=== lifeless_ is now known as lifeless [09:22] Happy Friday, folks! :-D [10:14] good morning! [10:58] ralsina, morning! [10:58] ralsina, so, what is going one with tarmac? that fsevents branch is bouncing due to mem issues.. [10:58] hi mandel! [10:58] on linux? [10:59] mandel: yes, u1-client tests are using a lot of memory since a ew days ago [10:59] mandel: so branches may bounce for no reason [10:59] mandel: we got it in the release manually [10:59] ralsina, ok.. bummer [10:59] ralsina, got something nice for the next release: https://code.launchpad.net/~mandel/ubuntuone-client/notification-center-support/+merge/122237 [10:59] BTW: high 5 on the release... o/ [11:00] ralsina, o/ [11:00] we may have some subset of that, the root daemon, the macfsevents code daemonized, share links working and the wizard fixed. [11:00] ralsina, I'm right now going to take a look on why the root daemon is not working [11:02] mandel: cool [11:02] ralsina, I need to talk with mmcc about how to add nice icons to the notifications but works :) [11:04] mandel: awesome! [11:04] mandel: I am going on school run and have lots of errands today (banks) so I will not be around a lot [11:04] ralsina, ok, no problem [11:04] mandel: but as usual, I have logs, so keep talking to me ;-) [11:28] good morning, all! [11:44] alecu, morning! [11:44] ralsina, maybe the reason why it looks bad on a retina display is due to the fix for the scrolling.. [11:45] alecu, new mac os x client feature: https://code.launchpad.net/~mandel/ubuntuone-client/notification-center-support/+merge/122237 [11:45] I wanted to get away form the files.. [11:50] mandel: the retina display bit sounds very reasonable. [11:53] mandel: that seems to be the problem, yes: [11:53] http://www.qtcentre.org/threads/50130-Qt-Creator-on-MacBook-Pro-Retina [11:53] and http://comments.gmane.org/gmane.comp.lib.qt.devel/4477 [11:54] alecu, nice [11:55] mandel: we should ask mmcc to include the relevant Info.plist items [11:55] mandel: and then ask zac to test it. [11:55] alecu, yes, is a problem that we don't have the hd to test it.. [11:55] hardware.. not hd :P [11:55] mandel: ask christian to buy one for you! [11:56] alecu, I'm close to my 3 years :) [11:56] mandel: you are closer to two! [11:56] alecu, also, heheh [11:56] alecu, but I'm close than I was :P [11:57] alecu, and feels like longer [12:20] omg, the new chat function in thunderbird is horrible. At least for IRC. Why didn't they concentrate on it eating all available memory, rather than build yet another chat client... [12:22] lunch time [12:22] * mandel lunch [12:26] thisfred: you are actually IRCing from thunderbird? That thing figuratively exploded whenI tried it. [12:27] I was just then, it worked for me, after a fashion, but the UI is crazy [12:27] especially unusable if you use tb halfimized, which all my windows are [12:27] the chat window if four characters wide in that case [12:28] give or take [12:28] thisfred: for the "twitter is too burgueois"crowd. [12:28] ralsina, well it does twitter gtalk and facebook chat [12:28] boy did I misspell that one [12:28] but then, so does everyone [12:29] so you can tweet in only 35 lines. neat. [12:29] hehe [12:29] I'll see if the other accounts work better [12:29] but definitely a no for IRC [12:30] wonder how the tweet app key limit hits thunderbird. They are supposed to have 10M users, and the keys are limited to 200K :) [12:31] maybe that's why they made it look like crap [12:32] oh, gtalk and fbook also totally unusable. [12:32] that was that then [12:34] also firefox 15 seems slower, and broke pentadactyl. Since vimperator already didn't work, I think I'm gonna go back to chrome [12:34] maybe it works great on Mozilla OS. [12:34] not a happy post-UIF day for thisfred [12:34] ralsina, well, if this was on quantal that would be one thing [12:35] this is my precise machine [12:35] thisfred: hahaha [12:35] thisfred: or rather... erm... bummer dude [12:35] :) [12:35] * ralsina ctrl-c's his apt-get [12:36] maybe there's a web browsing and email reading plugin for vim [12:36] but then I might as well switch back to emacs [12:37] just change your .xinitrc to "exec /usr/bin/thunderbird" and you'll be set [12:37] oh, oops past 8:30. Stop complaining/hacking on lp:gouda, start work :) [13:25] alecu: we need to come up with that estimate for joshua [13:25] * ralsina starts putting numbers into a hat [13:26] good morning briancurtin! [13:28] ralsina: I'm nowhere near close to providing a real estimate, so let me add this stack of old bus tickets in your hat too. [13:29] oh, a capicúa! [13:29] alecu: k, let's talk about it in the evening [13:32] hi ralsina [13:36] alecu: "the journey will take many moons to complete" [13:37] lol === teknico_ is now known as teknico [13:45] dobey: http://www.engadget.com/2012/08/30/lgs-ea93-29-inch-display-hands-on/ but I know, too low-res for you :-) [13:48] uh, yes [13:50] however, a 29" 21:9 monitor would look great on my desk === teknico_ is now known as teknico [13:52] it doesn't even have a VESA mount [13:54] if god intended us to use vesa mounts, he wouldn't have created duct tape [13:55] i want this new panasonic ips alpha display === tetris4_ is now known as tetris4 [14:10] ralsina: hrmm; we'll need a freeze exception for SSO perhaps as well (although technically it is a regression) [14:11] dobey: which bug? [14:11] https://bugs.launchpad.net/ubuntu-sso-client/+bug/974637 [14:11] Ubuntu bug 974637 in ubuntu-sso-client (Ubuntu) "Qt Registration and Log-in dialogs have no way to perform the other action" [High,Triaged] [14:12] can't believe i totally forgot about that bug :-/ [14:12] me neither [14:12] ok, let's file the exception then [14:13] I totally dropped the ball on that one [14:13] let me see if we'll need one [14:14] mmcc, I found the bug with the root daemon, what are the rights of the unix domain socket? [14:16] forget about it, there is another issue.. [14:47] howdy [14:47] hi mmcc [14:47] high 5 on release! o/ [14:47] ralsina: \o [14:47] [14:47] mmcc, morning! I'm looking at the root problem, it seems that there is a connection issues and the user is never added to the daemon, therefore the add-watch gets stopped there [14:48] mmcc, can I see the change you made for launchd? I'm trying to reproduce what is going on with some testing scripts [14:48] mandel: ok, do you need me to push my… yeah, ok [14:48] mmcc, also the source of the plist would be nice :) [14:48] mmcc, yes :) [14:49] mandel, why do you need the source of the plist? I have a hunch you won't like it :) [14:50] oh wait, never mind that :) [14:50] mmcc, that is the reason hehehe [14:51] if you're running your own copy of the daemon, then I bet you're getting an error about absolute paths in the socket name [14:51] to see what the plist is in the built daemon, do this: [14:51] otool -s __TEXT __launchd_plist com.ubuntu.one.fsevents | grep "^[0-9a-f]\{8,16\}" | xxd -r [14:52] this for the info plist: otool -s __TEXT __info_plist com.ubuntu.one.fsevents | grep "^[0-9a-f]\{8,16\}" | xxd -r [14:52] mandel, the catch is we're not allowed to pass arguments on the command line to the launchd daemon [14:52] mmcc: we are supposed to have everything needed to build on bzr so the sources for the plist belong there after some cleaning ;-) [14:52] so I had to make a change to make the launchd checkin the default action for the daemon [14:53] ralsina: oh, they will be - this is just handy debug double checking :) they're built into the executable [14:53] mmcc, hmm ok, lets take a look at the code, anyways I think there is an error somewhere else [14:53] mandel: ok I'll go push it asap [14:57] mmcc, mandel, dobey, thisfred, alecu, briancurtin: standup starts in 3' ends in 5' [14:57] aye aye [14:59] so this is *not* a great time to go make coffee, is what you're saying [14:59] thisfred: right [14:59] unless it's literally instant coffee [15:00] me [15:00] me [15:00] me [15:00] me [15:00] me [15:00] me [15:00] mmcc is last, go me! [15:00] DONE: mac release! a couple of quick branches for last-minute issues there, helped around, reviews, team call, 1-1 with alecu, design chat with lisette, dash call with alecu & neil, other calls, TODO: banks, errands, doctor visit, minor mac bugfixes, start on making the share tab work on mac, maybe also make it work better on ubuntu BLOCKED: too many errands NEXT: mandel [15:01] DONE: added support for notification center and growl for the mac client: https://code.launchpad.net/~mandel/ubuntuone-client/notification-center-support/+merge/122237 Look at the root daemon, my test script does not longer work which is bad news. [15:01] TODO: keep working on the root daemon debugging. [15:01] BLOCKED: no [15:01] briancurtin, please [15:01] DONE: reviews, debugging in a twisted path to find where their bytes/str needs to be changed. we're sending bytes, urllib requires bytes, and somewhere in between it's going wrong (rather, multiple places) [15:01] TODO: hopefully figure this out in a workable way and propose to antoine's t3k branch [15:01] NEED REVIEWS: https://code.launchpad.net/~brian.curtin/ubuntuone-dev-tools/py3-bytes-formatting/+merge/121711 and https://code.launchpad.net/~brian.curtin/ubuntuone-dev-tools/py3-dbus-str/+merge/121712 [15:01] NEXT: thisfred [15:01] DONE: basic auth u1db TODO: wrap up <- | move playlists to u1db BLOCKED: no NEXT: alecu [15:01] DONE: misc calls, a clean Q vm, started learning about the unity dash and nux codebase, got stuck running unity from trunk [15:01] TODO: get up to speed with dash devel [15:01] BLOCKED: no [15:01] NEXT: dobey [15:01] DONE: team call, ubuntuone-client-gnome fixes/release [15:01] TODO: reviews, icon generating magic, music store work [15:01] BLCK: None. [15:01] DONE: release alpha 1, testing, bug filing [15:01] TODO: push branches, fix bugs [15:01] BLCK: no [15:01] AWESOME standup team :-) [15:02] finished in 2 minutes too! [15:02] thisfred: go get coffee, the rest, comments? [15:02] hmmm....I need to change my BLOCKED above: [15:02] oops [15:02] BLOCKED: got stuck running unity from trunk [15:02] done [15:02] alecu: ok, you know who to ping on that [15:02] still on :02! [15:02] alecu: yes, no shaming! [15:02] ⁵! [15:04] ok EOM it is [15:05] 👋! [15:06] mandel, is your test script that's failing in the repo or can you paste it somewhere? [15:06] mmcc, is in the repo under tests :) [15:06] dobey, can I have a review for : https://code.launchpad.net/~mandel/ubuntuone-client/notification-center-support/+merge/122237 [15:06] mandel: good spot for it :) [15:07] mmcc, same there ^ and can you let me know where do I add the new dependencies? [15:07] for the buildout I mean [15:07] mmcc, the script you can make it connect to the daemon passing the path with -p and adding a folder with -f [15:07] mmcc, is in the README of the project [15:08] mandel: what are you trying to do with the buildout? didn't get that [15:08] t [15:08] t [15:08] time for a new irc client here [15:08] making me stutter [15:08] mmcc, I added new deps pync and the growl one, I'll like to add them to the buildout [15:10] mandel: how are they packaged? are they eggs that you can get from pypi or something? [15:10] mmcc, yes, you can use pip or easy_install with them [15:11] ok, then you should be able to just put the name of their egg in the list of eggs under the [development] section… [15:11] it's ok to have them in there for everyone I guess, we could make a mac only section but it doesn't hurt windows to download a couple eggs that it won't use [15:15] mmcc, ok, doing a branch for that [15:15] dobey, fixing it right now [15:25] hey, all! [15:25] Just got an SMS from ralsina, saying his connection got chopped off. [15:32] alright, need to get lunch. bbiab [15:42] dobey, branch fixed and images uploaded.. [15:42] * mandel restarts due to Q upgrade [16:04] hey mandel, I just proposed three branches of the daemon project: https://code.launchpad.net/~mikemc/ubuntuone-fsevents-daemon/mungeproj/+merge/122314 , https://code.launchpad.net/~mikemc/ubuntuone-fsevents-daemon/contextfix/+merge/122312 , https://code.launchpad.net/~mikemc/ubuntuone-fsevents-daemon/checkin/+merge/122313 [16:05] mmcc, on them [16:06] mmcc, talking with alecu we found the following: [16:06] mandel: that seems to be the problem, yes: [16:06] http://www.qtcentre.org/threads/50130-Qt-Creator-on-MacBook-Pro-Retina [16:06] and http://comments.gmane.org/gmane.comp.lib.qt.devel/4477 [16:06] mandel: the first MP is kind of big, but hard to split up because of the xcodeproj. [16:07] mandel: yeah I saw that in the backtrace. our info.plist already has NSPrincipalclass == NSApplication. did you see my comment in the bug I filed for that yesterday? Maybe I didn't send that bug to everyone… [16:08] https://bugs.launchpad.net/ubuntuone-control-panel/+bug/1044211 [16:08] Ubuntu bug 1044211 in Ubuntu One Control Panel "Graphics and text blurry on Hi-DPI displays like Retina Macbook" [Medium,New] [16:08] s/backtrace/irc backlog/ [16:08] * mandel reads [16:09] mmcc, yes, that is what I though, we added it so that the scrolling worked, right? [16:16] mandel: yes https://bugs.launchpad.net/bugs/1018918 [16:16] Ubuntu bug 1018918 in Ubuntu One Control Panel trunk "Scrolling shares on darwin does not work" [High,Fix released] [16:17] brb, quick baby watch duty [16:27] ok, EOD for me! [16:32] oh, apparently monday is a holiday in the US [16:47] wtf is up with tests.syncdaemon.test_action_queue.TestZipQueue.test_zip_acquire_lock on quantal? [16:54] mandel: should we wait for the icon for the notifications, or can we specify the icon as None to get no icon? [16:55] dobey, with None it works as I have uploaded in the images [16:55] dobey, so we can go with it and then I'll talk with mmcc on how to locate the icons correctly [16:56] mandel: but i don't see you passing that on to the upstream API at all. it seems that you're just calling the upstream API with no icon arg and those are just the default icons in those situations? [16:56] dobey, exactly, those are the default icons [16:57] mandel: i don't know the growl/nc APIs, but do they let you pass icon=None to them? [16:58] mandel, that merge has the fsevents.IN_ATTRIB change in it [16:58] dobey, the default is none already: https://github.com/kfdm/gntp/blob/master/gntp/notifier.py#L130 [16:59] mandel: but shouldn't your code be passing the icon argument on to the growl or pync API calls? as icon=icon for example? [16:59] dobey, for pync it gets the icon from the calling app [16:59] dobey, no, because sd is passing an icon that it won't find [17:00] mmcc, I must have merged it by accident, but it should have landed in trunk, let me check [17:00] it is? [17:00] dobey, what is up with this: https://code.launchpad.net/~mandel/ubuntuone-client/fix-wrong-fs-action/+merge/122099 [17:02] mandel: that test doesn't like quantal i guess :( [17:02] Love this note from pync homepage: "All the works are available under the MIT license. Except for ‘Terminal.icns’, which is a copy of Apple’s Terminal.app icon and as such is copyright of Apple." [17:02] "but I'm including it anyway because…?" [17:03] dobey, bummer.. [17:03] dobey, sd does not pass the icon and uses the default value.. I can change that if you want [17:03] mmcc, and yet he uses it :P [17:03] mandel: yes please [17:04] ok [17:04] where are the pync api docs? [17:04] oh, this time it failed because of memory [17:06] why the heck are the u1client tests using so much memory now? [17:06] i wonder if some recent change in the last couple months introduced some memory leaking [17:09] omg pync is awful [17:10] why does ubuntuone.status.AbstractNotification exist? [17:11] we never pass a different icon in the calling code, so it'll only ever call pync or growl with the string 'ubuntuone'… [17:12] mmcc: it's the base class for implementing the platform-specific notifications on top of === trijntje__ is now known as trijntje [17:15] "License is MIT except for this file that isn't included and doesn't belong to me, but is in the arbitrary zip file that this code will download and install on your system" [17:15] dobey, I get that, I just wondered why it was necessary. and why it was in a separate module, ubuntuone.status instead of ubuntuone.platform.notification [17:16] mmcc: there didn't used to be ubuntuone.platform.notification; it was all under status [17:16] and really there shouldn't be an ubuntuone.platform in there, but hey [17:17] hrm, yes. hey indeed [17:18] one day i will move everything to be under ubuntuone.syncdaemon in there; one day [17:18] so is there some reason for using an abstract base class for a class with one method (in a dynamic language to boot)? this is an honest question, not a passive-aggressive style complaint :) [17:19] thisfred: ^^ maybe you can give a better answer to mmcc since you wrote that [17:19] I didn't do it [17:19] what? [17:20] mmcc let me think back [17:20] thisfred: don't bother if it's going to interrupt something - not super important [17:20] basically it was sort of an attempt at using interfaces to force the different platforms to conform, I think [17:20] it may have been misguided [17:21] oh, to inject a little static checking into python, avoid seeing runtime type errors due to different apis, that kind of thing? [17:21] or it may have been intended to be expanded at some point to become more useful, and never did [17:21] mmcc, basically yes, but I'm not sure I'd defend it with my life now... [17:22] heh. well, I'll go look at abc some more out of curiosity. doesn't seem to be doing much here though :) [17:22] thanks [17:22] mmcc, it may also have been a case of new hammer syndrome [17:23] I'm off for the weekend, have a good one o. [17:23] o/ [17:23] bye mandel have a great weekend [17:27] hasta luego tío [17:29] mmcc: how familiar are you with the .icns files? i'm looking at the apple hig icon guidelines page, and it has a table with "filename" and "canvas size" but not sure what filename is exactly in this case (ie, key name in a dict or actual file names) [17:30] hrm, pync is a python wrapper around terminal-notifier, which is a .app that just calls NSUserNotification … maybe we should just cut to the chase and find a way to call that ourselves [17:30] dobey, file name is an actual file name [17:31] to generate the .icns file, you put those files in a folder called 'name.iconset' and run 'iconutil -c icns name.iconset' [17:31] and if you have exactly the right number of files and they are exactly the pixel size their names say they are, it will work without any awful cryptic error messages, and create name.icns [17:32] mmcc: and yeah, calling NSUserNotification would be preferred to pync I think. [17:32] so if you're generating icons on a non-mac, the best you can do is generate that .iconset and make the mac build do the last step [17:33] mmcc: there's a tool to build the .icns file on ubuntu too [17:33] but haven't tried it yet, as i wasn't 100% clear on the file naming issue [17:34] dobey: what's that tool called? [17:34] mmcc: package is icnsutils [17:35] and tool is png2icns [17:36] hello again! [17:37] dobey: interesting. curious to see how it handles the @2x stuff [17:38] so I would think that since pync requires terminal-notifier, which includes an Apple copyrighted icon (wtf), that'd be a problem with using it… [17:39] mmcc: i have more of a problem with it downloading a zip file from an arbitrary location and shoving things onto the users machine [17:40] dobey: oh yeah, that would break code signing, we can't do that - but it'll only do that if it can't find terminal-notifier, so if we bundled them together we could do it [17:40] and I guess we could remove the apple icon from the bundle… [17:40] i'm pretty sure apple doesn't allow redistribution of their icons [17:41] i think at that point though we could just use cnames or whatever and call the API ourselves [17:41] it might be time to add PyObjC [17:41] rather than shipping a python wrapper around a ruby script that wraps the API call [17:42] this would be two lines in pyobjc, and then we could get our native file dialogs back and maybe avoid the retina resolution issues all in one fell swoop [17:42] well, scratch that last one, I keep confusing those for some reason [17:43] but we could definitely get the file dialogs back [17:43] mmcc: which is a good reason, considering the non-native file dialog looks like ass [17:43] and not in a good way [17:44] mmcc: so even more reason not to use pync then :) [17:46] ralsina, are you saying it looks badass? [17:47] I'm grabbing pyobjc now. looks like he's still actively updating it, has commits for py3 and 10.8 [17:47] ralsina: btw, who should fix https://bugs.launchpad.net/ubuntu-sso-client/+bug/974637 ? [17:47] Ubuntu bug 974637 in ubuntu-sso-client (Ubuntu) "Qt Registration and Log-in dialogs have no way to perform the other action" [High,Triaged] [17:47] dobey: good question [17:47] dobey: I am guessing I should, since gatox is away [17:50] dobey: I'll put it irst in my queue [17:50] lisettte: ping if you are still around [17:50] did anyone with 10.8 try the alpha release? [17:51] mmcc: I am guessing urbanape did [17:51] mmcc: me [17:51] chaselivingston: did it complain when double-clicked? [17:52] mmcc: I don't think so, let me check. the sd does crash after a while though [17:52] chaselivingston: I'm assuming at some point you saw a warning about an app from an unknown developer [17:52] mmcc: yeah, just had to right click and hit open [17:52] if you already started it, it won't show it again [17:52] or maybe it will [17:52] ok [17:52] ralsina: pong not really but still [17:53] thanks, I guess i do need that official "Dev ID " cert for releases [17:53] lisettte: ok, homework for monday, can you take a look at bug #974637 and see what to do? [17:53] Launchpad bug 974637 in ubuntu-sso-client (Ubuntu Quantal) "Qt Registration and Log-in dialogs have no way to perform the other action" [High,Triaged] https://launchpad.net/bugs/974637 [17:54] lisettte: basically we need a path from "login" to "signup" and viceversa [17:54] lisettte: I am guessing just a link-like text somewhere, but I'd rather not guess too much ;-) [17:55] ralsina: have put it on the list for Monday [17:55] lisettte: awesome, thanks! [17:55] ralsina: but now, i am braindead [17:55] and dying for a beer [17:56] +1 [17:56] beer [17:56] :) [17:57] impossible to afford where you are! [17:57] beer sounds good [17:57] im saving during the week lisettte [18:00] jgdx: i see! :D [18:00] enjoy! [18:00] have a nice weekend! :) [18:00] * jgdx eows [18:00] * lisettte is outta here [18:00] enjoy your beer too dobey! [18:02] you too [18:10] mmmm beer! [18:10] OH A BEAR [18:11] Oh... no, a beer [18:12] There's an awesome new(ish) brewery in B'more called Stillwater. Highly recommended. It's not the usual super hoppy US IPAs, (which I also love) but more balanced European (mostly Belgian) style ales. Highly recommended if you can find them anywhere. [18:12] ralsina, bear hunting is apparently legal too in Maryland. [18:13] speaking of super hoppy IPAs, i got a case of 3floyd's Zombie Dust. not sure if you have it out east, but its really good and i hate IPAs [18:13] thisfred: nice! [18:13] ralsina, well. I hate h [18:13] unting just for sport [18:13] and nobody eats bear meat [18:14] thisfred: you don't hunt bears for sport, you hunt them because it's fair. [18:14] just kidding [18:14] and yes, some people eat bear meat [18:14] but no, killing bears sucks [18:14] It's fine with me if you hunt them with a knife ;) [18:15] thisfred: http://dayures.blogspot.com.ar/2007/01/caprice-food-from-bear.html [18:15] thisfred: guns for show, knives for a pro [18:16] sure, I give the bear a knife, I hunt the (bear with a knife) [18:16] bro's before pro's [18:16] thisfred: bear hunting is legal in certain situations, in all the states along the mountain ranges [18:16] yeah [18:16] besically the "it's coming right for us!" clause [18:17] dobey, I thought that was humans [18:17] works for both [18:18] * thisfred is lobbying congress for the constitutional right to arm bears [18:18] *crickets* [18:19] thisfred: I once read a scifi book called "The right to arm bears" [18:19] thisfred: about a planet of ursine aliens [18:20] thisfred: http://en.wikipedia.org/wiki/The_Right_to_Arm_Bears [18:20] aaaaand… friday! [18:21] pyobjc releases so far behind dev, and dev has no install instructions! this is fun… [18:21] mmcc: sigh [18:21] mmcc: for some reason that or similar things happen to about 50% of our dependencies [18:22] the pyobjc web site says it supports 10.3.9 [18:22] at least we don't have to use the mac dependencies on lucid [18:22] as in OSX 10.3.9??? [18:23] I know he's working on it, but it's one of those deals where he barely has time to do the code, let alone releases, let alone docs, let alone new website since 2009 [18:23] ralsina, that's right, last news was from Nov 09 [18:23] on the site, anyway. mailing list is active enough [18:24] ralsina, there is an awesome R.A. Lafferty story about a bear like alien. I forget what it's called though. [18:25] thisfred: haven't read much lafferty this century [18:26] nor have I, but I have a lot, is due for a reread, I think [18:26] thisfred: could be "Thieving Bear Planet"? [18:26] http://www.isfdb.org/cgi-bin/title.cgi?67528 [18:27] yeah I think it may be [18:27] in which case I have it, let me look [18:27] I am sure it's not "About a Secret Crocodile" because I remember that one :-) [18:30] ralsina, yeah, it's Thieving Bear Planet. In the collection Iron Tears, which I highly recommend. He's one of my favorite authors. [18:30] I think I may have read that at some point [18:30] He also wrote an awesome historical fiction book called Okla Hanali, about the trail of tears. [18:33] thisfred: have I already made you read "Year Zero"? I suspect you are sort of the ideal reader. [18:35] brb [18:37] ralsina, I think I put it on my goodreads to read list after you recommended it there [18:37] or ws that a different book? [18:37] thisfred: I dunno [18:37] thisfred: that may have been A Naked Singularity [18:37] ah [18:38] this is the comic novel about the music industry? [18:38] that's not the one I got in goodreads no [18:39] yeah, the comic novel [18:39] not the dramatic comedy novel about lawyers with swords [18:40] but this one mentions divynils and mike hucknall so it evens up [18:41] anyway, I'm procrastinating, and must stop. [18:41] * ralsina dives into code [18:44] ok, pyobjc shelved for now, waiting for help from dev list on how to build the thing [18:49] mmcc: I suppose ctypes is not enough for that kind of thing [18:50] ralsina: it could be done, and for something really small might be ok, but it'd quickly grow to rewriting pyobjc [18:51] basically everything would be a call to objc_msgsend [18:52] waiting for help on building pyobjc is almost certainly better -- once it's built, it works great! :) [18:53] mmcc: ok, trust you there :-) [19:03]