sbalneav | Morning all | 14:20 |
---|---|---|
sbalneav | I am so freamking mad I could spit. | 14:21 |
sbalneav | So, after spending MONTHS getting sabayon to work, and testing compiles in lucid, to make sure it works THERE as well, I did a push (with stgrabers help) up to lucid. | 14:22 |
sbalneav | sabayon works great, if started with sudo | 14:22 |
alkisg | sbalneav: relax man... what's troubling you? | 14:22 |
sbalneav | but now, gksu's broken!!! | 14:23 |
sbalneav | Whaaaaaa | 14:23 |
sbalneav | I can't catch a break :) | 14:23 |
alkisg | So if you start it with gksu it doesn't work? | 14:23 |
sbalneav | Yeah, under karmic, you can start sabayon fine under gksu, and under BOTH you can start sabayon with sudo, if you xhost+ before you start. | 14:24 |
sbalneav | yeah, doesn't work on lucid under gksu | 14:24 |
alkisg | Policykit1 issues, I imagine? | 14:24 |
alkisg | !info gpaint | 14:29 |
ubottu | gpaint (source: gpaint): GNU Paint - a small, easy to use paint program for GNOME. In component main, is optional. Version 0.3.3-3ubuntu1 (karmic), package size 114 kB, installed size 772 kB | 14:29 |
alkisg | !info gpaint lucid | 14:29 |
ubottu | gpaint (source: gpaint): GNU Paint - a small, easy to use paint program for GNOME. In component main, is optional. Version 0.3.3-3ubuntu1 (lucid), package size 114 kB, installed size 772 kB | 14:29 |
sbalneav | alkisg: not sure :( | 14:46 |
sbalneav | One of the things I used to like about Linux/Unix was how it was simple and made sense. Now with DeviceKit/PolicyKit/D-Bus, etc, it's getting... not simple. | 14:48 |
sbalneav | I'll have to fix it, I assume. | 14:48 |
stgraber | sbalneav: did you try with gksudo ? | 15:09 |
sbalneav | yeah | 15:11 |
sbalneav | both exhibit the same problem. | 15:11 |
sbalneav | gksu itself hasn't changed versions from karmic, but libgksu has | 15:11 |
sbalneav | so, obviously, something's gotten "mucked up" | 15:11 |
stgraber | sbalneav: what exactly is failing ? | 15:33 |
sbalneav | it fails with an assertion error that some string shouldn't be null | 15:35 |
sbalneav | By the time I got it last night, I was too p.o'd to dig into it. | 15:36 |
sbalneav | I'll have to figure out what's failing tonight. | 15:36 |
stgraber | sbalneav: looking at LP, it's not the only software that's this issue | 15:37 |
sbalneav | really? | 15:37 |
sbalneav | links? | 15:37 |
sbalneav | brb | 15:37 |
stgraber | sbalneav: jockey had the same issue | 15:37 |
alkisg | gksu output: http://paste.ubuntu.com/348614 | 15:39 |
sbalneav | yeah, that's it. | 15:44 |
sbalneav | how did jockey solve it, or haven't they yet? | 15:44 |
stgraber | sbalneav: looks like jockey had the issue because they used policykit but not completely correctly | 15:44 |
stgraber | I suspect it's not sabayon's case ? | 15:44 |
sbalneav | sabayon doesn't use policykit at all. | 15:46 |
sbalneav | I shouldn't NEED to use policykit: I'm using gksu | 15:47 |
alkisg | sbalneav: gksu gksu ls ==> has the same problem | 15:51 |
alkisg | So gksu doesn't work with gksu... | 15:51 |
alkisg | So if you want, make an "if I'm root, don't call gksu" | 15:52 |
stgraber | alkisg: well, it's failling as regular user too | 15:57 |
alkisg | stgraber: what is failing? | 15:57 |
stgraber | stgraber@castiana:~$ gksu sabayon | 15:58 |
stgraber | MainThread 2009/12/29 10:58:13.5550 (admin-tool): Creating profiles dialog | 15:58 |
stgraber | (gksu:27662): GLib-CRITICAL **: g_str_has_prefix: assertion `str != NULL' failed | 15:58 |
alkisg | Right, I mean this: | 15:58 |
alkisg | sabayon must have a gksu call inside its code, | 15:58 |
alkisg | so if you also run it with gksu, it fails | 15:58 |
stgraber | it doesn't | 15:59 |
stgraber | stgraber@castiana:~$ grep -r gksu /usr/share/pyshared/sabayon/ | 15:59 |
stgraber | stgraber@castiana:~$ grep -r gksu /usr/bin/sabayon | 15:59 |
sbalneav | No, it never sudo's within the code | 15:59 |
stgraber | it's just checking if you're root or not in the code, if you are it shows an error message | 15:59 |
sbalneav | it just checks to see if it's running as root | 15:59 |
sbalneav | it's a pythin script, and there's no way to make setuid python scripts | 16:00 |
sbalneav | that's why we need gksu | 16:00 |
sbalneav | and gksu SHOULD work. | 16:00 |
sbalneav | I don't buy into this whole "YOU SHOULD REWRITE YOUR ENTIRE APP TO USE POLCYKIT" philosophy | 16:01 |
sbalneav | especially seeing as policy kit is horribly horribly broken for remote X displays. | 16:01 |
sbalneav | see: can't run the user add window over a thin client. | 16:01 |
sbalneav | alkisg: Good point on gpaint. | 16:05 |
sbalneav | I'd support dropping it for kolourpaint. | 16:06 |
alkisg | sbalneav: thanks, well, it's nothing important, it's just package selection, but why include a broken app when there's a much, much better one available? | 16:07 |
alkisg | Anyway. Shouldn't there be an /etc/pam.d/sabayon file? I don't see it in Lucid... | 16:07 |
sbalneav | no, it's not really needed. | 16:07 |
sbalneav | I'm not sure why the heck sabayon includes one, seeing as how sabayon never talks to pam | 16:08 |
sbalneav | I think it may be an artifact. Let me check the code to be sure | 16:08 |
sbalneav | yeah, it was from 2005 when sabayon was using consolehelper | 16:10 |
sbalneav | so the sabayon.console and sabayon.pam files can actually disappear | 16:11 |
alkisg | sbalneav: btw, `gksu non-existant-app` has the same problem, and `gksu -w sabayon` works fine. | 16:31 |
sbalneav | -w? | 16:34 |
* sbalneav mans | 16:34 | |
sbalneav | ah, force it to use su, and not sudo | 16:35 |
sbalneav | ah, only problem with THAT is, you need a root password :( | 16:44 |
sbalneav | so that's not gonna fly. | 16:44 |
sbalneav | Whoohoo! | 16:44 |
sbalneav | Gotta love the gnome admins, one of them just did a redir for me so that projects.gnome.org/sabayon now points to live.gnome.org/Sabayo | 16:45 |
sbalneav | so the old, horridly out of date website's gone | 16:45 |
sbalneav | Sigh. It would be nice if there was a "policykit migration guide" out there. The docs on freedesktop.org talk about what policykit IS and how to use it, but give no good help on how to actually think about porting an existing app to use it. | 17:31 |
sbalneav | and googleing about doesn't seem to be turning up anything obvious for me. | 17:32 |
sbalneav | Tried "PolicyKit Migration" and "PolicyKit Porting", etc. | 17:32 |
sbalneav | So very very frustrating. | 17:37 |
stgraber | meeting in 45min | 18:17 |
stgraber | highvoltage: do we have an agenda written somewhere ? | 18:17 |
stgraber | highvoltage: for this one, I'd propose we follow-up on the spec work that has been done, maybe sbalneav will want to say a few words about sabayon and we may have to discuss the bug/wiki/doc days ? (not sure) | 18:18 |
highvoltage | stgraber: yep | 18:19 |
highvoltage | stgraber: I'll update the agenda | 18:19 |
highvoltage | stgraber: from my side I'd feel good if we could decide more or less what we'll do on the bug day so that we can announce it | 18:20 |
stgraber | right | 18:21 |
highvoltage | Edubuntu meeting in #ubuntu-meeting in ~10mins | 18:51 |
Lns | woah! | 18:56 |
Lns | i'm gonna make a meeting?! | 18:56 |
stgraber | Lns: heeh | 18:57 |
* Lns is not up on the current meeting schedule | 18:58 | |
Lns | i still have fridays on my cal | 18:58 |
HedgeMage | lol | 18:59 |
highvoltage | (meeting in #ubuntu-meeting) | 19:03 |
alkisg | sbalneav: ^^ | 19:04 |
sbalneav | back | 20:15 |
moldy | hi | 20:15 |
highvoltage | moldy: stgraber and myself can upload to the archives | 20:16 |
* Lns gets some coffee from the break room and walks in | 20:16 | |
highvoltage | moldy: currently, there's an archive reorganisation taking place which will allow us to have an edubuntu-dev group that will be able to upload | 20:16 |
moldy | highvoltage: ok, great. some months ago, i took a look at ubuntu python packaging... i hope that i remember enough to create decent ubuntu packages out of our python packages | 20:17 |
alkisg | Isn't there also a new app that packages python applications? What was its name, quickly? Can that be used? | 20:18 |
highvoltage | moldy: well feel free to ask if you ever need any help | 20:18 |
Lns | moldy, http://www.gnomefiles.org/app.php/TcosMonitor - wow, another app! i'll add it to the "other software" page on the tcm wiki | 20:19 |
highvoltage | alkisg: I believe that's just for software that's actually created using quickly | 20:19 |
moldy | highvoltage: thank you | 20:19 |
moldy | https://wiki.ubuntu.com/PackagingGuide/Python i used that guide last time | 20:20 |
alkisg | Lns: also here: http://wiki.tcosproject.org/Utils/TcosMonitor | 20:22 |
alkisg | There's also controlaula but I didn't like it at all | 20:22 |
Lns | hmm, tcosmonitor has vnc support, including 'demo-mode' screen broadcast | 20:23 |
alkisg | That's why I mentioned it, and it's in python, so it might be useful ;) | 20:23 |
nixternal | hey, when we have a set schedule for meetings, pass it on and I will put it on the fridge (if I can figure it out again) | 20:23 |
Lns | alkisg, yeah.. looks like it has a ton of functionality in it..is it stable? have you used it on ltsp? | 20:24 |
alkisg | Nope. | 20:24 |
alkisg | I've just read about it, I haven't ran it | 20:25 |
Lns | tcos seems like a great project, and something that ltsp people could share resources with... | 20:25 |
Lns | for instance it's supposed to have an incredible web config interface | 20:25 |
* alkisg would really prefer to give a hand to upstream italc development... it has the most potential... | 20:26 | |
moldy | i would probably join italc upstream if it wasn't in c++... | 20:26 |
alkisg | tcos can also use italc for vnc transport | 20:26 |
Lns | alkisg, don't you still have the 2 different config options for italc (install in chroot vs. not) ? | 20:26 |
Lns | that in itself sure is confusing to newcomers | 20:27 |
alkisg | Well choice isn't really the problem | 20:27 |
alkisg | Firefox can also be installed as a localapp, we don't blame firefox for that... | 20:27 |
alkisg | italc can also be used in a mixed environment, with ltsp, standalone, and even windows clients | 20:28 |
Lns | but i'd say 90% of the time people will want to be able to fully control clients instead of just a user session, right? and that requires manual chroot configuration and again, last time i checked it was very buggy to get that going | 20:28 |
alkisg | What do you mean with "full control"? | 20:28 |
Lns | as in, you don't have to have a logged-in-user to control the client | 20:28 |
alkisg | Well, you don't | 20:29 |
alkisg | I can power on my clients and I don't have italc installed in the chroot | 20:29 |
Lns | but can you power them down at ldm? | 20:29 |
alkisg | Sure, if I put a tiny patch in italc | 20:29 |
alkisg | That's why I was saying I'd join italc upstream... | 20:30 |
Lns | italc currently has its own communications facilities for that right? as in, not ssh/dbus/etc? | 20:30 |
Lns | ics or what not | 20:30 |
moldy | i havn't used italc myself. it seems to have some nice features. but i regard c++ as a real problem. slow development, many bugs. | 20:30 |
alkisg | Yes, it supports encrypted communication | 20:31 |
alkisg | It'd be nice if it used telepathy, but it wouldn't be cross platform then | 20:31 |
moldy | italc has a service running on the clients to do stuff like locking screens? | 20:32 |
alkisg | Yes | 20:32 |
stgraber | yep | 20:32 |
highvoltage | sbalneav: thanks for adding the times that you'd be able to attend for November 2009 :) | 20:33 |
alkisg | moldy: here's a good screenshot: http://italc.sourceforge.net/screenshots.php?img=italc-1.0.0_2.jpg&label=4 | 20:34 |
highvoltage | sbalneav: unfortunately, because of the Ubuntu Temporal Prime Directive, I'm not allowed to alter meetings that has happened in the past | 20:34 |
alkisg | lol :) | 20:34 |
* Lns can only imagine how slow that screenshot actually ran :) | 20:35 | |
alkisg | What do you mean? | 20:35 |
Lns | 20 concurrent vnc sessions heh | 20:35 |
alkisg | I've used it with 12 sessions with no problems... | 20:35 |
sbalneav | Buh, | 20:35 |
sbalneav | I thought the dates were funny. | 20:36 |
alkisg | Although (in the lab where the broadcasting mode always crashed) plain x11vnc performed much better. | 20:36 |
=== Guest652 is now known as Baby | ||
sbalneav | Does anyone know of a good "how to polkit your application" document? | 20:39 |
alkisg | highvoltage: are we going to ship edubuntu-menus in lucid? We could hide qt3-assistant with those... | 20:39 |
sbalneav | I've googled around for a good tutorial or migration strategies... but to no real avail | 20:40 |
moldy | sbalneav: i polkitted tcm some time ago (reverted to no polkit for now since then) | 20:41 |
moldy | sbalneav: trying to re-google the documents i used... | 20:41 |
moldy | http://hal.freedesktop.org/docs/PolicyKit/ http://techbase.kde.org/Development/Tutorials/PolicyKit/Helper_HowTo | 20:42 |
highvoltage | alkisg: I think so | 20:43 |
alkisg | Nice | 20:43 |
moldy | sbalneav: i had to read lots of the polkit library reference to understand stuff... i don't think there is a really good tutorial... those kde docs helped me, though | 20:44 |
sbalneav | yeah, the freedesktop.org link's good for the tech stuff, but not on a "how to migrade an existing app" | 20:44 |
sbalneav | I'll look at the second. | 20:44 |
alkisg | moldy: it think it should be polkit-1 nowadays... policykit is deprecated | 20:44 |
moldy | alkisg: what's polkit-1? | 20:45 |
alkisg | http://drfav.wordpress.com/2009/12/22/polkit-and-kde-lets-make-the-point-of-the-situation/ | 20:45 |
alkisg | From Karmic and on, ubuntu uses polkit-1, the successor of policykit | 20:45 |
sbalneav | Bleh. yeah, there seems to be no way to "simply convert" an existing application. | 20:45 |
sbalneav | complete. rewrite. of. sabayon. | 20:46 |
moldy | alkisg: ah. well, i guess the general concepts did not change that much, though? | 20:46 |
sbalneav | yeesh | 20:46 |
alkisg | moldy: I don't know, I just read about them over a related bug I experienced... | 20:46 |
alkisg | I haven't done any developement related to them | 20:46 |
sbalneav | 'spose I could just toplevel it. | 20:47 |
sbalneav | instead of executing "sabayon", the whole freaking app becomes one gigantic service. | 20:47 |
sbalneav | "PolicyKit was created exactly to make the whole process easier and more secure" | 20:49 |
sbalneav | AHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHHHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHA | 20:49 |
sbalneav | *snort* | 20:49 |
highvoltage | sbalneav: that's like any project with "simple" or "easy" or "lightweight" in it's name :) | 20:49 |
highvoltage | alkisg: stgraber just informed me that the changes have been pushed so gpaint is now gone | 20:51 |
sbalneav | \o/ | 20:51 |
alkisg | highvoltage: much appreciated ;) | 20:51 |
stgraber | I'm actually waiting for Jordan, I have a small question about the meta packages | 20:51 |
stgraber | the seeds are ok but we need to sync the meta on it and I can't find a bzr branch for it, would like to confirm that we don't have one somewhere | 20:51 |
alkisg | sbalneav: file:///usr/share/doc/policykit-1-doc/html/index.html | 20:53 |
highvoltage | stgraber: since we're on LP issues... shouldn't we have an "edubuntu" project registered in LP? | 20:54 |
highvoltage | stgraber: I think Jordan mentioned it before but I can't remember what the outcome was | 20:54 |
sbalneav | alkisg: that's the same stuff off the freedesktop.org site | 20:57 |
sbalneav | and... unless I'm missing something, I can't see python bindings for polkit | 20:58 |
alkisg | http://hal.freedesktop.org/docs/PolicyKit/ is for the old version | 20:58 |
alkisg | The new version isn't backwards compatible... | 20:58 |
sbalneav | http://hal.freedesktop.org/docs/polkit/ | 21:00 |
sbalneav | that's what I was looking at | 21:00 |
alkisg | OK | 21:00 |
alkisg | I saw the link moldy posted above.. | 21:01 |
sbalneav | the second one moldy came up with is better, but still not perfect. | 21:02 |
moldy | what's perfect in this world? ;) | 21:03 |
sbalneav | oddly enough, when I googled for "PolicyKit tutorial, that doc was nowhere to be found | 21:03 |
moldy | google often misses important stuff, in my experience | 21:04 |
sbalneav | I'm still not seeing any direct polkit python bindings, which means to interact with polkit in sabayon, i'm going to have to talk directly with dbus, which sucks. | 21:04 |
moldy | there are python bindings | 21:04 |
alkisg | google for link:http://hal.freedesktop.org/docs/polkit/ => 8 results | 21:05 |
alkisg | I don't blame google for this.. | 21:05 |
moldy | well, i am not blaming them, i am blaming our culture that imho relies too much on google :) | 21:05 |
moldy | sbalneav: http://pypi.python.org/pypi/pypolkit/0.1 | 21:05 |
sbalneav | aptitude search pypolkit | 21:06 |
moldy | sbalneav: for that "helper" scenario, pypolkit is missing a binding, though. i think i have a patch lying around here that adds that binding... | 21:06 |
sbalneav | no results | 21:06 |
moldy | sbalneav: i don't remember wether it was included in ubuntu or not | 21:06 |
sbalneav | I like how I'm told in #ubuntu-devel I should convert to polkit, when the freaking tools I'd need TO CONVERT TO POLKIT aren't there. | 21:07 |
sbalneav | and johnny poo-pooed me for calling polkit the flavour of the month | 21:08 |
sbalneav | cripes, this is no way to run a railroad. | 21:08 |
moldy | sbalneav: if ubuntu is using polkit-1 or what's it called by now, i guess the python bindings won't work at all, anyway... someone has to write some then. | 21:09 |
sbalneav | ah, so for the new polkit, there's no bindings for polkit-1 | 21:09 |
alkisg | sbalneav: just file a bug for gksu, and leave it there. In the next LTS you can look again to polkitify it if the tools are there... :) | 21:10 |
moldy | it seems so, at least i cannot find any | 21:10 |
sbalneav | that's what I thought. | 21:10 |
sbalneav | alkisg: The problem is, I want to announce to people to test sabayon | 21:14 |
sbalneav | they can't in it's current state. | 21:14 |
alkisg | Sure... "Due to THIS FREAKING GKSU BUG, you need to run sudo sabayon to test it" | 21:15 |
sbalneav | Assminng people see the post. | 21:16 |
sbalneav | I'll just fix gksu tonight, and get stgraber to upload it. | 21:16 |
alkisg | I mean as part of the announcement... But anyway it's strange why e.g. gksu gedit works, and gksu sabayon doesn't. Where' the difference? | 21:16 |
sbalneav | or why gksu gksu ls doesn't | 21:17 |
sbalneav | there's no rhyme nor reason. | 21:17 |
alkisg | Right | 21:17 |
alkisg | Even `gksu unknown-app` has the same problem... | 21:17 |
alkisg | Anyway... has anyone looked at ebox? | 21:18 |
sbalneav | heh, the solution being proffered in ubuntu-devel is "write your own python wrappers and write raw dbus calls to policy kit" | 21:22 |
sbalneav | "no thanks" | 21:22 |
sbalneav | I just find it really amusing that the genral trend in Linux today is: | 21:23 |
sbalneav | 1) break old thing | 21:23 |
sbalneav | 2) Tell you to use new thing | 21:24 |
sbalneav | 3) New thing isn't completely written/tested/debugged yet | 21:24 |
sbalneav | 4) Wonder why things seem to be falling apart | 21:24 |
sbalneav | 5) ??? | 21:24 |
sbalneav | 6) Profit! | 21:24 |
sbalneav | Anyway, kiddies, Go through the package bugs, and see what you want me to fix. Screem looks like it needs some love. | 21:26 |
sbalneav | kdeedu might need some love too. | 21:26 |
Lns | sbalneav, Linux has gone through massive growing pains, imho partly due to canonical/ubuntu mission to "adopt early, fix bugs later" strategy..which has really gotten us far, but in the meantime of fixing all this stuff people are wondering how to find a stable system | 21:27 |
Lns | hopefully things will really settle down and we'll start solidifying our direction with things overall | 21:28 |
Lns | and hopefully by that we've chosen the right packages to adopt early... | 21:28 |
sbalneav | Lns: Well, you and I have been arguing against the "fix bugs later" thing for some time. | 21:29 |
sbalneav | it definitely seems to be getting worse to me in the last year. | 21:29 |
Lns | it's hard going from a 'we'll release when we're ready' model to a 'every 6 months' thing | 21:30 |
Lns | well coming from you, an experienced programmer, that's something for people to pay attention to... | 21:31 |
sbalneav | phht, I'm nobody special. | 21:31 |
Lns | not sure what the solution is besides, "let's stop adopting new underlying technologies and stick with what's been working" | 21:31 |
Lns | you're a helluva programmer to me! I'm in awe of what you (and everyone else in our little community) has done | 21:31 |
sbalneav | I'm a problem solver. I go down to brazil for fisl all the time. I see the underpriv'd kids there, and what they have to work with. | 21:33 |
sbalneav | All I want to do is help these people. | 21:33 |
sbalneav | I want kids to have the tools so that they can learn. | 21:33 |
sbalneav | that's all. | 21:33 |
Lns | =) | 21:34 |
sbalneav | And it's frustrating to me that I have to spend time re-fixing problems that have already been fixed/solved in the past, rather than making something better. | 21:34 |
sbalneav | I don't mind hard work, but nothing p*sses me off more than running in place. | 21:35 |
Lns | yeah...i have the same feeling fixing pwned windows boxen constantly | 21:35 |
Lns | hate to hear even a similar situation in the gnu world | 21:35 |
sbalneav | and that's what I find annoying. Because every second I spend re-fixing something that used to work is cheating thousands of underpriv'd kids out of my time. | 21:36 |
Lns | well hopefully we've hit a good spot with all this stuff being ripped out from underneath us. I mean, how much can we really keep changing that's going to cause coders to pull their hair out and re-implement stuff? | 21:37 |
Lns | unless the m$ way of "well you were employee of the month so you get to have your say in how xyz works this release" really is the way people want to keep heading in | 21:38 |
Lns | i watched a good interview w/shuttleworth the other day which seemed to kind of calm my frustrations..lemme find the link | 21:39 |
Lns | http://www.youtube.com/watch?v=9abjUi_H0LY | 21:39 |
Lns | comment posted on that vid: "I'm really hoping the Ubuntu community, beginning with 10.04LTS, will start locking down what new underlying methodologies are used...there have been a lot of growing pains with things like PA, polkit and others that many developers have had to re-structure their existing applications. Hopefully the right things have been adopted, so we can build upon a very solid base for many years to come without worrying th | 21:59 |
Lns | at our efforts will be squashed by another carpet-yank underneath us.." | 21:59 |
alkisg | Lns: what client are you using? it nicely cutted the message in half, while pidgin just drops the rest of the message... :) | 22:01 |
alkisg | Is that xchat? | 22:01 |
Lns | yeah | 22:01 |
Lns | the 2nd msg: at our efforts will be squashed by another carpet-yank underneath us.." | 22:02 |
LaserJock | hi all | 22:03 |
Lns | hey LaserJock =) | 22:04 |
alkisg | Hi LaserJock! | 22:05 |
LaserJock | sbalneav, highvoltage: you guys around? | 22:05 |
sbalneav | hey LaserJock! | 22:07 |
LaserJock | so I was reading my email | 22:07 |
LaserJock | and thinking a tiny bit while stuck at the Registry of Motor Vehicles | 22:08 |
LaserJock | it seems to me that the structure of the Edubuntu metapackages is kinda backwards | 22:08 |
LaserJock | where edubuntu-desktop depends on everything else | 22:09 |
LaserJock | it seems like it would be more logical for people to pick the ubuntu-edu-* that they are interested in that in turn deps on some Edubuntu core set | 22:09 |
alkisg | (10:51:38 μμ) stgraber: I'm actually waiting for Jordan, I have a small question about the meta packages | 22:13 |
alkisg | (10:51:58 μμ) stgraber: the seeds are ok but we need to sync the meta on it and I can't find a bzr branch for it, would like to confirm that we don't have one somewhere | 22:13 |
sbalneav | Seems reasonanle | 22:13 |
LaserJock | maybe there should be an edubuntu-core package or something | 22:13 |
LaserJock | what would be ideal is to have ubuntu-edu-* dep on edubuntu-desktop and then have some sort of edubuntu-full that could be used for getting packages on the DVD | 22:14 |
LaserJock | stgraber: we generally don't put the -meta in bzr because it just gets regenerated at every upload | 22:14 |
LaserJock | the problem with the above scheme is that all the build tools are kinda hardwired to use edubuntu-desktop | 22:16 |
alkisg | LaserJock: I tried the current edubuntu daily build, and the autologon feature was broken, so I had to create a new user from the console to be able to login. | 22:18 |
LaserJock | bummer | 22:19 |
alkisg | I haven't tried the ubuntu daily build though | 22:19 |
LaserJock | that sounds like an Ubuntu problem | 22:19 |
alkisg | Is there anything we do different than Ubuntu? If not, we should be ok.. | 22:19 |
LaserJock | we shouldn't do anything different | 22:19 |
LaserJock | the only problem is if the Ubuntu preseed or something get's changed but Edubuntu's doesn't | 22:20 |
alkisg | OK, it should get fixed shortly then.... | 22:20 |
LaserJock | but I don't think that would cause the autologin to break | 22:20 |
pygi | hi | 23:02 |
stgraber | LaserJock: ok, I'll just regenerate and upload then | 23:20 |
stgraber | uploaded the new meta packages | 23:33 |
stgraber | dropped lpia at the same time as it's not longer in ubuntu | 23:34 |
sbalneav | hahahaha | 23:47 |
sbalneav | so gksu-polkit asks for the password on the command line, no gui to ask for the password! | 23:47 |
sbalneav | Hey, wait, I know, we could use ldm as a front end, we've got that prompt parsing code... :) | 23:48 |
stgraber | you mean the one we want to get rid off ? :) | 23:49 |
sbalneav | yeah, that one! | 23:49 |
Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!