[00:31] evand: isn't adding Jockey to Gobuntu going against its purpose? [00:32] nixternal: I think its due to the seed inherritance. Let me check the seeds to be sure. [00:32] AFAIU, it’s suitable for free third-party drivers as well. [00:33] ya, that could lead to very bad publicity [00:34] Well, I think its in the gobuntu seeds proper, as its not in the platform seeds. Pulling gobuntu seeds to have a better look. [00:35] jockey is also supposed to be a general-purpose driver manager now [00:36] but unseeding it shouldn't hurt either [00:36] slangasek: ahhhh, general-purpose is cool...I thought it was just going to be the new 'restricted-manager' [00:37] nixternal: nah, it got renamed because the scope has expanded to include management of free drivers [00:37] groovy [00:37] OTOH, I'm not sure there currently are any free drivers that it would suitably manage for gobuntu [00:38] hi all [00:38] slangasek, how much time do I have to add to the alpha 6 release notes? [00:38] mgunes: around 12 hours, if everything else goes according to plan [00:40] slangasek, thanks, I'll see what I can do [00:40] slangasek: can I add a section about windows integration (likewise-open) even if the package is in universe ? [00:40] slangasek: to do the Alpha6 release notes [00:41] mathiaz: sure; you're looking to get more testing feedback, I suppose? [00:41] slangasek: yes ! [00:41] we've pointed people at packages while they were still in universe before; e.g., ufw, ff-3 [00:42] slangasek: ok - I'll update the Release Notes then [00:42] slangasek: oh - and also a section about iscsi [00:44] * Hobbsee waves [00:48] It's be cool if zul could upload ebox in time to make the release notes too. [00:49] * Hobbsee gives irc +1 === TheMuso__ is now known as TheMuso [00:59] hi, is it possible to add a script to the image which is outside the compressed part and execute it on startup? [00:59] for the live cds [01:00] i would like to execute something before X starts [01:00] Kano: Have you looked at the way casper works? [01:01] well i looked but did not see it [01:02] do you know more about casper [01:02] Kano: For a start, there are the several scripts in scripts/casper-bottom that do a lot of stuff. [01:02] thats too early for me [01:03] i need internet access [01:03] slangasek: I've updated the ReleaseNotes for Alpha6 [01:03] Kano: Then you probably have to modify the squashfs image. [01:04] best would be after networking/network-manager and before x [01:04] also i see no option to stop network-manager from running [01:05] you could maybe delete the startup links for network-manager(-dispatcher) if a nodhcp option is used [01:05] mathiaz: cheers [01:06] Perhaps, but it would need discussion, probably at the next UDS for example. [01:06] uds? [01:07] Ubuntu Developer's Summit. [01:07] when is that [01:07] Kano: I thought you had been around this community for long enough by now to know what that means. [01:08] i am no ubuntu developer, i only use parts currently [01:08] It is in May. [01:13] what are ubiquity-hooks? [01:14] * Hobbsee suggests RTFMing [01:15] or looking at the source [01:15] which manual for casper? [01:17] * Hobbsee thought there only was one... [01:18] do you call your installer ubiquity? [01:22] a really logical name... === amit is now known as amitk [01:25] could somebody remove patch 03* from syslinux? because of that the menu.c32 countdown is always 0 [01:26] http://kanotix.com/files/fix/casper-terminalserver/casper-terminalserver-kubuntu.jpg [01:26] usally 0 is never shown, 1 would be the last [01:27] in that test i boot kubuntu via pxe with a modified pxelinux.cfg/default derived from isolinux.cfg [01:28] if you drop patch 03 it counts down the 10 timeout i set [01:32] * Hobbsee suggests bugs work better than irc, particularly during au day. [01:35] Hobbsee: Not just particularly during au day, but any time. [01:56] <_MMA_> Kano: After watching you over various Ubuntu IRC channels I wish you would start working *with* this community and its processes and not just take. You've been asked many times to file bugs on issues, on various channels. If you cant be bothered to do it, I wouldn't expect people to bother with you. [01:57] _MMA_: i filed already too many bugs [01:57] "too many bugs"? [01:58] most of em are against the kernel [01:58] * cody-somerville pokes asac in the tummy. [01:58] <_MMA_> I would also say being a part of various mailing lists are a good thing. -kernel, -devel, -devel-discuss, -motu [01:59] mailing lists are something what i absolutely hate [02:00] cody-somerville: if it's WRT n-m, it's fixed in -0ubuntu3. [02:00] :) [02:00] <_MMA_> Kano: A sometimes unfortunate necessity. Patches with bugs also go a long way. And in the end, people might not agree with the patch and it doesn't go in. Just the way it goes sometimes. [02:02] i have got already a fixed version. just a i do not create live cds based on ubuntu yet it is a bit difficult to change what i want [02:02] crimsun: um? what's broken with n-m? [02:03] slangasek: uninitialized priv struct in nm-supplicant.c [02:03] to explain how that works with kanotix -> identify bug -> fix bug -> package updated [02:03] crimsun: should I be stopping this CD build I'm running specifically to pull in the n-m update, then? [02:03] no launchpad or something needed [02:03] slangasek: very much so. You want -0ubuntu3 on the CD build. [02:04] grmbl [02:04] * cody-somerville nods. [02:04] slangasek: the effect being that all wireless broke [02:04] and when will -0ubuntu3 be available? I currently only see -0ubuntu1 in the archive [02:04] heh, "published 38 seconds ago" [02:04] ok then [02:05] I was going to say, I saw it hit hardy-changes a few minutes ago. [02:06] * ScottK would have avoided that bug by virtue of having lost the wireless card for his Hardy laptop. [02:07] * cody-somerville is sitting on the floor by his router with a 15" ethernet cable :( [02:09] cody-somerville: lol [02:09] cody-somerville: did you ever think of using a router with dd-wrt or similar in client bridge mode? [02:09] * cody-somerville did not think of that, no. [02:10] works beautifully, no drivers needed for wlan cards, you can connect as many clients as you like to [02:10] for 2-3 pc in the same room even cheaper than wlan cards [02:11] cool. [02:12] i have got wlan cards, but use em only for driver testing [02:18] * cody-somerville really hates the changes to the theme. [02:20] ah yes, the new theme [02:20] black text on dark shiny windeco really isn't a good combination [02:20] cody-somerville: that reminds me; do you want to have a look at what needs to be done to get xubuntu i386 alternate down to size for alpha-6? [02:21] slangasek, Sure [02:22] cody-somerville: a metapackage update would probably help some, I removed from fonts from the platform seed that ArneGoetje says aren't needed; but i386 is currently 29MB oversize, so more is needed [02:22] * cody-somerville nods. [02:23] s/from/some/ # need some rum [02:26] slangasek, I noticed in the manifest for the live cd, there are duplication applications [02:26] slangasek, I think xubuntu-meta needs to be regenerated and uploaded [02:27] cody-somerville: hrrm, you're not MOTU? [02:28] Hobbsee: Would you please ack Bug #198160? [02:28] Launchpad bug 198160 in rails "Please sync package 'rails' 2.0.2-1 from debian sid" [Undecided,New] https://launchpad.net/bugs/198160 [02:28] slangasek, Not yet. [02:28] Note that despite the title. it's a merge. [02:28] hah [02:28] The reporter thought it would be a sync, but we got that worked out. [02:29] cody-somerville: right; well, do you want to have a look at whether any other seed changes are needed? then I can upload xubuntu-meta for you [02:29] ScottK2: looks fine [02:30] OK. Are you going to mark it in LP or do you want me to? [02:30] Hobbsee: ^^ [02:30] Hobbsee, ScottK: knetworkmanager seems to still need a rebuild against libnl1, now that n-m is (theoretically) working with it, do either of you want to check that a no-change rebuild will DTRT? [02:30] Tonio_: or you, if you're around ^^ [02:31] slangasek: Well see my earlier comment about lost the wireless card for the laptop I have Hardy on. [02:31] slangasek, I just got a phone call and I gotta jet now but off the top of my head, example-content can be removed. [02:31] I can check for wired use, but I dunno if that'll help or not. [02:31] slangasek: I can do that on friday [02:31] slangasek: no time for this now... [02:32] ScottK: ah, right; as we've seen, wired-only testing of n-m does not give us good coverage ;) [02:32] slangasek: not easily from here. will probably be tomorrow as well [02:32] enolaptophere [02:32] Tonio_: ok - I'll just leave the package alone for alpha-6 then [02:32] slangasek: no problem :) [02:33] 3:30 am and I have to wake up at 7...... and I'm still not done with the job :/ [02:34] * Hobbsee ponders actually doing some work [02:38] dear 'doze, please stop munging the formatting. if you have to do so, at least be consistent! [03:02] slangasek: Due to the universe/main backport problem (db 4.6 this time) postfix backport in Gutsy also dep waits. For that one I can trivially do a source backport to switch back to db4.3 (what the released postfix uses). How about I do that instead of you messing with republishing it in Universe? [03:12] uhm, hi y'all. =] I'd just like to say thanks for all your work. <3 [03:19] 1/aw === keescook_ is now known as keescook [04:29] wow, quiet [04:52] //c [04:53] Hrm. I just got some weird keyboard behavior, after updating, restarting, and logging back into GNOME. [04:53] Using gedit, after a while when typing text, I started typing characters that appeared to be for another language... [04:54] it's the start of text-input-via-brainwaves. Stop thinking in other languages. [05:03] Hrm something appears to be broken, even on a fresh hardy install, using en_AU.UTF-8 locale, with a US layout. [05:03] * TheMuso updates from a.u.c to be sure. [05:11] TheMuso: Have you been accidentally hitting shift-space? That brings up some scim dialog which then eats a keypress & switches languages. [05:12] One of my ex-workmates used to run SCIM on his machine so he could type Japanese, and the amount of times I'd accidently hit shift-space suprised me a lot. [05:12] Go _away_, scim-bridge! [05:12] Good night all. [05:32] ScottK: sounds fine to me... :) [06:59] Riddell: hummm, you verified network-manager-kde still works before uploading it, right? It's been rebuilt against the new upstream version of libnl now, and I'm in the midst of rerolling alpha-6 images [07:07] slangasek: all fine now for nm? [07:07] -gnome didn't need an update to still work ... why does -kde need it? [07:08] asac: ah, I assume it's fine now, because no one's told me about (or milestoned) any bugs saying it's broken [07:08] asac: -kde didn't need an update to work, it /got/ an update for something else and as a result picked up libnl1 [07:08] slangasek: yeah. we shouldn't need it [07:09] so I have no idea if the old -kde works ok with the new libnl [07:09] slangasek: it should. i don't think it uses libnl on its own (at least i cannot see it in Depends: of my apt-cache) [07:10] anyway, theoretical questions ;) [07:10] it does depend on libnl [07:10] Package: network-manager-kde [07:10] Depends: [...] libnl1-pre6 [07:10] oh really. what a messy thing. it should interact with nm and not libnl [07:11] if its fixed. fine [07:11] Good morning [07:11] "theoretical questions" - these kind of questions are part of why you freeze before milestones... [07:12] good morning [07:12] i ment theoretical because you said that it works now. [07:12] er? [07:12] I have no idea if network-manager-kde works [07:12] I have no idea if it's been tested at all [07:25] slangasek: wfm [07:26] ok [07:28] well, images are posted then - now we just need to get people testing them... :) [07:29] slangasek: where? its likely that i get a new notebook any minute. could use that to test ;) [07:29] asac: Ooh. What's due? [07:29] standard lenovo thing 12'' [07:30] asac: X300?-) [07:32] not yet :( [07:34] dang :) [07:34] moin [07:43] asac: http://iso.qa.ubuntu.com/qatracker/build/all [07:46] Good morning [07:47] hi pitti [07:48] slangasek: waiting for ack> the hal FF === \sh_away is now known as \sh === hunger_t is now known as hunger [08:14] ah, I think I know why my keyboard is all broken for a few days [08:14] * pitti blames scim [08:14] martin 11958 0.0 0.0 0 0 ? Z 08:40 0:00 [scim] [08:14] bah [08:15] pitti: any idea why gnome-power-manager always show 100% (Current rate and Design rate stay at 0 :( ) ? I tried with the new hal, reseted my gnome-power-manager settings and gconf, ... [08:15] my next try will be to test with alpha-6's livecd (downloading now) [08:15] why do we install scim by default now for European languages? [08:16] seb128: ^ that might be one reason why the keyboard switch applet doesn't work any more? [08:16] pitti: I've no clue about scim so not sure [08:16] stgraber: what kernel version? [08:16] ArneGoetje: why do we install scim by default now for European languages? [08:17] slangasek: 2.6.24-11-generic on amd64 (HP Compaq nx7400) [08:18] stgraber: ok, no idea then :) [08:19] I was having g-p-m issues before the last time I restarted; don't know if those were kernel or hal though [08:20] that's weird because it correctly detects manufacturer, technology and the charge rate but is unable to give me the charge status (stuck at 100% all time) [08:29] pitti: this is because of update-alternatives... it is called by scim to set the preferred modules.. and that one automatically enables scim for all_ALL [08:30] pitti: we need to call 'sudo im-switch -z all_ALL -s none' explicitely to disable it. [08:34] Hi all. I was wondering if we are planning to take advantage of the network manager's GSM capabilities (http://blogs.gnome.org/dcbw/2008/03/02/networkmanager-your-gsm-mobile-and-you/) by adding appropriate information to .fdi files. [08:43] ArneGoetje: or we should change the alternative priorities in scim.postinst - update-alternatives is just doing what it's told :-) [08:43] * seb128_ kicks network-manager dropping network on upgrade [08:45] hehe, mine did that while I was connected to the net over VPN :) [08:46] cjwatson: exactly. but if we change the priority to none, scim will never work for non-cjk locales. [08:49] cjwatson: the way it should interact with language-selector is: to disable scim, call 'sudo im-switch -z all_ALL -s none', to enable scim call: 'sudo im-switch -z all_ALL -a' (this sets it to automatic, which will use the priorities as they have been defined. [08:51] cjwatson: if we want to disable scim by default for all_ALL, we need to call 'sudo im-switch -z all_ALL -s none' *after* scim _and_ im-switch have been installed. [08:52] the reason why this behaviour has not been triggered before is simple: im-switch was not installed by default (which disabled scim altogether, not only for non-cjk locales). [08:54] so, suggestion is: add im-switch to Pre-Depends in scim and call im-switch to disable it for all_ALL by default in scim.postinst after the priorities have been set. [08:55] if this solution is acceptable, then I will change the scim package right away. === Sp4rKy_ is now known as Sp4rKy [09:01] ArneGoetje: is it desirable for scim to work for non-CJK locales? [09:02] I suppose if you are in a German locale and writing Chinese [09:02] cjwatson: yes. [09:02] is anyone else having issues with rsyncing the iso's from cdimages.ubuntu.com? I keeps dying on me :( [09:02] it just seems like a bit of an edge case, and caused confusion on #ubuntu-devel within hours of being activated [09:02] I think it should be off by default [09:03] and perhaps enableable in some GUI way [09:03] cjwatson: basically any overseas CJK speakers and CJK language learners would want it to work for their locale. [09:03] cjwatson: that's what language-selector is supposed to do. [09:04] but this feature has not been properly implemented before. It was just not triggered, because im-switch only got installed when adding a CJK language pack. [09:05] ok [09:05] so, if my proposed solution is OK, I will change the scim package that way. === \sh is now known as \sh_away [09:10] so, comments please... I'm waiting :) === \sh_away is now known as \sh [09:22] ArneGoetje: I think the defaul for all_ALL should be to point to "none" in im-switch [09:22] ArneGoetje: I think that is how it was back in ~dapper :) [09:23] ArneGoetje: we then can use language-selector to turn it on. what do you think? [09:23] mvo: but this would be overwritten by scim.postinst, won't it? [09:25] ArneGoetje: hm, wouldn't it be enough to remove the "ua_inst all_ALL scim 0" from the scim postinst ? or modify it to point to "none" instead of scim? === bigon` is now known as bigon [09:26] mvo: no. that would disable scim permanently for all_ALL. it would never work until you set another module with a higher priority for all_ALL [09:27] * ogra doesnt understand, cant you start it on a per desktop/user base ? if so, why not use an /etc/xdg/autostart/ with a scrit wrapper that checks for a gconf key or so so the system wide stuff to globally enble the setup comes from postinst, but the .desktop file does the user related rest [09:28] ArneGoetje: hm, my impression was that "im-switch -z all_ALL -s scim" overwrites that - we would use that in language-selector then to activate scim for all_ALL if the user wants it [09:28] mvo: no. wrong. [09:29] mvo: the command called by language-selector should be 'im-switch -z all_ALL -a'! [09:29] mvo: this switches it into 'auto' mode, means it uses that module with the highest priority in u-a. [09:29] ArneGoetje: but that would mean that we have to leave scim enabled otherwise -a will not pick it up [09:30] (for all users, even those who are not interessted in it) [09:31] mvo: the priorities are set by scim.postinst (scim-bridge 60, scim 50, scim-immodule 0). and this is used when set to 'auto' mode. [09:31] mvo, see my comment above .... let scim be started by an xdg file and check the lang selector value during login if there is nothing requiring scim it wont be started [09:31] mvo: to disable scim, use '-s none'. this forces scim to use 'none' instead of 'auto' [09:33] ArneGoetje: if we do not want to have scim enabled for all_ALL (that is what colin mentioned earlier), then were do we call '-s none'? [09:33] mvo: if you use '-s scim' to enable scim, you will force the 'scim' module instead of 'auto'. means if the user prefers to use scim-bridge for example, it will be overwritten by language-selector... and we don't want that. [09:34] mvo: if you uncheck the checkbox in language-selector, '-s none' should be called. [09:34] ArneGoetje: I think we are talking past each other, the problem we want to solve is how to make all_ALL not enabled by default for everybody, right? [09:34] (even with im-switch installed by default) [09:34] mvo: I have proposed a solution a couple of pages up... [09:35] ArneGoetje: "let language-selector" do it? [09:35] 16:54 < ArneGoetje> so, suggestion is: add im-switch to Pre-Depends in scim and call im-switch to disable it for all_ALL by default in scim.postinst after the priorities have been set. [09:36] aha, I think your wrote that before I logged in, sorry [09:37] this way the default for all_ALL would be '-s none'. if the user want's scim, he can enable the checkbox in language-selector, which would call '-a' to set it to 'auto', which in turn uses the module with the highest priority... [09:38] I have uploaded a new scim package to rookery (~arne/scim/), with these changes. So, if anyone wants to take a look, please go ahead. [09:39] mvo: I have updated my l-s/arne branch last night. did you take a look already? [09:40] making im-switch a pre-depend (its currently a recommends) sounds a bit drastic to me - I would prefer to make it the other way around (default none and explicitely enable it). but its your packages [09:40] ArneGoetje: no, sorry. haven't looked at it yet [09:40] mvo: the im-switch package does not set any defaults. [09:41] right, I was thinking about changing them in scim [09:42] mvo: but to set any defaults in sim.postinst, im-switch must have been installed already, otherwise the 'none' file doesn't exist and the symlink would fail! [09:42] we also need to be sure that no tool runs "im-switch -a", otherwise we would get the (possible) unwanted side-effect of suddently having scim enabled for all users [09:42] mvo: currently no package call '-a' [09:43] ther eis currently a line: ua_inst all_ALL scim 0 [09:43] in the scim.postinst [09:43] it dosn't need im-switch, it just sets up alternative in /etc/X11/xinit/xinput.d/ [09:45] mvo: I suggest you do 'dpkg -L im-switch' and 'dpkg -L scim' and see which files get installed into /etc/X11/xinit/xinput.d/ by which package. I hope this makes it clear once and for all. :) === doko_ is now known as doko [09:47] mvo: further more, im-switch install /etc/X11/Xsession.d/80im-switch , which is needed to eanable the locale specific im-switch settings. Without it scim won't work at all, no matter what is set in u-a. [09:47] ArneGoetje: (pre-depends is only needed if you're doing something in the preinst, or need something to be in place before the package is unpacked; it isn't needed for use in the postinst) [09:47] ArneGoetje: (for the latter, plain depends is just fine) [09:48] cjwatson: ok, will change that... hang on. [09:48] ArneGoetje: it just seems weird to call im-switch explicitly when setting appropriate alternative priorities would have the same effect [09:49] the default should be identical to what you get when everything is in auto mode [09:49] cjwatson: scim doesn't ship the profile 'none'. if it did, he could just u-a on the .postinst [09:51] mvo: does compiz session management work for you? how do you save your session? [09:51] if i'm looking into this correctly. a solution would be to ship 'none' in scim, and remove it from im-switch [09:51] pitti: yes, I had to modify gnome-session-properties to set it to "save session on logout" [09:51] cjwatson: the user can decide which moule he wants to use by default by adjusting the priorities in u-a. The priorities set by scim.postinst is just a suggestion (and in our case the default setting). if we hardcode a specific module as default by givig it a higher priority when enabling the checkbox in l-s, we will setup those users which have tweaked the settings to their liking. [09:51] mvo: ah, using 'gnome-session-save' or the button 'save now' in g-s-props doesn't work? [09:52] mvo: (usually I just want to save it once) [09:52] pitti: that should work too [09:52] pitti: is it not for you? [09:52] hm, I used the button last time [09:52] with no effect [09:52] pitti: you saved it when the new compiz (with session plugin) was on the system already [09:52] ? [09:52] yes, of course [09:52] hmmm [09:53] cjwatson: that's why I suggest we don't touch the priorities for 'auto' mode and just toggle scim on/off by setting im-switch to -a (== on) or -s none (== off) [09:53] mvo:how do I tell if the plugin is enabled? [09:53] pitti: ccsm will tell you, I set it to enabled by default so if you haven't modified your plugins then it should be on [09:53] pitti: its called "session" (surprise :) in ccsm [09:54] * pitti boots laptop and installs ccsm [09:54] ArneGoetje: "the user can decide which moule he wants to use by default by adjusting the priorities in u-a." that's not the way update-alternatives works [09:54] cjwatson: then how should it work? [09:54] ArneGoetje: the package sets priorities, the highest of which is the default; the user may override by picking a single one manually, not by changing the priorities [09:55] ArneGoetje: the priorities should always be set by us to represent what we think should be the default [09:55] ogra: about your suggestion, currently langauge-selector runs as root (yes :( and this is why we can't set per-user settings. something we should fix for hardy+1 [09:55] ArneGoetje: it doesn't make sense to have this complicated system of priorities and then have the package override it at the end [09:55] mvo, i'm planning a spec to get rid of all unused background porcesses for prague [09:56] we waste a ton of ressources with crap we start but the user will never need, that should be fixed imho [09:56] cjwatson: then how would we enable scim by checking the checkbox in l-s? which module should be used by default? and why should we force to run 'im-switch -s $his_preferred_module' every time after he enabled the checkbox in l-s? [09:56] force the user... [09:57] ArneGoetje: checking the checkbox in l-s could select the next one down in priority order that isn't 'none' or 'default', I guess [09:57] certainly the user should not have to run some other command by hand as well as using l-s [09:58] ogra: what are those? [09:58] cjwatson: that's the same bastardized usage of im-switch as I proposed just the other way round. [09:58] ArneGoetje: in a multi-user environment where multiple users may have different locales, we simply can't use im-switch in the postinst sanely [09:58] mvo: ccsm> is that the simple-ccsm package? [09:58] im-switch has to be a per-user thing [09:59] pitti: if you install simple-ccsm it will get pulled in (as a dependency) [09:59] mvo, example: enabling bluetooth even if we dont have the HW etc ... its jut silly, we have udev that can detect hotplugged bluetooth HW so we can switch it on the fly if needed and dont need to start a system service for it all the time [09:59] (if there is no such HW on boot) [09:59] many things could benefit from better automation here [09:59] ogra: yeah, same for some hplip stuff I guess (not sure if this is still running) [10:00] cjwatson: in that case we wouldn'e be able to use l-s either, because that one changes the settings system wide. [10:00] although I guess that using the -z switch to im-switch makes that less of a problem [10:00] if i see the classmate dektop where i actually can get a workable gnome then i also see that 80% og the stuff i have runniong by default is useless in this HW [10:01] and the wasted ram is not in a kilobyte range rather tens of MB [10:01] cjwatson: if im-switch is called in a user environment it will overwrite the system settings anyways. [10:01] which is how it should be [10:01] cjwatson: we are talking about running im-switch as root [10:02] I understand that, yes [10:02] certainly, language-selector is conflating system-wide configuration (installing packages) and user configuration (enabling input method support) [10:02] so eventually it probably ought to use policykit to escalate privileges just for the system-wide bit [10:02] cjwatson: yes, absolutely [10:02] but I doubt that'll happen for hardy now, unless mvo has a trick up his sleeve [10:03] cjwatson: yes. and that enabling input method support uses im-switch as root to set sane defaults. [10:03] mvo: installed, called ccsm, it immediately crashes on me :/ [10:03] which is a mistake, because it means that me playing around with input methods forces them on my wife using the same computer [10:03] cjwatson: its not too much work to port it so that it runs as the user, but there is little time :/ [10:03] pitti: what is the backtrace? [10:04] mvo: "import ccm" -> 'No module named ccm' [10:04] mvo: simple-ccsm, line 31 [10:05] ArneGoetje, btw, do you have any insight how the new hal input methods will influence the scim stuff ? with intrepid we're likely to switch over to that [10:05] pitti: *sigh* I thought that got fixed a couple of days ago, let me check. please install compizconfig-settings-manager [10:05] ok [10:05] ogra: I have no idea about hal and I doubt that sci will support it. [10:05] mvo: it's supposed to be 'import ccsm' by chance? [10:05] ArneGoetje, its likely that xorg will need it [10:06] mvo: hm, no [10:06] pitti: no, the library is provided by compizconfig-settings-manager, simple-ccsm makes use of it, its a missing dependency === tkamppeter__ is now known as tkamppeter [10:06] ArneGoetje, it will switch to no config by default and rely on hal for everything (mice keyboards and layouts) [10:06] mvo: ah, that did it [10:06] ogra: scim is at a higher level [10:06] ArneGoetje, so its likely to massively influence scim [10:06] cjwatson, right [10:06] mvo: I don't see anythign session-like there? [10:06] ogra: switching to hal shouldn't affect scim [10:07] ogra: then it will probably not be possible anymore to use scim. [10:07] ArneGoetje: I don't think that's true - it just changes how xkb is configured [10:07] i think it changes other stuff as well [10:07] * cjwatson puts up a big "DON'T PANIC" sign [10:07] cjwatson: ok, then it should not be an issue indeed [10:07] anyway, nothing for hardy [10:08] i just dont want us to have somjething in the LTS thats probably not supportable later anymore ... would be good to have an idea about it in advance [10:09] ArneGoetje: anyway, don't let the debate about u-a priorities vs. calling im-switch stop you fixing the basic problem - if you feel it has to be by calling im-switch then your call [10:09] pitti: checking [10:12] pitti: I have it here in 0.6.99+git20080228-0ubuntu1 (plugins-main) as libsession.so, is that available in your package too? [10:12] cjwatson: the point is that the priorities are set and modified by any IM package which supports im-switch, not only scim. For example, nabi and gcin also set priorities in u-a. So, IMHO the best way would be to leave the priorities alone and use 'none' as explicit default and 'auto' when the user enables the checkbox in l-s (which is a system wide setting by design). if the user runs im-switch in a console he will overwrite the whole thing [10:14] cjwatson: so, on a multi-user system the best way would be to leave scim disabled in l-s and use the im-switch command on a per user base to set the preferred module or 'auto' if the defaults sound sane to the user. [10:14] mvo: I mean, in simple-ccsm I only see "info", animations, effects, workspaces, and a11y [10:15] mvo: I do have libsession.so, yes [10:15] pitti: aha, sorry. please run "ccsm" instead of "simple-ccsm" [10:16] I re-uploaded the scim package to rookery BTW. [10:16] mvo: ah, it's there and enabled [10:16] mvo: 'save legacy apps' is off, but the plugin is on [10:16] mvo: but gnome-terminal is hardly a legacy app? [10:16] pitti: no, g-t is saved just fine for me [10:18] mvo: trying with the 'save on quit' checkbox now [10:18] pitti: thanks [10:19] erk, WTH was that; the login sound was dramatically bouncing [10:19] mvo: it looks differnetly disturbed, but still not correct [10:20] mvo: the multi-tab terminal had the wrong size, and all terminals have wrong position [10:20] pitti: so the apps are not restored at the right place? [10:20] at least the workspace is correct [10:20] pitti: totally random or just a couple of pixels off? [10:20] couple of pixels [10:20] maybe 30 or 50 [10:21] my terminal with mutt on 3rd workspace was sticking on the screen bottom instead of being at the top [10:21] ArneGoetje: I think this discussion made it clear that language-selector should run as per-user, this would solve most of the problems as I understand it. I will have a look into the code now to see how much work it is [10:21] * mvo considers the current behavior a bug [10:21] mvo: yes [10:21] mvo: how do you install langpacks then? [10:22] make apt-get suexec *hides* [10:22] pitti: hm, I wonder if it somehow confuses the decoration when it calculates the sizes [10:22] mvo: at least for the 'default system locale' and 'enable/disbale complex character input'. language packs and -support need to be instaled by root anyways [10:22] pitti: it will run synaptic as backend with gksu, that should be fine [10:22] mvo: the size of the 'no tabs' terminals was correct; the terminal size with two tabs was totally off [10:24] mvo: maybe two modes for these two settings. User and Admin... Admin sets system defaults. [10:24] <\sh> bah [10:24] <\sh> how can I disable this damn scim module which is running now by default [10:24] \sh: Hey Stephan :) [10:24] \sh, apt-get remove scim [10:25] ArneGoetje: not for hardy, if I port it, it will look exactly the same (UI freeze), it will just setup stuff like input methods per-user instead of system-wide [10:25] or so [10:25] \sh: sudo im-switch -z all_ALL -s none [10:25] better :) [10:25] ArneGoetje: sorry for that, we can do better for hardy+1 [10:25] mvo: yeah [10:26] <\sh> Sebastian: hey...what a surprise :) What gives us the pleasure of your Visit? Not zend-framework, right? :) [10:26] \sh: No :) Just started lurking here after I have been lurking for a couple of months in #ubuntu+1. [10:27] \sh: Maybe I should create eZ Components packages at some point. [10:29] <\sh> Sebastian: That would be great :) Cool that you joined the Ubuntu Force :) [10:29] \sh: I'm still just a user. Although a very happy one. [10:29] <\sh> ArneGoetje: thx :) [10:30] <\sh> ArneGoetje: na that didn't work [10:30] \sh: It did for me. [10:30] \sh: need to re-login [10:31] <\sh> ArneGoetje: ah that's the magic [10:31] <\sh> Sebastian: still in siegburg area? :) [10:31] \sh: Yes. And today I am actually at home. [10:31] \sh: Been in Zurich and Munich the last week and leaving for Montreal tomorrow. [10:32] \sh: And after Montreal I will be near Berlin for a week. [10:32] \sh: Wine works great thanks :) I can do my course stuff now :) [10:32] <\sh> davmor2: good to hear :) [10:33] \sh: You're with Webzooms now? I had a coaching/training gig there last year. [10:33] <\sh> Sebastian: I heard about that :) And the people here were surprised, that I already knew you :) [10:33] \sh: hehe [10:33] Uh, the PHPUnit package is outdated. Maybe I should start by fixing that before getting into new packages. [10:34] <\sh> Sebastian: I'm happy to sponsor :) [10:34] Ah, cool. [10:34] <\sh> Sebastian: just join #ubuntu-motu :) the coolest crew in the world :) [11:06] heya [11:20] Where is hidden messages in gksu dialogs? I can't find which module I should translate [11:28] mjg59, do you remember the suspend to ram issues in loopinstallations (fuse)? Well have 2 users reporting they can now suspend! [11:28] not sure what happened there but, I am very pleased [11:29] I won't be able to test that myself until tonight though [11:29] xivulon, whats the base device under the loop mount ? [11:30] the normal windows HD ? [11:30] guess so, ntfs in both cases, will ask about disk specs [11:30] * ogra_cmpc wonders if amitk already added the usb presistent patch [11:31] xivulon, ah, thats something different than what i am waiting for [11:31] i thought i missed a change === \sh is now known as \sh_away [11:31] what is your issue? [11:31] i have an usb jey as bottom device ... [11:31] *key [11:32] they get automatically powered down by the kernel on suspend [11:32] in my case it was problems suspending when root was in a loopfile on top of an ntfs host (i.e. the host device was mounted using fuse) [11:32] if you resume your device is gone [11:32] looks nasty [11:32] it is [11:33] by the way, I'd assume that hibernation does not work with swap on file, correct? [11:33]