[00:00] nhandler_, Thanks, I following/learning from the instructions on the ubuntu wiki which only seems to deal with the most basic of packing tasks [00:00] *I'm [00:03] csilk: are you following through the PackagingGuide at https://wiki.ubuntu.com/PackagingGuide/Complete ? [00:03] I just went through it the other day (several times) and found it really helpful [00:04] radix, yes I am, Only just realised I've been looking at the "packaging from scratch" section for so long that I overlooked the fact that all the more advanced details are further down the page [00:04] yeah, it's long :) [00:05] It's like snow blindness, you look at it for so long you can't see it anymore [00:05] ;) [00:05] * csilk takes coffee break [00:06] so, does anyone have advice about why this cdbs-using package just gives me "make: Nothing to be done for `binary'." when I try to build it? It's got autotools.mk and gnome.mk included and this is a standard autotools/gnome project [00:07] radix: stab in the dark, but you may need debhelper.mk as well [00:07] james_w: oh, I've got it, sorry for not mentioning it [00:07] there's debhelper.mk, autotools.mk, utils.mk, and gnome.mk, in that order [00:07] nothing else in rules [00:07] hmm, are you actually building any binary packages? [00:08] i.e. is debian/control sane? [00:08] * radix looks [00:08] Is there any guide available to see what section a package woudl fit into? [00:08] *woudl [00:08] *WOULD [00:08] looks sane [00:08] oh craaap [00:08] james_w: thanks for pointing me in that direction [00:08] it's *not* sane [00:09] it's got "Architecture: i386" ;-) [00:11] csilk: http://www.debian.org/doc/debian-policy/ch-archive.html#s-subsections has a list of all of the sectoins. That might help you choose. [00:11] radix, is there something like a cheat sheet to give me an idea as to waht types of packages go into what section of the repositories? [00:11] * radix points csilk at nhandler_ [00:11] yeah i hit return just as that was posted [00:13] Hmm, I still don't really know which of the sections would be the most apppriate for this package [00:13] I'm tempted to go for misc [00:13] csilk: Are there any other packages in the repositories that are similar to the one you are working on? [00:14] no [00:14] there isnt any other software for linux similar to this [00:14] csilk: if you describe it then others can give their opinions [00:15] csilk: I'm not an expert at this, but I would suggest looking at the types of packages with a section of 'misc' and 'utils'. Those sound like they might be the most appropriate. [00:15] The autor claims it to be: MegaTunix is the only tuning software for Unix. It's an engine tuning utility for measuring variables within a combustion engine such as fuel to air ratio etc etc [00:19] I was just reading up about the Documentation String Freeze (https://wiki.ubuntu.com/DocumentationStringFreeze), what is defined as Documentation? [00:19] nhandler_, it's between electronics and misc [00:21] csilk: It is really your call. You know the application better than us. If a MOTU feels that it should have a different section, they will mention that in their comment on REVU. [00:26] thanks nhandler_ , radix . I'll put it down as misc, people in #debain seem to aggree, and like you said revu can always change it if need be [00:27] You're welcome csilk. Good luck with the packaging. I look forward to seeing it on REVU and in the repositories. [00:28] Thanks, it's my first package. Now that I've got into the flow of managing university work I'm really looking forward to spending more time learning the internals of ubuntu [00:30] Hi! I now it's not an official support channel, but you're the best people to answer my little question. :) I have 7.04 and want to upgrade to 8.04. I know it's not a supported way of upgrading, but is it safe to replace "feisty" -> "hardy" in sources.list and make a dist-upgrade? [00:32] E17: I would suggest upgrading to 7.10 (Gutsy) and then upgrading to 8.04 (Hardy). Modifying the sources.list is not a safe method of upgrading. [00:34] I wanted to get assured. Thanks. :) Will it be necessary in the future to do the same? Many people don't upgrade to every Ubuntu version. [00:36] E17: That is the safest method. You can also simply do a fresh install. [00:40] I don't want to do so. Thus, I will be having Gutsy. :] [00:43] "Many people don't upgrade to every Ubuntu version" - that's fine, but the only upgrade path from 7.04 to later versions is to upgrade 7.10 -> 8.04; if you don't want to have to upgrade every 6 months when the version you're running is end-of-lifed, you might want to do both of those upgrades now... [00:47] slangasek: if you don't want to upload every 6 month, upload every 2 years - LTS -> LTS [00:48] ok, but is it possible to skip versions since 7.10? It's interesting. [00:50] You can only skip releases if you're going LTS->LTS [00:50] ie. 6.06 LTS -> 8.04 -> 10.04 [00:50] right. If you're on 7.04 now, the upgrade path to an LTS is 7.04 -> 7.10 -> 8.04 [00:50] but you can do back-to-back upgrades... [00:51] Do you mean downgrades? [00:53] E17: No, he means upgrade from 7.04 to 7.10 and 7.10 to 8.04 in quick succession. [00:56] But it's still ~600M for 1. upgrade and ~600M for the second. [00:56] It's good I have 2Mbps. About an hour of downloading one "part". [00:56] Right. [00:57] E17: You need to remember, it is a year's worth of changes you are installing. That is why it takes so long. [01:04] Yes, I understand this. However, if it was possible to have in older Ubuntu versions the newest (but less important only) packages (in official repos, -backports maybe), upgrade would be smaller. [01:04] For example, there's newest wine in Hardy, but older in still supported Gutsy. It's OK for important packages, but not so friendly to the other ones. [01:05] I don't moan, just type what I think. I really like Linux, Ubuntu and the Free/Open Source world. [01:11] * E17 says good night / good bye, depending of your country [01:13] if I've got an AUTHORS file like this > http://pastebin.com/m2171c9ab should i just credit the lead/main developer or should i list everone in that AUTHORS file even though some of them didn't do any developement, they did things like donate hardware etrc? [01:13] *everyone [03:39] If I am requesting an upgrade of a package to a new upstream release (not in Debian), I know I need the .diff.gz. Do I also need a debdiff? [03:40] nhandler_: use revu [03:40] nhandler_: revu.ubuntuwire.com [03:41] nhandler_: and attach the link to the bug report [03:41] nhandler_: the whole package is prefered [03:41] nxvl: I've heard mixed things about using REVU for new upstream releases. The package is already in the repositories. Should I use REVU, or should I just upload the files to a Launchpad bug report? [03:42] nhandler_: Attaching the .diff.gz to a bug is the official way. If a potential sponsor wants it on review, you might as well put it there so you get sponsored. [03:42] exit [03:43] Ok, thanks ScottK-laptop. Do I need to provide a debdiff along with the .diff.gz? [03:43] nhandler_: it's a matter of preferences, i just check it if it's in revu :D [03:43] No. It's pretty sensless for a new upstream. [03:45] Anyone got a clue why pbuilder keeps throwing me an error > make: *** [config.status] Error 1 [03:45] says it cant find "flex" [03:46] Ok, thanks ScottK-laptop and nxvl. I already have the files attached to a LP bug report. If a sponsor requests that I move it to REVU, I will. The advantage to keeping it on LP is that I only need one MOTU to look it over and upload it, vs the 2 needed for REVU [03:47] nhandler_: One versus two is new upstream versus new package. Doesn't matter if it's on revu or not. [03:48] csilk: then add flex to the build depends [03:48] or flex-dev or something like that [03:48] i did [03:48] then you added the wrong package [03:49] it's looking for flex as library [03:49] not as package (i think) [03:49] what package is it again? [03:49] flex [03:49] \o/ my first patch to the -installer [03:49] \o/ [03:49] csilk: package flex? [03:49] yup [03:49] csilk: can you send me the lp link to the package please [03:49] (and version) [03:52] ScottK-laptop: did you know if there is a policy for bzr branches sponsorship? === LucidFox_ is now known as LucidFox [03:52] ScottK-laptop: just propose the merge with the oficial one and subscribe the team to that? [03:53] nxvl: There is not. [03:53] AFAIK MOTU sponsorship process is built around debian packages. [03:54] ScottK-laptop and nxvl: I just re-read the REVU wiki page "If you packaged a new upstream version (of a package that is already in the archives), and you want to get it sponsored, do not use REVU to get it sponsored. Instead see SponsorshipProcess." [03:55] ScottK-laptop: it's a main sponsorship [03:55] Ah. [03:55] I was confused by what channel we're in. [03:55] Dunno what the rules might be for that then. [03:57] ok [03:57] :D [03:57] i will need to ping colin on monday [03:58] need to go [03:58] god damn [03:59] dont you just hate it when life gets in the way of productivity [04:30] success, one passed out girlfriend recovered from one friends bathroom [04:38] nxvl, i fixed the flex problem, i was just overlooking several dependencies [04:39] added them to the control file now and all is well [05:24] argh, I missed the MOTU meeting [05:24] I seem to be alseep whenever we have them [05:27] me too [05:40] * NCommander listens to the theme of House MD [05:43] ... [05:43] NCommander, are you peaking at my last.fm? [05:44] No, youtube [05:46] * NCommander thinks House is awesome [05:49] I know revu don't usually accept packages that have no man pages, but what if the app has no written documentation in the first place? [05:52] csilk: All the more reason you should write a man page for it. [05:53] ScottK, thing is, i cant test this software as it requires specialist hardware [05:53] so there is no way i can write the documentation [05:53] Use the source luke. [05:54] haha if only it were that simple === fabrice_sp_ is now known as fabrice_sp === fabrice_sp_ is now known as fabrice_sp [07:30] NCommander: there was a MOTU meeting that actually took place? [07:31] geser, that seems to be the popular opinion [08:48] <\sh> moins [09:22] siretart: fyi: almost done merging fai 3.2.11. Will try to find time to test (later) [09:22] * allee afk [09:52] Good morning. [10:04] allee: excellent, thanks [10:09] morning iulian [10:16] Hey jpds [10:45] morning [10:45] morning RainCT [10:46] Hi RainCT. [11:14] Does anyone know of an example of tarball-in-tarball packaging off the top of their head? [11:14] I can't find any documentation about how to do it [11:15] volkris, tarball-in-tarball packaging? [11:15] volkris: vim. [11:16] directhex, I have a package where the makefiles are distributed separately from the code, so one has to unpack the makefiles and put the code in the src subdirectory of that tree [11:16] oh. erk. [11:16] So I asked about it here yesterday and was advised to go tarball-in-tarball [11:17] "erk" basically sums up my thoughts every time I'm asked to build this monstrosity :) [11:29] volkris: I can think of coreutils, glibc and gcc off the top of my head, but I doubt those are "simple" examples [11:31] Well I see vim has tarballs in its source distribution [11:32] it's not the simplest example either, but probably mainly because it builds multiple packages [11:38] does someone have a Python bug for me? :P [11:39] gah. stupid debian with their stupid mailing list culture [12:17] the command "dpkg-buildpackage -us -uc -rfakeroot" gives error "debian/rules: 3: include: not found" what could be the issue here ? [12:18] It starts with "#/bin/sh" rather than "#!/usr/bin/make -f" ? [12:18] StevenK: "#!/usr/bin/make -f" is required ? [12:19] Yes [12:19] It is required to be the first line in the file [12:19] StevenK: was following this guide https://wiki.ubuntu.com/PackagingGuide/Python [12:20] I think the wiki format misses the shebang line [12:21] StevenK: heh, that was it [12:21] StevenK: now it says python: can't open file 'setup.py': [Errno 2] No such file or directory [12:21] Then your source package doesn't include a setup.py file? [12:21] StevenK: but https://wiki.ubuntu.com/PackagingGuide/Python does not talk about any setup.py [12:22] StevenK: no, any guides on making one ? [12:22] I've corrected the wiki page to show the #!/usr/bin/make -f [12:22] include /usr/share/cdbs/1/class/python-distutils.mk [12:22] That assumes python distutils [12:22] StevenK: python-distutils-extra deb package? [12:23] No, it assumes the software you're packaging uses distuils [12:24] StevenK: Err.. i have a few scripts in python that i would like to package and share, hence doing this [12:25] tab_shift: Then you aren't using distuils, I'd suggest dropping the -distutils.mk include and using straight debhelper [12:26] StevenK: using debhelper means this command dh_make -c gpl -s -b ? [12:26] Don't run dh_make again. And it doesn't. [12:27] tab_shift: I'd suggest you read https://wiki.ubuntu.com/PackagingGuide [12:29] StevenK: thanks, also i don't have a makefile as well, is it mandatory? [12:31] tab_shift: No, you can deal without it [12:34] StevenK: I'll read up & try to figure it out, thanks [13:04] james_w: New debdiff for bug 257007. Thanks for your help :-) [13:04] Launchpad bug 257007 in revelation "[Intrepid] crash / error when using revelation-applet" [High,Triaged] https://launchpad.net/bugs/257007 === Kmos_ is now known as Kmos [13:36] * StevenK kicks java [13:36] mcasadevall: Here? :-) [13:37] StevenK, for some definitions of hear [13:37] *here [13:37] mcasadevall: So, I'm trying to update cacao to 0.99.3, and it fails. Mind helping me poke at fixing it? [13:38] why does it fail? === mcasadevall is now known as NCommander [13:38] class file for java.lang.Object not found [13:38] And a whole bunch steming from that [13:39] Which makes me think the bootclasspath isn't getting set. [13:39] Now, setting it to /usr/share/java, /usr/lib or /usr hasn't fixed it [13:39] sounds like loads of fun [13:40] Indeed [13:41] StevenK, it's all yours ;-) [13:42] * NCommander runs [13:42] Hmph [13:42] StevenK, post your work to a PPA, I'll look at it when I wake up [13:42] unless your just trying to compile the straight debian package [13:43] Well, actually, 0.99~rc5 from Debian works. 0.99.3 from upstream doesn't [13:46] heya [13:52] StevenK, maybe there is a reason 0.99.3 isn't in sid or experimental [13:52] morning emgent [13:54] ahoi emgent [13:54] StevenK: did you see my nfdump debdiff? [13:54] StevenK, how are you trying to compile it [13:56] StevenK, and what are you using for --with-classpath-install-dir [14:06] sebner: I did not [14:07] NCommander: I think I know what is sort of going on -- they renamed all of the ./configure options for 0.99.3 [14:07] StevenK: ah, I attached a debdiff and thought you will notice it because you are subscribed to the bug [14:07] StevenK, that needs to be set to the where classpath is [14:08] StevenK, or in other words: http://xkcd.com/293/ [14:10] * StevenK beats NCommander to death [14:10] NCommander: I've been looking at stuff for an hour before I asked, you know [14:10] Ok [14:10] I apologize [14:10] I suspect thats the problem however [14:10] I'll see if I can just compile the tarball here [14:10] I know it was in jest, so shrug :-P [14:11] NCommander R.I. P [14:12] sebner, well, it just happened that StevenK pointed me to that why I was reading xkcd [14:12] StevenK: will you review it or should I subscribe u-u-s? [14:12] NCommander: xD xD xD :D [14:12] Which bug is it? [14:12] I might care [14:13] StevenK: bug #268811 [14:13] Launchpad bug 268811 in nfdump "nfdump 1.5.7-5 introduces an empty -dbg package, while the main package ships unstripped binaries" [Undecided,Confirmed] https://launchpad.net/bugs/268811 [14:13] sebner: I'm not subscribed [14:14] StevenK: argh sry. the other steve xD [14:16] Muahaha. [14:16] * StevenK wins [14:16] Hew: looks good thanks. I'll upload after lunch. Did you warn Jef about the issue? [14:17] james_w: now you are a hardcore u-u-s guy :D // I'll upload a new debdiff for wordpress later :) [14:18] StevenK, what's your configure string [14:18] sebner: cool, what approach are you going to take? [14:18] james_w: I emailed him before just before you posted. Now that you've checked the new debdiff I'll send off another mail. [14:19] james_w: I have to investigate further but I'm wondering since the fix is from (beginning) 2007 [14:19] NCommander: --with-java-runtime-library-prefix=/usr --disable-disassembler --libdir=/usr/lib/cacao --with-jni_h=/usr/include/classpath --with-jni_md_h=/usr/include/classpath [14:19] Hew: great, thanks. I haven't downloaded and patched to check thoroughly, but I trust you to tell the difference between a tab and a space :-) [14:20] sebner: yeah, I'm not sure I agree with the analysis in the other bug report. I would have to read the CVE and the patch again to know whether we should pull it in [14:20] james_w: Heh, yes. I noticed it before and had a dilemma, whether I should change his patch or leave it for consistency. Tabs are definitely there now :-) [14:20] james_w: I know and I'll do that :) [14:21] Hew: it's something you should change as it can cause failures, I just can't tell you under which circumstances. [14:23] james_w: from a quick look it seems that we really don't need this fix as long as we stick to 2.5.x and not 2.6.x [14:23] james_w: Ah ok, good to know. [14:23] StevenK, it builds fine from source here [14:24] are you build-deping on classpath-common [14:26] james_w: at least we merged the debian version which has the fix ^^ [14:29] james_w: /me is confused. I merged the actual version from Debian and Debian *has* this fix included so we get it as well when we merge it [14:48] NCommander: Yes [14:48] NCommander: It builds, so I'm good [15:13] Hew: uploaded, thanks. [15:13] james_w: excellent, thank you :-) [15:14] ping === allee_ is now known as allee === emma is now known as emma42 === emma42 is now known as emma [16:42] I'm having trouble using pbuilder. I am backporting packages using it (I am on Intrepid, and am trying to build some packages for hardy), but I can't get one package to build because it depends on the others. I tried the method in PbuilderHowto on the wiki for setting up a local repository for pbuilder and using it with pbuilder, but pbuilder still doesn't use it and complains about the packages missing. [16:56] anybody have actual working flashplugin-nonfree in intrepid? [16:57] LaserJock: me [16:57] jussi01: it doesn't take 100% CPU? [16:58] LaserJock: I havent looked exactly, but other programs are running fine with it... I can take a look if you like [16:59] well, I was getting pegged CPU and flickering with 64-bit Intrepid [16:59] so I reinstalled with 32-bit and still the same thing [16:59] LaserJock: nope, its fine here on 32bit [16:59] so I'm assuming it's not specific to my particular install [17:00] I suppose it could be hardware related... [17:00] LaserJock: Im using it in firefox on kubuntu if that helps [17:00] I'm using FF on Ubuntu [17:01] ok, so looks like either ubuntu or your machine specific [17:03] flashplugin-nonfree might also be a good target for a Hug Day [17:03] 174 open bugs and 90 in New status [17:04] heh i jsut came to ask about the flashplugin backport. it says it's 10 but it's really still 9 [17:04] why is that? [17:04] LaserJock, mine was acting similarly because my pulseaudio got messed up again [17:04] on 32 bit [17:04] it seems to be having lots of troubles, but i've not looked into it yet [17:04] superm1: what's "messed up"? [17:05] only works after i kill the daemon, rm ~/.pulse* and then restart the daemon [17:05] and then only works for a limited undetermined period of time [17:07] mcquaid: how do you know that it's 9? [17:08] about:plugins [17:10] mcquaid: and what does dpkg -l | grep flashplugin-nonfree give you for the version? [17:10] 10.0.1.218+10.0.0.525ubuntu1~hardy1+really9.0.124.0ubuntu2 Adobe Flash Player plugin installer [17:11] heh [17:11] heh, ya it's in the file name. what's that all about +really9.0.124 wtf? [17:11] how about 'really' package 10 and i dunno call it 10 [17:11] so my guess is that we had to revert back to 9 [17:11] ah [17:11] mcquaid: it's not exactly that easy [17:13] i understand. i just wanted beta 10, i've never had a problem installing flash manually but can't get it working for some reason [17:13] and i saw there was a backport, i thought ah cool, but then it's really 9, so I'm back where i started [17:13] right [17:14] Flash 10 *was* uploaded to -backports [17:14] but then was reverted to 9 [17:14] do you know the reason for reverting back? [17:14] just curious [17:14] mcquaid: it was crap on hardy [17:15] too bad. read a lot of surprisingly good comments on 10. like fullscreen finally using vid overlay properly [17:15] that's all i wanted [17:15] damn i hate flash [17:16] mcquaid: well 26 days to go ;) [17:16] so flash 10 is fine in intreped? [17:16] intrepid [17:17] mcquaid: well, we have the beta and it's half crappy. maybe the rc or final will be better [17:18] heh. well one can hope [17:18] 'flash 10! now with less crap!!" [17:18] ha [17:19] mcquaid: flash 10 will be a lot better than flash 9. but dunno if still crappy or not ^^ [17:19] give swfdec or gnash a go [17:20] how is swfdec or gnash in regards to vid playback? fullscreen? overlay? [17:20] orly_owl: I tried it and it wasn't that usable for me [17:20] did you compile latest trunk? [17:21] mcquaid: dunno. try it and find out [17:21] #gnash and #swfdec btw [17:22] flash 10 in Intrepid for me sucks because it's taking 100% CPU and flickers [17:23] LaserJock, sigh same old story eh. how hard can it be to use xv? [17:25] LaserJock, is that with video or do you mean other types of flash, like a game or funky menu's etc [17:25] any flash [17:25] well that sucks to hear [17:25] I think anyway [17:26] hello [17:26] folks I have a package I could put into the repositories [17:26] anyone want to advise? [17:26] I think it should go in the multiverse [17:27] everyone in bed? [17:27] hedkandi: well, would it depend on anything in multiverse and what's the license? [17:27] ok [17:27] it depends on two lgpl things [17:27] the licence is pretty much gpl except [17:28] for a clause which says that modifications may not be distributed [17:29] That's enough that it has to go in multiverse [17:29] does it? [17:29] I wasn't actually sure from reading [17:29] the ubuntu pages (which are confused and badly written) [17:30] I actually think some clarification would be good, as to what restricted is about [17:30] when's the next LTS out? [17:30] 2010 [17:30] but anyway, I think I'll be happy with the multiverse [17:30] Yes. Since any modified package would be undistributable, it's not Free software. [17:30] april 2010? [17:30] That's the schedule. [17:31] ish, yes [17:31] ok [17:31] is 8.04 going to stop at 8.04.4? [17:31] hedkandi: is your question Multiverse vs Restricted or Multiverse vs Universe? [17:31] ScottK: well I disagree with your assumption that free means you can distribute modified versions [17:31] hedkandi: It's not mine. It's Debian's and Ubuntu's, but I agree with it. [17:31] hedkandi, yu can disagree, but you're wrong [17:32] hedkandi, http://en.wikipedia.org/wiki/Four_Freedoms [17:32] i'm usually right 'cos I'm very clever. [17:32] wait, http://en.wikipedia.org/wiki/The_Free_Software_Definition [17:32] LaserJock: restricted I think. [17:33] freedom 3, "# improve the program, and release your improvements to the public, so that the whole community benefits (freedom 3)" [17:33] hedkandi: ah, well that I can kinda see where there could be some confusion [17:33] hedkandi: but Restricted is really only used for a few special packages [17:34] so basically assume you're not going to get anything into Restricted directly [17:34] is this the place for recommendations? [17:34] resticted is for non-free apps supported by canonical, which basically means nvidia drivers. why do canonical want to offer support for your package? is it that critical? [17:34] LaserJock: ok [17:34] directhex: I don't think that's even quite true [17:34] but I'd be okay in the Multiverse then? [17:34] Restricted is primarily for drivers and non-free "blobs" [17:35] So how would I get my package into the multiverse? [17:35] not just any non-free software, generally speaking [17:35] but even things in multiverse/restricted need to be distributable [17:35] my program is distributable. [17:35] hedkandi: same way you would to get it into Universe [17:36] i would like for code::blocks to move into universe, is there a guide for these things? [17:36] LaserJock: call me ignorant, but could you just remind me of the process please? [17:37] hedkandi: package it, put it on REVU, get 2 MOTU +1 votes [17:37] !REVU [17:37] REVU is a web-based tool to give people who have worked on Ubuntu packages a chance to "put their packages out there" for other people to look at and comment on in a structured manner. See https://wiki.ubuntu.com/MOTU/Packages/REVU [17:37] LaserJock: thanx [17:37] good thanks. I didn't know about revu. [17:39] And it's way too late for Intrepid. [17:39] right [17:40] point of general discussion: I think it's bad that other people can wreck my software and distribute it as [17:41] if they've "improved" (haha) it. [17:41] The fact is that there are a lot of incompetent programmers and designers hanging around ubuntu [17:41] and I don't want them playing with my program and introducing bugs and inconsistencies. [17:42] fine, then don't put it in Ubuntu [17:42] I think this is entirely sensible,and the motu would do well to accept that. [17:42] accept what? [17:42] Ie if the people running ubuntu were a little smarter, they would allow software under my licence into ubuntu main. [17:43] I think the motu are being a bit anal in refusing my licence. [17:43] hedkandi, what is your license? [17:43] hedkandi: well, sorry, but we have established principles [17:44] my licence says modified versions may not be distributed [17:44] and one of them is that we are free to modify [17:44] indeed and that's different [17:44] there's no point to having an oss license if you can't redistribute it [17:44] no [17:44] that's like, a fundamental right [17:44] hedkandi: so, you want something that aint free mislabeled and supported by canonical? [17:45] my licence comes under the "free" banner. [17:45] well, by any definition of Free software I've seen, it doesn't [17:45] i aint free to change it and redistrobute it, which is one of "three freedoms" [17:46] hedkandi: One thing that is supported is a license condition that if you change it you have to rename it. [17:46] if gpl'd software was "free" I'd be able to modify it, and then sell it. So I think your definition of free is contrived [17:47] hedkandi: aint you? [17:47] ScottK: tell me more?? [17:47] That way modifcations can be distributed, but the 'name' of your package doesn't get 'tainted'. [17:48] There is an interesting distinction to be made, actually, between a 'modification' and a 'modified version' [17:48] or better yet, you could work with Ubuntu to make sure the modifications are "good" [17:48] step 1: introduce a debian/ folder [17:48] step 2: IT'S MODIFIED, GAME OVER! [17:48] hedkandi: See DFSG #4 http://www.debian.org/social_contract#guidelines [17:49] LaserJock: indeed, but ultimately that decision should be made by me. [17:49] hedkandi: That's exactly why it's not Free software. You are retaining control over it. [17:50] I'm allowing modifications, so I'm not. [17:50] you're not allowing modifications. you're providing a read-only copy of the source [17:50] wake up folks [17:50] and conveniently ignoring that you can't make a package without introducing "changes" in the form of the debian/ package structure [17:50] I'm allowing modifications, but I'm forbidding their distribution. [17:50] hedkandi, you know this has been covered more than once in debian's history, right? there's precedent? [17:51] hedkandi, you're talking about the PINE license. [17:51] no, actually. [17:51] well, the old version [17:51] sorry I'm ignorant of the pine licence [17:51] hedkandi: Ubuntu is a distribution, so you're forbidding Ubuntu from having any changes. [17:51] I guess so, yep. [17:52] that means technically we can't even package it [17:52] http://en.wikipedia.org/wiki/Pine_(e-mail_client)#Licensing_and_clones [17:52] LaserJock: why is that please? [17:52] LaserJock: Typically debian packaging isn't considered a modification. [17:52] hedkandi, how do you package it if we can't insert any files into it with packaging metadata? [17:52] ScottK-laptop: *any* patch would be [17:53] Yes, but this is common in multiverse. [17:53] So it'd either work unmodified or not, but we couldn't fix it. [17:53] well can I make the package myself? [17:53] ScottK-laptop: he was talking about it should be allowed in Main though [17:53] Yes, but it can only go in the Multiverse repo. [17:53] Right, it clearly can't. [17:54] hedkandi, YOU WIN THE PRIZE! http://mako.cc/copyrighteous/images/trophy.png [17:54] best image of all time [17:54] ok wait a mo. [17:55] ScottK-laptop: even in Multiverse though, we have no way of enforcing that license really [17:55] ScottK-laptop: what prevents a MOTU from uploading a patch, we can't expect people to look at every license before they upload just to see if they are allowed to do so [17:55] If it's in Multiverse, they really need to. [17:56] Which is why many motu just ignore multiverse. [17:56] I'm not entirely sure [17:56] If it's distributable, it can go in multiverse. That's the rule. [17:56] no, it's not exactly [17:56] there isn't any hard rules for Multiverse [17:56] No? [17:57] we know that it *has* to be distributable, but that's not necessarily the only restriction, the Archive admins could reject for other reasons if the wish to [17:57] if you *really* want to prevent standard procedure - e.g. fixups for linking locations etc - then go down the j0000rg route. either permit distribution but retain copyright over the name (like iceweasel in debian) or the cdrecord route for spewing lots of "you're an evil hippie, you should be using solaris" every time an 'unapproved' build is used [18:00] ScottK-laptop: obviously when there is no source we can't modify it, but if it's got source I expect that Ubuntu should be able to modify it [18:00] well I just read about pine. thanks to directhex [18:00] we've had cases were *only* Ubuntu can modify it [18:00] LaserJock, firefox? [18:00] firefox is tighter than that, but kinda yeah [18:01] I don't understand the trophy I'm afraid. [18:01] that wasn't an example I was thinking [18:02] hedkandi, joerg schilling claims to be a free software advocate, but is abbrasive, refuses to accept patches or fixes, and releases undistributable code. in the end most linux distros dropped his software in favour of better-licensed solutions [18:02] hedkandi, ooh, remember xfree86? how many people use xfree86 these days? [18:02] well, basically [18:03] http://en.wikipedia.org/wiki/Xfree86#2004:_Licensing_controversy [18:03] you're free to do what you want with your code [18:03] it would probably make it ok into Multiverse [18:03] but if you're not going to allow us to modify it I'd personally rather you not put it in Ubuntu at all [18:03] just give out .debs on the project site or something [18:04] i'm unconvinced it's acceptable for multiverse, and i don;t think anyone would touch it within debian or ubuntu [18:04] well, it'd probably be no worse than a lot of Multiverse [18:05] but my guess is that we'd end up violating the license inadvertently and hedkandi would get mad [18:05] joerg schilling [18:06] I'm surprised that more people don't take this stance however. Is it unusual? [18:06] LaserJock, there's also the ion3 issues within debian caused by the same attitudes as j00rg [18:06] directhex: sure [18:06] I don't remember xfree86 I'm afraid. [18:07] hedkandi, it's no an unusal stance, no. and it's an acceptable one. but saying black is white annoys people. if i can make your app 10x better but it's illegal for me to give LaserJock those improvements, then that's not free software [18:07] it's closed software with a NDA'd copy of the source [18:07] directhex: We have binary only stuff in multiverse, so this is definitely within the scope of what's allowed. [18:08] except I consider the binary stuff as being better since there's no chance of modification [18:08] what's closed software with nda? the stuff in multiverse? [18:09] hedkandi, some of it, yes [18:09] if you're not gonna let me make the modifications, then don't give me the source [18:09] hedkandi, actually, i think cdrecord reappeared in multiverse. it's right up your street [18:09] ? [18:09] hedkandi, what's the difference between a closed source app, and an open source app i can't improve? [18:10] directhex: you can modify my swe, but you can't pass that on. [18:10] hedkandi: but "passing it on" is *what* we do [18:10] as I've said, I don't mind people modifying it for their own use. [18:10] i see. well well. [18:10] The discussion should now turn to the difference [18:10] between a modification and a modified version. [18:11] I don't mind people distributing diff files and patches for my swe [18:11] Then allow that in your license. [18:11] That would make it allowable for Universe/Main. [18:11] mhm [18:11] well is a patch considered a derived work? [18:12] when the patch is applied I'd think so [18:12] I didn't think that it was legally possible for me to say anything about patches [18:13] That's one of the conditions mentioned in DFSG #4. [18:13] When you apply a patch, you create a derived work, but you aren't distributing a derived work. [18:13] actually, i know this license [18:13] not just PINE. there's a closer parallel [18:14] can someone just tell me quickly what debian is all about please? is it like gnu? I'll go read. [18:14] Right, but once I upload it to Ubuntu, then I am distributing it. [18:14] Microsoft Reference Source License (Ms-RSL) [18:14] Debian is the distro that Ubuntu is derived from. [18:14] We get most of our packages and our policy from them. [18:15] ScottK-laptop: link to DFSG? [18:15] http://www.debian.org/social_contract#guidelines [18:15] What,exactly is the difference between debian and ubuntu? [18:16] and can I run debian linux on my pc? [18:16] Ubuntu is a derivative of Debian. [18:16] hedkandi, debian is a distro. ubuntu is a distro. ubuntu is based on debian. this is possible because debian is Free software, and all the packages in it can be redistributed [18:17] hedkandi, okay, here's the laundry list of problems, there may be more. [18:18] hedkandi, 1) a security problem is found in your app. we can't fix it because that's a patch and is TEH BANNED [18:18] hedkandi, 2) a new lib which your app depends on has an altered API. we can't fix it because that's a patch and is TEH BANNED [18:18] directhex: When was the last time you saw a security fix in multiverse? [18:19] ScottK-laptop, well, indeed. multiverse is a security hole [18:19] So if he wants to be in multiverse it's fine. [18:19] ScottK-laptop: that we don't often see them is not exactly the same as saying that can't happen though [18:20] True, it's just that generally we can't fix stuff there and so it just sits. [18:20] or dies [18:21] non-free stuff that isn't considered "vital" (like nvidia, no flame wars please) dies, or at least atrophies badly [18:23] hedkandi, 3) a user files a bug report. we can't fix that bug. [18:23] directhex: wrong about (2) [18:23] the lgpl requires that you can update the linked library [18:23] for this purpose, I provide source code for mine [18:24] The "security problem" is another way of saying "we think there's a bug" [18:24] And then I'm afraid it my opinion of what is a bug that matters. [18:24] oh no wait (2) [18:24] hedkandi, part of what this ties into is the concept of stable releases. if you release version 2 of your app, which fixes bugs in version 1, that isn't good enough - a stable release is expected by users to not suffer sudden unexpected changes, so versions are not changed in stable releases - problems (and ONLY the problems) are patched, but the version remains [18:26] directhex: ok so a "stable release" is a released branch [18:27] directhex: this thing about "stable rElease" does destroy your point (2) however [18:27] as in a stable release of an api, only the bugs will be fixed and the api won't change. [18:28] ask ncommander about the ada 4.2 transition [18:28] Can I Mention something I just found on the debian site please? [18:28] http://www.debian.org/intro/free [18:28] this page references the artistic licence [18:29] and it is called "free" at debian [18:29] just take it up with debian-legal [18:29] you obviously aren't interested with anything said here [18:30] ? [18:30] but, short answer, your app WILL NOT be considered free software, for inclusion into debian main, or ubuntu main/universe. and nobody within MOTU will give a rat's arse about your package if it's in non-free or multiverse (you'll need to take on that burden yourself) [18:30] you're probably better off just packaging it yourself [18:30] Anyway, if you look at the artistic licence 2.0 [18:30] and keeping it on your own site [18:30] it distinguishes between standard and modified versions [18:31] and there are restrictions on distributing the modified versions [18:31] ok well I can package it myself I think! [18:32] is there a public repo for the files found in a package's debian dir? [18:32] like if I want to get qemu's, but don't want the whole qemu source too [18:33] hedkandi: Yes, those are very close to the types of restrictions allowed under DFSG #4. [18:33] I rather like the artistic licence 2.0 [18:33] so tell me, can we compromise. If I use al2.0 can I have my swe in main? [18:33] mainly so I can ask a questions about it: how do I get it to do ./configure --target-list=x86_64-softmmu [18:33] Artistic 2.0 is DFSG free, so it would be OK for Universe. To get in Main, Canonical has to want to support it. [18:33] what are usually parameters considered while granting freeze exception? [18:34] CarlFK: well, you could get just the .diff.gz? [18:34] slytherin: It's generally a question of risk versus potential benifit. [18:34] It's not an exact science. [18:34] CarlFK: some packages are maintained that way. but not all. [18:34] yes I did say AL was dfsg free, and directhex blew a fuse. [18:35] ScottK-laptop: Ok. I am considering applying for exception for gnomeradio. Let's see what happens. [18:35] AL is. AL2 is not. [18:35] directhex: Why not? [18:35] and as per the authors, AL is for use only when dual-licensed with something better [18:35] ScottK-laptop, non-modification terms. check which license is in your /usr/share/doc/common-licenses/ [18:35] I know which is there. [18:36] I don't see anything in those restrictions that violates DFSG. [18:36] They are distasteful, but pass DFSG #4 as I read it. [18:36] "You are permitted to use the Standard Version and create and use Modified Versions for any purpose without restriction, provided that you do not Distribute the Modified Version." [18:36] ok - was hoping there was something similar to apt-get source qemu [18:36] is the term from AL2. and not in DFSG-free AL1 [18:37] directhex: There is also the section "Distribution of Modified Versions of the Package as Source" that I think makes it OK. [18:38] at any rate, WHAT IS IT WITH PEOPLE WHO INSIST THEIR NON-FREE APP MUST GO INTO MAIN? [18:38] seriously, it just gets tiresome [18:38] if your app is non-free, then just say it's non-free and go within the little pen set aside for that [18:38] well as you said, no-one will give a rat's if it's in multiverse [18:38] what is with people who insist that their application go into main :-P [18:38] directhex: no need to shout :-) [18:39] I think al2.0 is free [18:39] LaserJock: [18:39] LaserJ: can I get al2.0 into main? [18:39] * directhex wants wolfenstein: enemy territory in main :'( [18:39] hedkandi: Has to be in Universe to even be considered for Main. [18:39] hedkandi: I wouldn't say that about multiverse. Far more people use Sun JDK even though it is in multiverse. [18:39] Gotta run. [18:40] ok well for the time being I'll aim for the multiverse then. [18:40] I'm gonna have to go soon folks... [18:40] directhex: grow up buddy, play wesnoth. :-D [18:40] hedkandi: I don't know specifically about al2.0 [18:40] ok [18:40] * siretart finds 'joerg schilling' in the backlog... [18:40] not to worry. [18:40] but it does need to be in Universe before it can go into Main [18:41] okay then. That's fine. [18:41] I'll get my package sorted out and "wing it" later on. [18:41] sounds good [18:41] thanks for all the advice folks. Doubtless I'll be back sometime [18:42] is main a status symbol? [18:43] directhex: I am sure it is to some [18:43] hmm, I wouldn't exactly call it that [18:43] frankly my life would be easier if i could get sponsorship from u-u-s instead of u-m-s [18:43] but there's main, and there's main in the default install [18:43] for me it's always been a core subset of Free software that Ubuntu chooses to support more extensively [18:44] software in Main goes through security audits and generally seeks active upstreams [18:45] on a related note, where's my pet debian developer [18:45] directhex: why would it make your life easier? [18:46] is u-m-s not very responsive? [18:47] it's smaller, so tracking down a main sponsor who has the time to do my packages (and agrees to do them) is tricky sometimes [18:47] took me weeks to get something as bloody trivial as a documentation package sponsored [18:48] well, there are things to be improved for sure [18:48] recently Canonical has started having all their Core Devs doing 1hr/week in the sponsorship queue [18:49] I think more particularly though [18:49] Main tends to be more maintainer focused, Debian-style [18:49] so if a Main package doesn't have a designated maintainer it can pretty easily fall through the cracks [18:50] there are a few Main packages that are not as well maintained as most Universe apps [18:50] or that maintainer is overly busy. or on holiday. or busy learning to cook [18:50] I'd like to see my QA work in identifying those packages and getting people to help make sure the cracks close up [18:51] s/my/more/ [18:53] LaserJock, i did a fair bit of the work on mono for intrepid, which is in main. if there's anything which can make a core dev really really far too busy all of a sudden, it's packages with "mono" in the name# [18:53] * sebner laughs :) [18:54] Any MOTU free enough to sponsor the fix to resindvd plugin I backported from upstream CVS? [18:55] directhex: mono-develop? [18:56] LaserJock, that one hasn't been touched for a while since there hasn't been a stable release for a while, and we in the pkg-mono team don't want to offer support for a package marked "alpha" by upstream [18:57] I just don't like sponsoring things I don't know much about, especially in Main [18:58] it's not that I dislike the software itself, I just don't like blindly uploading things [18:58] LaserJock, well, i kinda understand that attitude, but i think there's a difference between random apps, and stuff actively team maintained in alioth by debian & ubuntu people alike [18:59] i mean, if canonical want to PAY me to do it, then fine :p [18:59] why don't the Ubuntu people sponsor it then? [18:59] the Ubuntu people in the team that is [18:59] ahh, that's the last of my angostura gone. now i can start on the PROPER stuff [19:00] LaserJock, we only have one core dev, and he's rather busy at the best of times with the other teams he's in [19:00] who? [19:00] slomo [19:01] oh, and he's mostly debian these days anyway [19:01] ah yeah [19:02] directhex: at least MD trunk is pretty stable :P and I touched it in past :) [19:03] sebner, i know it's a bit unstable right now. i know. but offering years of stable release support to a random svn checkout of a large IDE? [19:03] bleh :/ [19:04] I'm not a huge fan of mono right now, so many good apps that eat all my memory :-) [19:04] LaserJock, better than java on that front. and it should improve a fair bit in jaunty, with mono 2 [19:05] yeah? that'd be cool [19:05] directhex: was just a joke :) [19:05] I love gnome-do, Tasque, and Tomboy, but they eat up so much RAM [19:05] LaserJock, oh, and ~20 meg of space on the install cd saved too [19:05] hmm /me only uses MD [19:05] LaserJock, it's mostly shared pages though [19:06] LaserJock: you also don't like banshee? [19:06] sebner: I dont' know, never really tried it [19:06] i like banshee, but i think it needs some more work before i put a proposal to the desktop team about looking at it versus rhythmbox [19:06] directhex: right, but those apps are only actively used occasionally and take up 1/3 of my in use RAM [19:07] LaserJock: more ram! yesterday I built a new Pc for my parents with 4gb ram :P [19:07] sebner, 4 is the right umber for a new pc these days, certainly with windows on it [19:07] directhex: ubuntu 8.10 on it :P [19:07] LaserJock, how much ram do you have? [19:08] 2GB [19:08] that seems a VERY large number [19:08] tomboy+tasque+gnome-do run ~100-150 MB [19:09] for me [19:09] that was on 64-bit [19:09] you're sure that's taking shared pages into account? [19:09] I believe so [19:09] f-spot is also pretty strange O_o [19:09] f-spot is the bane of my life [19:10] gnome-system-monitor shows 16.5M for tomboy, which i think is fair enough [19:10] now I'm running 32-bit so it's roughly half the usage [19:11] on 64-bit I would just get into gnome and maybe fire up FF and I was already at 1GB used [19:11] LaserJock: O_o [19:11] now it's more like 500MB [19:11] well this xchat window is showing more than twice the RAM use of tomboy [19:11] so i don't think blaming mono is entirely fair [19:11] * sebner has now 565mb ram for a normal pc use [19:12] for me Firefox is the only app that uses more than the mono apps [19:12] LaserJock: how much? [19:13] sebner: how much what? [19:13] LaserJock: firefox takes ram for you [19:13] ~ 100MB at startup [19:13] firefox is a hungry beast [19:13] i'm showing 25 meg with 1 tab... LaserJock, how exactly are you measuring your ram use? [19:13] at lot better than version2 [19:14] so far the difference between 64-bit and 32-bit is pretty significant [19:14] directhex: gnome-system-monitor at the moment [19:14] ff 3.1 does only take 135mb (while watching youtube) [19:14] but I also use top [19:14] oh, whoops, i forgot, this isn't ubuntu firefox, it's 32-bit from tarball [19:15] 38 for hardy firefox with 1 tab [19:15] in 64-bit intrepid each of Tasque, gnome-do, and Tomboy took about ~40MB each [19:15] on intrepid? really seems a lot to me? [19:16] which would be OK, IMO, if they were doing a lot of work or something [19:16] but they're little convenience apps [19:16] banshee clocks in at 45 meg to me [19:16] 25 for f-spot [19:16] 48 for monodevelop [19:17] those are chunky apps though, and i'm positive g-s-m doesn't tell you if 2 processes are sharing the same chunk of memory [19:17] right now on 32bit intrepid I'm getting 17MB+15MB+15MB for the 3 [19:18] so 32-bit makes a hug difference [19:18] *huge [19:18] well, yes, it does. but not as big a jump as you suggest IME [19:19] but they're still more than most other apps [19:19] well, I all I can report is the numbers I got :-) [19:19] how does pidgin weigh in? [19:19] for me, on 64-bit, it shows as marginally heavier than tomboy [19:19] right now it's heavy at 36MB [19:20] Pidgin is #3 for me right behind FF and Xorg [19:20] anyway, isn't evolution our default mail client? [19:20] i mean really, you could load every mono app in ubuntu and still use less ram than evo [19:20] ;) [19:20] on 64 bit the mono apps were using a lot more than Pidgin [19:20] I don't use evolution [19:20] not for me! really! [19:21] directhex@mortos:~$ dpkg-architecture | grep DEB_BUILD_ARCH= [19:21] DEB_BUILD_ARCH=amd64 [19:21] like I said, right behind FF and Xorg would be gnome-do, tomboy, and Tasque [19:22] and I'd have 900MB-1GB of used RAM [19:22] for 32-bit I'm running 500-600MB [19:23] so the mono apps seemed like the obvious target [19:24] well, python is ubuntu's language of choice, and looking at http://shootout.alioth.debian.org/u64q/benchmark.php?test=all&lang=csharp&lang2=python ... [19:25] right, python apps do much better for me [19:25] I was thinking of trying to port Tasque to python, but it's a bit beyond the amount of time I've got [19:27] though Ruby 1.9 looks really quite fast [19:27] I wonder what it's memory usage is like [19:27] ironpython! worst of both worlds! [19:27] python's crummy syntax, plus mono's system requirements! [19:27] lol [19:27] and, for extra measure, (c) microsoft [19:27] :) [19:27] PATENZZ [19:27] directhex: ;) [19:28] well, I don't care much, I just don't have time to learn another language [19:28] directhex@mortos:~$ grep Copyright /usr/share/doc/ironpython/copyright | head -1Copyright (c) Microsoft Corporation. [19:28] I'm not a programmer, nor will I probably ever be [19:28] *professional programmer, I suppose [19:29] so for now i stick to Fortran, C, and as much Python as I can get away with [19:29] all the cool kids use fortran [19:29] heck yeah [19:29] too muxh time spent supporting people who live on mpif77 :/ [19:29] I so hate fortran [19:29] LaserJock: just FYI wrt flashplugin-nonfree - mine works fine for youtube, but on cnn videos basically freezes up. [19:30] but whatever, that's what my code is written in so .. [19:30] and it doesn't work with gfortran [19:30] LaserJock, are you a u-m-s then? [19:30] so I got to *forward-port* gcc 3.4 [19:30] directhex: I'm a Core Dev [19:31] I don't think I'm currently in the u-m-s team since I'm trying to finish my dissertation [19:31] LaserJock, if i need something mono-related sponsored, can i add you to my list of friendly people to talk to? i don't want to abuse pitti's good graces [19:32] hmm, good question [19:35] LaserJock: ah core-dev! :D mind sponsoring a merge? It seems they forgot about me [19:36] what's the package? [19:37] LaserJock: tuxtype, bug #261991 [19:37] Launchpad bug 261991 in tuxtype "Merge tuxtype 1.5.17.dfsg1-3 from Debian(Unstable)" [Undecided,Confirmed] https://launchpad.net/bugs/261991 [19:38] sebner: oh, right, is that still not done? [19:38] LaserJock: *nope*, otherwise I wouldn't ask :) [19:39] sebner: that was somewhat of a rhetorical question ;-) [19:39] LaserJock: ah ^^, mind sponsoring it? [19:40] well, it's not easy [19:41] I emailed the Debian maintainer a while back about it [19:41] sebner: I had a couple issues with your debdiff [19:42] LaserJock: yes? O_o [19:43] sebner: did you check the .desktop file for validity? [19:43] I'm pretty sure the original Categories= was fine [19:43] it at least shouldn't have Applications [19:43] not sure about GNOME [19:44] Hey guys, I'm trying to find an archive administrator to have a look at bug 274276 and hopefully sync it. [19:44] Launchpad bug 274276 in salasaga "Please sync salasaga 0.8.0~alpha4-1 (universe) from Debian unstable (main)." [Wishlist,Confirmed] https://launchpad.net/bugs/274276 [19:44] sebner: also you say you merge greek and malayalam but there's also diff to svenska [19:45] iulian: archive admins are usually found in #ubuntu-devel [19:45] LaserJock: this maybe was autogenerated. I'll recheck.. also the .desktop file [19:47] slytherin: I said this few hours ago in -devel too and it seemed that no one answered. [19:48] iulian: right, but the bug report is just fine [19:48] iulian: they must be busy. besides it is weekend. :-) [19:49] the archive admins are subscribed and the sync is acked [19:49] so they'll get to it when they can [19:53] LaserJock: Well, slangasek told me a couple of days ago that some of the bugs are not going to be closed, IIRC. This is the reason why I'm worrying about. [19:55] I don't think I said they won't be [19:56] note to self: never "just" finish the bottle [19:56] I probably declined to commit to getting them closed before release [19:56] * directhex is happy with the shape of his little slice of ubuntu [19:58] slangasek: Ahh, I misunderstood then, sorry. [19:59] directhex: hmm? [19:59] zul: Hi, have you some time to sponsor bug #264554 or should I try to find an other core-dev? [19:59] Launchpad bug 264554 in xen-3.3 "libxen3 and libxen3-dev both include /usr/lib/libblktap.so" [High,Confirmed] https://launchpad.net/bugs/264554 [20:00] ahoi geser \o/ [20:00] Hi sebner [20:00] sebner, which bit? the "monodoc merge happened so mono in intrepid is a complete, well-formed 1.9.1 stack" or the "just because you want to finish the angostura to move onto the santiago, never drink rum from a pint glass" bit? [20:00] directhex: ah, I understand :D [20:01] sebner, WHICH BIT? >_< [20:01] directhex: both, of course ;) [20:03] sebner, :o :D 8D [20:03] directhex: But I still don't know why I had monodoc 1.9.1-1.deb installed xD [20:07] sebner, you use monodoc. perhaps you installed the debian deb, for better more recent docs? [20:08] monodevelop, sorry [20:08] directhex: maybe, but I already told you that I *really* can't remember. I think I changed from ubuntu 1.0 to trunk :\ [20:51] I recall that it is possible to upload stuff in a way that automatically gets launchpad to mark associated bugs as "fix-committed" (via the changelog comments? bzr --fixed?) but I can't find a way that works - am I wrong? [20:52] I don't know about fix committed bug for "fix released" there is a way [20:52] nealmcb: You need an entry in the changelog of the form LP: #xxxxxx. But please note that this marks the bug fix released. [20:53] slytherin: thanks - is that only for ubuntu? or can I set it up so launchpad does that for an independent software project? [20:53] nealmcb, thats only for ubuntu, for bazaar, you can use --fixes when commiting to do pretty much the same thing [20:54] nealmcb: I am not sure how it would work for other projects. [20:54] I tried that with bzr and it didn't work, and the folks in #launchpad seemed unsure. hmmm === _boto3 is now known as _boto [20:55] geser: do you have time to sponsor a debdiff for gstreamer0.10-plugins-bad? [20:55] nealmcb: i think if you 'bzr commit -m "foo (LP: #XXXXX)"' it works [20:55] nealmcb: oe if you use ppa's it certainly does [20:55] s/oe/or [20:56] Ng: around? [20:57] nxvl, if you upload to a project PPA with (LP: #XXXXX) in the changelog, it works on assiocated bugs? [20:58] NCommander: AFAIK yes [20:59] neat feature [20:59] NCommander: the LP bug number is not ubuntu specific [20:59] so if on a project you have #123456, you can be sure that is not in ubuntu or another project [20:59] nxvl, that I know, but I thought the changelog LP: feature was only good for non-PPA packages [20:59] i.e., that never works backports, we have to manually close bugs there [21:00] NCommander: well, for example with terminator, Ng releases it using the ppa, and has some bugs related to packaging that needs to be fixed [21:00] nxvl: hmm - what status does it end up with in a ppa - fix-committed? [21:00] NCommander: and those are terminator specific [21:00] nealmcb: released, just as with ubuntu [21:01] i can be wrong, but i'm almost sure it works that way [21:01] NCommander: but i don't think it can close bugs reported against ubuntu using a ppa [21:01] * NCommander feels like doing some packaging work [21:03] when a bug has multiple tasks, a package uploaded only closes against that task [21:03] yes [21:03] ...i ... think [21:03] nxvl: hmm - but a ppa isn't an official ubuntu repository, as described at https://wiki.ubuntu.com/Bugs/Status - right? [21:04] right [21:05] nealmcb: i'm not sure what you are trying to accomplish, but you can't change a bug reported against ubuntu using a PPA IIUC [21:05] nealmcb: you can manage project bugs using project's ppa [21:06] how does it figure out which task to close based on the PPA? there's no linkage between a project an a PPA as far as I'm aware [21:07] nxvl: ahh - good - so do you know what needs to be set up for that to happen for a project? [21:08] or where it is documented.... [21:10] nealmcb: not actually sure [21:11] nealmcb: Ng made it, and i just use it [21:11] :D [21:26] nxvl: ish [21:27] Ng: how is manage the LP closing of bugs with bzr/ppa in terminator [21:27] Ng: nealmcb is asking for that [21:28] :) [21:28] managed* [21:28] I don't know of a way to do it for PPAs, but I would love it if there is a way [21:28] You can't do that currently [21:29] but I've filed a bug request asking for that feature [21:29] Ng: really? we don't do that in terminator? [21:29] nxvl: I try to put the LP bug numbers in bugfix commits, but it doesn't make anything happen automagically [21:30] mm i thought it did [21:30] for me, it would be nice if upstreams in LP could do that, but also tie a release to a milestone such that creating the release makes all Fix Committed bugs for that milestone be Fix Released [21:31] but I am vaguely working on a launchpadlib script where feeding it a specific bug search URL will mark all the Fix Comitted bugs as Fix Released [21:31] Ng: from bzr we can't either? [21:31] nxvl: hmm, dunno [21:31] that's the problems of the weekend, the alcohol affects my brain [21:32] :S [21:32] I think not though [21:32] Ng: so what is --fixed for in bzr - hmmm [21:33] just a hook, with nothing on the lp backend? [21:33] ooh interesting, but my guess is that it just stores some metadata === nm-rocker is now known as asac [21:33] oops - that's "--fixes" on commit command in bzr... [21:34] with specific mention of launchpad.... [21:34] nealmcb, it works with other bugtrackers; its documented in the documentation how to use --fixes [21:35] NCommander: which ones does it work with? and what exactly does it do? the doc is vague iirc [21:35] Launchpad, Bugzilla, Trac, and I *think* one other type of tracker I can't remember [21:35] (maybe RT, but I'm drawing a blank) [21:36] * nealmcb re-reads http://doc.bazaar-vcs.org/bzr.0.18/configuration.htm [21:39] oops - google is still finding an old one. try http://doc.bazaar-vcs.org/bzr.1.3/en/tutorials/using_bazaar_with_launchpad.html [21:41] ... and I still don't understand what is written there.... === iron is now known as _1r0N === _1r0N is now known as iron