] fader_: does the same thing happen trying to log in to launchpad.net? [18:10] dobey: Nope, I'm able to log into launchpad on all three systems just fine. [18:11] When I click 'sign in' I get redirected to ubuntuone.com/files and then immediately back to login.launchpad.net [18:13] fader`: We've had a bunch of bugs reported about this same thing recently, so I don't think you are alone. [18:14] jamesh: Yo! Have there been any recentish changes to the SSO system? [18:14] jblount: I'm happy to try anything that will help you here, provide logs, etc. :) [18:14] fader`: i think if you delete your cookies, and empty the cache, it might work around the problem for you :) [18:15] jblount: I can't think of anything that would cause that problem. [18:15] jamesh: That's what I figured. Thanks though. [18:18] dobey: No effect, I'm afraid :( [18:18] Plus I had never logged into U1 from the OSX machine until I tried a few moments ago and I see the same behavior there [18:19] weird [18:21] fader`: I was just able to reproduce, although I'm not sure exactly what is going on yet. [18:22] Whew, I was starting to think U1 didn't like me personally :) [18:22] fader`: I have it on good authority that U1 wants to be your friend. [18:22] :) [18:26] obama said that, now he's turning the country socialist! [18:29] statik: ping? [18:29] * jblount realizes now that ping should probably not have a question mark. [18:31] dobey: you mean the "I was just able to reproduce" bit? [18:33] * Chipaca codes while watching over the coals for an asado [18:33] dobey: https://bugs.edge.launchpad.net/ubuntuone-client/+bug/412360/comments/7 [18:34] Launchpad bug 412360 in ubuntuone-client "impossible to sign in in web interface" [Undecided,Confirmed] [18:35] Chipaca: huh? [18:35] fader`: Are you a recent subscriber? [18:35] jblount: No... I have been on it since slightly before the karmic UDS [18:35] dobey: jblount said "I was just able to reproduce, [...]" [18:35] dobey: and you said "obama said that, now he's turning the country socialist!" [18:36] Chipaca: you missed an intermittent line from jblount [18:36] dobey: it's more fun this way [18:36] Chipaca: the one where he said "i have it on good authority that u1 wants to be your friend" [18:37] fader`: Hmm. Thanks for the info, although it pokes a hole in my first theory. [18:38] fader`: I'll update the bug I linked to a moment ago with info, I'm guessing that jdo will be able to help us figure it out. [18:38] jblount: Last night I purged and reinstalled the client on my karmic machine if you think that's relevant, though I was able to log into the web after doing so [18:38] fader`: the desktop client is irrelevant to logging in on the web :) [18:39] fader`: What dobey said, outside of a Oauth authentication process, the desktop client and the authentication bit on the website don't interact. [18:39] jblount: Please feel free to /msg me if you need me to do anything to help out even if I'm not in this channel. I'm generally on 0800-1700 EDT (which I think is 0400-1300 GMT) [18:39] fader`: Will do, thanks again! [18:39] jblount: Thank you! [18:42] READY_WAITING_WITH_CONTQ --[SYS_META_QUEUE_WAITING]--> READY_WAITING_WITH_METAQ [18:43] gotcha!! [18:43] * Chipaca dances [18:45] * jblount dances because Chipaca is dancing, but doesn't really understand [18:45] jblount: facundobatista found a weird bug where syncdaemon thought it had nothing to do, with 1k things queued in its content queue [18:45] ^ that transition is responsible [18:46] it dropped the _WITH_CONTQ [18:46] it should be [18:46] READY_WAITING_WITH_CONTQ --[SYS_META_QUEUE_WAITING]--> READY_WAITING_WITH_BOTHQ [18:46] :-D [18:46] :D [18:48] some day I'm going to build up (a) my courage, (b) time, (c) lucio's courage, and I'm going to refactor that dumb fsm into something smarter [18:49] (d) tests for each darn transition, before the refactor [18:51] Chipaca: This sounds like a very good plan, you should go swim with sharks or similar to build up your courage. [18:51] Although I hear parenting is sometimes just as dangerous. [18:52] * Chipaca suscribes to the school of extreme parenting [18:52] heh [18:53] urbanape: Any luck landing your branch, or are the tests still running locally? [18:53] I updated my system and had to rebootie [18:54] urbanape: Can I suggest just re-trying pqm-submit? I had to do the same on an earlier branch, and I'm guessing you won't be able to recreate the test failure locally. [18:54] sure [18:54] Nice, thanks. [18:55] resbumitting [18:55] spelled correctly, even [18:56] spelling, smelling. [18:57] fader`: Yo, can you try going to https://ubuntuone.com/files/old ? [18:57] jblount: That seems to load just fine. [18:58] fader`: I thought so, we'll have the rest of the site fixed shortly. Just forgot to remove a decorator when switching over to the new files interface. [18:59] fader`: Thanks again for helping troubleshoot. [18:59] jblount: Thank you! Off to share files. :) [19:20] facundobatista, I made a patch for Python trunk, and two problem descriptions. [19:20] facundobatista, http://sandbox.chad.org/python-subprocess/ [19:21] mmm.... i know that it is offtopic... but maybe any1 know how i can set port of IRC server, when connecting to it from jabber client? [19:21] facundobatista, You did indeed get my June patch in for the pipe leaking fix. [19:22] facundobatista, Thank you. [19:22] CardinalFang, this is another stuff? [19:23] facundobatista, Yes. Two different, but related problems. [19:23] CardinalFang, you should open bugs in bugs.python.org about this [19:23] Rgr. Never done that. I will try. [19:24] inkvizitor68sl: I have no idea about that, what client are you using? [19:25] jblount, gajim... for example i connecting to #ubuntu-ru%zsh.su@irc.zsh.su [19:25] but irc server at zsh.su on 6668 port [19:27] just i have local ZNC and want use it from Jabber >_> [19:28] I think generally you do something like zsh.su:6668, but am completely unfamiliar with the client. [19:32] jblount, i tired already) "wrong symbols in conference name" [19:32] tried* [19:32] facundobatista, Attached to existing bug. http://bugs.python.org/issue1068268 [19:44] jblount: success [19:44] show-me-filenames merged cleanly this time [19:45] urbanape: huzzah, one step closer. [19:57] will desktopcouch work with 9.04? [19:57] haven't tried it [20:03] CardinalFang: ^^ ? [20:07] wow i totally broke the web UI last night [20:08] jblount, Eh? 9.04? It should, urbanape. [20:09] statik: yes you did! :P [20:09] would he need to add karmic universal? [20:09] urbanape: shouldn't [20:09] statik: how so? (beyond flipping on the new UI)? [20:09] urbanape: i don't know if we build jaunty desktopcouch packages currently, though [20:10] urbanape: i forgot to remove the decorator that only allowed people in ~ubuntuone-hackers to access it [20:10] d'oh [20:10] I was thinking, "Works for me!" [20:10] haha, no woder i could log in [20:11] yeah, i stayed up late last night to test it after the rollout, and tested it twice today trying to figure out the bug reports that were coming in [20:11] it was working great for me [20:11] oh, man. [20:11] Maybe that decorator should emit something about the restriction. [20:12] CardinalFang: +1 [20:12] CardinalFang: i presumed it would just 404 [20:13] yeah, it was a weird combination. most places we use it we return 404 instead of 401 [20:13] CardinalFang: but i didn't realize it was there for /files/ :P [20:13] but the other redirection stuff that we have in place to send anyone who logs in to /files/ interacted badly with that [20:13] right [20:13] well at least i was right in suspecting a probable issue with the new files ui :) [20:14] urbanape: you need python-desktopcouch packages for jaunty? [20:14] I got pinged on twitter from a firefox dev who knows extensions and places and wanted to try it out [20:14] * jblount sighs as he sees jdo's branch fail pqm the first time (just as urbanape and my branches did before it) [20:15] but he installed jaunty [20:15] statik: Why does pqm hate new code? Is it my fault? I can change! [20:18] statik, do any of us own python-couchdb ? I filed a bug last night, and I think I can fix it pretty easily. [20:19] Er, maybe not. [20:19] Hrm. [20:20] CardinalFang: nope. for a bugfix that needs to go in quickly, the process is to patch the package, and also forward the patch to the upstream project (either the bugtracker or the mailing list). Filing a bug on the python-couchdb package in ubuntu, attaching a debdiff, and getting a sponsor to upload it is the fastest way to get a fix pushed out [20:27] statik, speaking of debdiff :) do you know if testresources ever got updated for karmic? [20:27] as happy as i am to have a working video card in my laptop now, i sort of wish i had one day to enjoy it before karmic updates broke my video [20:28] jdo: this page says no: http://packages.ubuntu.com/karmic/python-testresources [20:29] if anyone on the U1 team had earned MOTU rights by now, we could upload it ourselves ;) [20:29] statik: you should poke james_w or someone to look at your bug real quick :) [20:30] statik, if none of the U1 team has MOTU rights, I would be the last one to get it [21:15] fader`: Yo! We just got that fix rolled out, the web interface should be fully functional for you now :) [21:16] jblount: Awesome, thanks! You guys rock! :) [21:17] fader`: Thank you, we wouldn't have been able to track down the problem without your help. [21:18] No problem, complaining is what I do best. [21:18] ;) [21:23] facundobatista, CardinalFang: thank you both for fixing python bugs upstream, that makes life better for everyone === fader` is now known as fader|away [21:51] statik, I got a big monitor, too. It's almost oppressive. "MORE CODE", it demands. [21:53] CardinalFang: but what dpi is it? [21:57] CardinalFang: What sort did you get, I remember that it is a 30" behemoth [22:08] it must be one of those giant low resolution things :P [22:10] urbanape: the version of desktopcouch that was in karmic has now been uploaded to https://launchpad.net/~ubuntuone/+archive/beta for your firefox dev to play with [22:10] rock [22:18] * thisfred is developing a serious rootbeer habit [22:18] hey, it said beer, and by the time I discovered it had no alcohol at all, I was well and properly hooked... [22:18] joshuahoover: dunno if you saw earlier, but your GPG key is all set in pqm now so you can resubmit at will [22:24] thisfred: there is good rootbeer, and then there's a lot of crappy rootbeer [22:25] dobey: I probably haven't even gotten to the really good stuff yet [22:25] in the Netherlands I had to go to a specialty asian import supermarket to get any [22:26] heh [22:28] rmcbride: the Iced Earth cover of Number of the Beast is really good too [22:30] thisfred: surprisingly, the organic/cane-sugar root beer that i can get at the grocer near me is pretty damn good [22:31] dobey: well, any kind without H-F corn syrup is an improvement. I think they have some at our supermarket too. This time I just fell for the 12 can box set :) [22:32] thisfred: i just love the corn syrup advocacy commercials they have now. "It's ok, in moderation." [22:33] of course, almost everything that uses it, has it as the first or second item in the ingredients list on the label... which is sorted by quantity [22:33] It's even in some of the chips I discovered to my dismay [22:33] and the bread [22:33] yeah [22:34] i'm just like "wtf do you need that in Doritos for... the fact that they're corn chips isn't enough?" [22:34] f'n lobbyists [22:34] seriously [22:35] shame they couldn't really get the cars to run on it efficiently, that may have been a better way to get rid of oversubsidized unneeded corn. Other than getting it to starving people of course, but that's crazy talk. [22:36] Also loving Chipotle-anything. Not a flavor we had in NL. Nor *any* real mexican food really. [22:36] cars do run on it efficiently [22:37] well not efficiently enough to make up for the cost of extraction/production [22:37] I thought [22:37] they can, yes [22:37] but not efficiently to warrant the gov't pushing for real change [22:38] because why push for changes when you're a shareholder in the oil industry? [22:38] Yea avoiding HF corn syrup is something I've gotten quite adept at [22:40] dobey, only 100DPI. [22:40] thisfred: the other major issue is that pure ethanol is 100 proof liquor, which sort of introduces some odd legal things [22:40] err [22:40] 200 proof i mean [22:40] so there's a bit of required dilution [22:41] but i'm tempted to set up a still in the garden :P [22:41] hehe [22:42] jblount, Dell 3007WFP-HC [22:44] G'night, all. [22:45] later CardinalFang [22:48] woah, there's a funky aquamarine hue to the storm clouds [22:48] dobey: that means hail [22:48] light refracting through airborne ice [22:49] yeah [22:49] though it's not precipitating yet [22:49] clouds are just sort of hovering [22:50] that means they're full of eels [22:50] indeed [22:51] and now more rain [22:51] guess i probably won't get to see the perseids tonight [22:58] alright, i am outta here for now [22:58] later! [23:39] Hello people, Ubuntu One uses something like delta to send only changes to the server ? [23:47] diogo: nope === Chipaca_ is now known as Chipaca [23:48] diogo: not yet; for now, it sends whole files [23:48] Chipaca That send hole file to the server ? If Yes, do you will some day use changesets or something like that ? [23:48] diogo: in fact, the client as it stands now doesn't even resume uploads if they are interrupted [23:48] nor downloads [23:48] Chipaca I was using dropbox, however I'm so excited with Ubuntu One and I contracted the service [23:49] diogo: the idea is to at some point use deltas, yes [23:49] diogo: the plan was (a) make it work, (b) make it fast [23:49] we're still in (a) [23:49] and deltas is definitely in (b) :) [23:49] Chipaca I'm believing in the project and I will keep using it [23:50] diogo: the client is still buggy; usable, but don't use it for critical things [23:50] diogo: we *mean* the "beta" part [23:50] diogo: I love that you believe in the project :) so do I :-D === inkvizitor68sl is now known as ink_away [23:51] I'm using the last ppa and for now I will keep dropbox and ubuntu one until a stable release [23:52] diogo: mind you, we're not just another dropbox wannabe :) [23:52] diogo: file sharing is only one vector [23:53] Chipaca hehehe, that is good