=== mpt_ is now known as mpt [01:23] slangasek: poke mvo if you would; otherwise remind me during the week [01:26] ok [02:33] * calc uploaded openoffice.org 1:2.3.0-1ubuntu5.2 to gutsy-propsed [02:33] er gutsy-proposed [02:34] fixes 11 bugs :) [02:35] calc, Did you file an SRU? [02:37] somerville32: i reused the SRU for 5.1 since it was defective [02:37] or am attemping to reuse it, heh [02:38] calc, Are you a MOTU? [02:38] somerville32: core-dev [02:38] Cool [02:39] to bad pitti isn't awake or he could accept it for me [02:39] slangasek: do you do accept SRUs? [02:39] er remove second do [02:39] hmm looks like just pitti and cjwatson do them according to lp [02:43] https://bugs.launchpad.net/ubuntu/gutsy/+source/openoffice.org/+bug/153132 <- debdiff is there if anyone is interested in what i have fixed for the new upload [02:43] Launchpad bug 153132 in openoffice.org "Openoffice splash screen includes Ubuntu logo" [Medium,In progress] [02:44] i fixed the broffice.org not installing bug, the bug where forms/dictionary/etc don't work without openoffice.org-base installed, the python-uno can't be imported bug, the gtk themes crash OOo bug, and proper bitmap depth for splash screen bug [05:08] I tried building a package build-depending on sun-java6-jdk, and got this error: "sun-dlj-v1-1 license could not be presented" [05:11] Yup, that's right. You get an EULA in the postinst. I forget if people can get around that - you might want to check #ubuntu-motu, too. === LaserJock is now known as LaserRock === stu1 is now known as stub === Shely is now known as MJ086 === Shely is now known as MJ086 [10:09] is someone here? [10:09] i need someone to help me with a give-back [10:12] nxvl: sure, which package? === LucidFox is now known as Ilpalazzo === Ilpalazzo is now known as LucidFox === zasf_ is now known as zasf [10:18] Mithrandir: is uploading possible right now or should I wait till LP is back? [10:18] Wait until Launchpad is back. [10:26] Mithrandir: advi [10:28] nxvl: I asked Hobbsee yesterday to give it back [10:31] geser: oh! thnx :D [10:32] geser: but it's still on the FTBFS list [10:32] how often it refreshed? [10:32] nxvl: That gets updated about once a day, but it's always best to check in LP, as the build queues might be lengthy. [10:33] * Hobbsee waves [10:33] More verbosely, if you're looking at a FTBFS, you'll want to make sure there's not a pending build prior to hunting too deeply. [10:33] build queues are lenghty for i386 [10:33] Morning, Hobbsee. [10:33] Hobbsee: The new -intel seems to be stable. [10:34] I've been running it all day without issues. [10:34] persia: where on LP? at the release information on the package page? [10:34] palmer is broken since 20th [10:34] Hi Hobbsee [10:35] nxvl: check the build status for the package (once LP is back) [10:35] nxvl: Errr.. Something like https://launchpad.net/ubuntu/hardy/+builds and then search for "All states" and the package name. There's a better URL, but I usually find it by trial and error, and LP is being maintained now. [10:36] oh! thnx i didn't knew it exist :D [10:36] Fujitsu: excellent! [10:36] Fujitsu: downloading it now [10:36] persia: i dont think there's much of a better URL [10:37] sun is now shining and don't want to sleep even a little [10:37] :S [10:37] Hobbsee: There is a package specific build URL, which doesn't require a querystring. [10:37] persia: oh, that's a point [10:37] (mind you, I don't think there are any links to that URL anywhere) [10:38] no, of course not. [10:38] if links were there, it'd be easy to find things :) [10:38] https://launchpad.net/ubuntu/+source/advi/1.6.0-13/ shows you the build and their statuses [10:39] Fujitsu: wow, it does seem better. is scrolling better too, or is it just me? [10:55] Hobbsee: Just you, I think. [10:57] Other than scrolling, it is working quite admirably. [10:58] Fujitsu: nice. oh, and the redraw's still slow, and creates artifacts [10:59] Hobbsee: You mean creation of new windows not updating textures in time? === cprov-lunch is now known as cprov [10:59] Or the occasion where bits of the desktop will appear in the middle of the screen? [11:00] the former [11:14] is there any mailling list where i can ask for give-backs or only here? [11:16] nxvl: Usually here is only where its done, and usually only during the week. [11:16] nxvl: As it is, LP is still down I think. [11:17] mm [11:17] ok [11:17] then a geve-back is needed for apcupsd [11:20] TheMuso: it appears to be back [11:20] Hobbsee: ah ok. [11:20] TheMuso: oh, and it's when buildd-admins are around [11:20] which, as you say, is usually during the week. [11:21] Hobbsee: Are you using -synaptics? [11:21] is there any build-admins list out there? [11:22] nxvl: lp.net/~buildd-admins [11:22] iirc [11:22] nxvl: it's ~launchpad-buildd-admins [11:23] nxvl: given back [11:23] Fujitsu: yes [11:25] Hobbsee: Right, just wonderinf if you had a fix for the problem I was seeing. I'd manually added the synaptics stuff to a new xorg.conf, and the USB mouse was going crazy, but copying an extra couple of lines from the old config seems to have fixed that. [11:25] Fujitsu: i've not modified it at all, WFM. [11:26] although i dont have a usb mouse connected [11:26] You've probably got an old config. The new debconf scripts seem to not detect -synaptics. [11:27] probably. === andreasn_ is now known as andreasn [11:28] Hobbsee: Ah right. Goes to show how rarely I have to do it. :p [11:29] TheMuso: :P [11:30] Hobbsee: synaptics was dropped from dexconf because input-hotplug should cover that. too bad that it's not actually used yet [11:31] (commit by debian, fwiw) [11:31] right [11:32] also, you coud try 'Option "AccelMethod" "XAA"' in your Device-section to see if the bugs you are seeing are because of EXA [11:34] tjaalton: I'll try XAA now that I have synaptics properly working. [11:34] Fujitsu: great [11:36] * TheMuso decides to upgrade to hardy on a non-critical box. [11:38] tjaalton: It much better, but still not perfect. [11:38] I can scroll PDFs now! [11:39] +is [11:40] Fujitsu: did your touchpad work at all before you made your changes? [11:41] tjaalton: Yes, but without scrolling or double-tapping. [11:41] Fujitsu: ok, in that case synaptics can wait for input-hotplug ;) [11:42] When I copied the synaptics section verbatim from the previous xorg.conf, events from my USB mouse were all detected twice, but that was fairly easily fixed. [11:42] yep, that was one of the reasons it was dropped [11:43] Ah. === Sikon is now known as LucidFox === \sh_away is now known as \sh === nand`_ is now known as nand` [12:51] tjaalton: ROCK ON! [12:51] tjaalton: using XAA goes back to working fine === \sh is now known as \sh_away === cprov is now known as cprov-away === asac_ is now known as asac === LaserRock is now known as LaserJock [17:47] hehe [17:47] LaserRock :) [17:47] ;-) [17:47] did jorge show you the pic ? [17:48] oh, i just notice he blogged it ... [17:48] * ogra should read planet more regulary :) [17:54] ogra: hehe [18:11] * lamont wonders who retried ebug-http === j_ack_ is now known as j_ack [18:51] hi [19:10] Hi all. Anyone know (roughly) how long it's likely to be until Xorg 7.3 is gonna be in the ubuntu packages? [19:39] Jooles: it's in Hardy already. [19:40] so it's downloadable using the package manager? I haven't played around with ubuntu much you see and I'd rather not compile it from source as I'm setting the machine up for a friend who's not used linux [19:42] it's in the development release [19:42] not in the stable [19:46] cool. Thanks guys = ) === Lure_ is now known as Lure [22:36] Refresh my memory: Is it ok for source packages to contain blobs that are built from free software, but not built from source in that package? Example: The kvm package contains a version of bochsbios with a few tiny patches applied and built by kvm upstream and shipped in their tarball. [22:37] I'm guessing: hell no. [22:37] soren, Maybe in multiverse? [22:38] Sure, but that's not really interesting at all :) [22:38] soren: it's the bochsbios binary? [22:38] LaserJock: Well, it's built from the bochsbios source with a few tiny patches applied. [22:39] LaserJock: So no, it doesn't match a file from the bochsbios package. [22:39] soren: it's a supportability issue, surely? [22:39] slangasek: Indeed. [22:39] slangasek: I've changed the bochs package to build this special version of it, and having qemu depend on that, and I intend to do the same for kvm.. [22:39] soren: well, but is there source for it or no? [22:40] soren: thats the right thing to do [22:40] yeah, I would think so [22:40] I just want to know if it's kosher as per Debian policy, so I know what to put in my bug report to Debian. [22:40] LaserJock: Not in the kvm package. The kvm package has the patch, the bochs package has the rest of the source. [22:41] soren: the key question is 'is it a licence violation' [22:41] LaserJock: Allegedly, if you apply said patch to the bochs source and build it, you get that very blob. [22:41] lifeless: It's not a policy violation? [22:41] soren: that is, if the blob is e.g. a derived binary it will trigger GPL protection [22:41] soren: .jpg's and other things are built, its still an open freaking sore wound [22:41] lifeless: That's an issue, too. Agreed. however, that's not what I'm interested in right now. [22:42] soren: if the blob is derived from bochs but the source for it is not delivered when you apt-get source kvm, then its a violation of the GPL (assuming bochsbios is gpl'd) [22:42] soren: and this takes precendence over policy :) [22:42] I'll rephrase my question: Should I just tell the debian maintainer that I disagree with his approach, or should I pound him in the head with some part of Debian Policy or the dfsg or something? [22:43] oh, if it's the fault of the Debian maintainer, pound him in the head [22:43] soren: FWICT pound him in the head with the GPL. [22:43] lifeless: That too :) [22:43] soren: DFSG says that we have to ship source for all programs in Debian main [22:44] slangasek: Right. [22:45] slangasek: even if that source is split into different packages? [22:45] slangasek: But depending on how you interpret that, you could argue that the source is somewhere in Debian main, so all is good. I wouldn't agree to thtat, though. [22:46] slangasek: ...so to do the right thing in terms of supportability and such, you could just change the build system to not use it, but in terms of complying with policy, would it also be necessary to repack the tarball without any blobs at all? [22:46] er... policy and the gpl.. :) [22:47] soren: hmm, in the past where I've had binary objects in an upstream tarball, I've stripped them out. If the code is GPL, this is preventative license compliance, because in the future you may not be able to reproduce that blob from sources available in the archive [22:48] soren: no, you can't argue that its somewhere in main [22:48] lifeless: Didn't think so. [22:48] soren: because if you follow that argument, 'apt-get source bash' has to download all the source for all of main to match the gpl's distribution at same time requirement [22:50] lifeless: well, in effect, "apt-get source bash && apt-get build-dep bash"... [22:51] headers and whatnot [22:51] slangasek: in this case it's apt-get source kvm && apt-get source bochbios, which is different. [22:51] slangasek: reproducible binaries and used-source are different arguments, but I think we're in basic agreement [22:52] lifeless: biochs is lgpl, by the way. That doesn't change anything in this respect, afaict. [22:52] lifeless: bochs, I mean. [22:52] lifeless: yes [22:52] soren: no, it won't, because we're not linking to it, its the output itself thats being used, [22:53] lifeless slangasek: thanks guys. I'll send an e-mail to the Debian maintainer explaining the errors of his ways :) === mpt_ is now known as mpt