[00:05] * sistpoty needs to go to bed now... gn8 everyone [00:16] hey guys, question about the "runit" package on 7.10. No /etc/inittab exists on my system, so the install fails b/c runit wants to add an entry to /etc/inittab. Suggestions? [00:16] it's left in "iF" state instead of "ii" [00:18] check for its existence, create/modify conditionally, etc. [00:20] up_the_irons: you probably want to see debian/ in hardy's runit source package. [00:22] crimsun: if I make an /etc/inittab, how do i "fix" the package? should I just remove and reinstall? [00:23] up_the_irons: sure, you could do that. Or you could backport hardy's runit source package. [00:23] crimsun: ok gotcha [00:23] crimsun: yeah backporting it might be the way to go [00:24] thanks [01:25] Hi room! [01:26] :) well I am looking for a set of instructions that will guide me with making a .deb file for distribution. I am not a major programmer so I guess I would be a beginner with this stuff. Can anyone point me in a good direction. Most of the stuff Ihave found is so technical I cant grasp it very well. [01:30] whileimhere: https://wiki.ubuntu.com/PackagingGuide/Complete [01:32] Thanks! [01:33] np === victor___ is now known as victor__ [01:56] Heya gang [01:57] sispoty: yes [02:33] * RAOF wonders what his 'mail organ' is :P [02:35] RAOF: probably your postfix port ;-) [02:35] * jdong braces for more innuendos [03:15] * ScottK is here if someone is looking for an ack to upload (~6 hours to go). [03:17] ScottK: I'm available though cautious about what's going to be tagged to my name now :D [03:19] jdong: Do you know anything about using bzr? [03:19] yeah I use it a lot [03:20] jdong: Would you be willing to grab the fix in Bug #220385 from bzr and upload it? dholbach told me he'd take care it, but apparently didn't. [03:20] Launchpad bug 220385 in envyng-core "envy needs to pull drivers from multiverse, not PPA" [Medium,In progress] https://launchpad.net/bugs/220385 [03:21] I really don't want to release with something in the archive that allows non-developers to arbitrarily update end user systems. [03:21] Note who filed the bug if you have any doubts about we want it fixed. [03:21] ScottK: oh I have no doubt we want this fixed [03:22] lemme take a look at famed revision 45 :) [03:22] Thanks (or maybe there's a later one that's better, dunno). [03:26] ScottK: looks good, uploaded [03:26] jdong: Thanks. [03:44] jdong: Accepted into the archive. Thanks for taking care of that. === calc_ is now known as calc [03:49] ScottK: sure thing [04:20] jdong: Got time for another one? [04:20] I guess :) [04:21] If you look at Bug 220863 there's a patch in the linke sourceforge bug. [04:21] Launchpad bug 220863 in dkim-milter "dkim-milter rejects validly signed mail" [Medium,In progress] https://launchpad.net/bugs/220863 [04:22] I just discovered that the debian/rules don't appear to actually quite apply the patches (even though dpatch is at least partly there). [04:22] My brain is totally fried. [04:22] I'd greatly appreciate it if you would grab that patch, fix up debian rules, and upload it. [04:23] ScottK: sure, lemme take a look [04:23] Thanks. [04:29] * jdong checks his luck :) [04:35] applying patch 01-partial-signatures to ./ ... ok. [04:35] ah, beautiful [04:35] ScottK: ok, uploaded [04:36] jdong: Thanks againg. [04:36] again [04:38] * Hobbsee waves [04:38] * jdong waves back [04:40] sky fallen in yet? [04:42] Hobbsee: no but I am with a perl book right now :) [04:43] I think that's close enough [04:43] haha [04:43] true [04:44] Hobbsee: We ought to be nice to jdong just for this one day. He pulled the EnvyNG changes out of bzr and uploaded them, so that lowers his crack average some. [04:45] lol [04:45] ScottK: oh goody. [04:45] I am surprised that slipped past REVU though [04:45] jdong: i'm not. How many bits have you seen go in by canonical people, or reviewed by canonical people only, that haven't gone thru REVU? [04:45] and it doesn't help that tehy can't use openID [04:46] Hobbsee: yeah that is a bit concerning at times [04:47] jdong: but you can't bring it up - certain canonical people get very annoyed, and keep declaring that there is no difference between canonical people and community people, procedure-wise [04:47] jdong: so I guess that's here to stay. [04:49] * TheMuso waves, as he performs more iso testing. [04:50] uhm ubuntu seems vulnerable (by default) to fork bomb [04:50] Kewl. [04:50] emgent: that's not much of a vulnerability. [04:50] heheh i know [04:51] jdong: Accepted. Thanks again. [04:51] ScottK: not a problem [04:51] but i'm thinking to edit /etc/security/limits.conf [04:51] jdong: You may get the last two uploads for the release. [04:51] ScottK: haha is there a trophy for that? :D [04:51] emgent: if you are a sysadmin on systems prone to forkbombing that is a good idea [04:52] emgent: but otherwise, messing with limits.conf for all Ubuntu users is not a good idea [04:52] jdong: i know, grsec by default block it. [04:52] emgent: an Apache server can have 200 good behaving children, while I guarantee with 4 processes I can make you cry. [04:52] we can stop this if we add "* hard nproc 200" [04:53] emgent: you don't think 200 spinning processes can make your life miserable? [04:53] I only suggest, anyway no problem :) [04:54] emgent: my point is there really is no sensible default that won't impact some subset of our userbase [04:54] emgent: by default I think it's entirely reasonable that ubuntu grants whatever resource requirements a user asks for [04:54] emgent: though for the administrator of a multiuser system, more limits may be appropriate [04:55] * emgent nodes [05:53] Anyone around from motu-sru? [05:53] yeah [05:55] Last release (when there was no motu-sru), motu-release stepped up and said the "must be fixed in the devel release" rule was waived until the new repo was up and running. [05:56] It seemed to work out well. I recommend it. [05:56] I think that it's up to motu-sru to decide, so I thought I'd mention it. [05:56] If you are, I suggest you (motu-sru) make an announcement soonish. [05:57] Bug #158670 looks like it could still be sponsored if someone was up for it. [05:57] Launchpad bug 158670 in quodlibet "Quod Libet uses wrong playback device in Sound Preferences" [Low,Triaged] https://launchpad.net/bugs/158670 [06:02] ScottK: lemme take a look at that bug :) [06:14] ScottK: quodlibet debdiff mangled/uploaded [06:20] good morning [06:21] hi dholbach [06:21] hi ajmitch [06:21] hi Hobbsee [06:22] 'ello Hobbsee, ajmitch dholbach. [06:22] hiya RAOF [06:28] I am a novice programmer, familar with perl, c, c++. Never really worked on a large project. Is it reasonable to think I can contribute to Ubuntu? I read some of the wiki, but just wondering what someone of my level could do to contribute. [06:29] DanikarPN: that's great - you can definitely contribute to Ubuntu! :) [06:29] did you read http://wiki.ubuntu.com/MOTU/GettingStarted already? [06:30] Yeah, I read that and some of the articles it links too. Massive amounts of information. lol. Been looking into Ubuntu over the past week or so. [06:30] OT, is it a bad sign that I'm reading a Perl book and liking the language? [06:31] very much so [06:31] lol [06:31] * dholbach shudders slightly :) [06:31] I'm already starting to write unreadable 1-liners! [06:31] haha [06:31] That is what makes perl fun =P [06:31] makes it extremely handy for quick and dirty scripts for everyday tasks [06:32] .*/a?*!43.&^//:!.*^%!U [06:32] what makes that fun to read? :p [06:32] nixternal: it's a WOL [06:32] made that up by just pressing shift :p [06:32] (Write Only Language) [06:32] nixternal: the point is you hope you NEVER need it again after the 10 seconds you spend writing it and 30 seconds spent using it [06:32] heh, your 10 seconds would be my 10 hours [06:33] :) [06:34] * StevenK was employed as a Perl programmer for six years [06:34] /^.*?\.php.*?\?(.*)$/ That was mine for today [06:35] ^.*?\.php.*?\?(.*)$ That was mine for today [06:35] /^Pos:\s*(\d+)(?:.\d)?s\s+(\d+)f\s+\(\s*(\d+)%\)\s+(\d+(?:\.\d+)fps)\sTrem:\s+(\d+min)\s+(\d+mb).+\[([\d:]+)\]/) was mine [06:35] nice [06:35] (/^Pos:\s*(\d+)(?:.\d)?s\s+(\d+)f\s+\(\s*(\d+)%\)\s+(\d+(?:\.\d+)fps)\sTrem:\s+(\d+min)\s+(\d+mb).+\[([\d:]+)\]/) was mine [06:35] was mine. [06:35] mencoder CVS changed output format again [06:35] so acidrip.pm has to follow :) [06:35] It is very pretty. [06:35] lol [06:35] the backslashes are soothing [06:36] why he doesn't just write a perl binding to the actual API.... [06:36] I guess this is more fun :) [06:37] I guess this is a usage of the word 'fun' I was not previously aware of :). [06:39] It isn't fun till you throw something against a wall. [06:40] is it just me, or does every 2-character permutation of punctuation marks mean something in Perl? [06:40] ".!" [06:42] jdoing: haha, pretty much all of em probably yeah. [06:42] rofl [06:42] I suck at tab complete [06:43] and the default perl lint-checking is about as generous as the PL/1 compiler [06:43] i.e. you can sit on the keyboard and run it, and Perl wouldn't complain [06:44] jdong: How does it handle "if"? [06:44] My cat wrote a awesome perl script yesterday. === asac_ is now known as asac [06:45] RAOF: perl -e 'if(awefawjkf){ awejf; kawlej; kfaw }' [06:45] generates no errors whatsoever :D [06:46] something tells me this can lend to some ridiculous code bugs [06:46] use strict; [06:46] DanikarPN: yeah seems like it's a must-have for any serious more than 5 line codebase [06:47] Yeah, I had to debug about every script my company uses when I added use script up at the top. [06:47] There are three [06:47] use strict rather [06:47] use strict; use warnings; and running perl with -w [06:47] StevenK: yeah i use -w and use strict; didnt know about warnings. ill have to look into that [06:48] DanikarPN: -w and use warnings; might do the same thing [06:48] isn't there some perldiag thing that makes error messages more verbose too? [06:49] jdong: Soft. perl -e 'if' returns an error code :) [06:49] Line 7: It broke. [06:49] RAOF: yeah, compilation error at EOF [06:50] RAOF: so it isn't the PL/1 compiler. [07:29] Good morning === twanj_ is now known as twanj === _Czessi is now known as Czessi === Syntux_ is now known as Syntux [11:14] Hey [11:19] submitted Bug 220952 [11:19] Launchpad bug 220952 in ubuntu "Out of Range for 800x600 full screen games" [Undecided,New] https://launchpad.net/bugs/220952 [11:25] kahrytan: You may want to ask in #ubuntu-bugs. [12:24] hi [12:51] Heya RainCT [12:56] Hi Iulian === rzr is now known as rZr === fta_ is now known as fta [13:13] Hi! I'm trying to build the asterisk package (asterisk_1.4.11~dfsg-1) on gutsy using pbuilder, but 'pbuilder build asterisk_1.4.11~dfsg-1.dsc' fails with a few error messages about a broken h323 installation. See alsohttp://lists.digium.com/pipermail/asterisk-dev/2006-December/025345.html Could someone help me to build this package, I would like to patch it. [13:23] Noone? [13:25] flohack: we're very close to release, I suspect that everyone is just busy at the moment I'm afraid. [13:26] I see, that's fine. [13:26] flohack: do you have libopenh323-dev installed? [13:29] that should be a build dependency of asterisk, shouldn't it? [13:31] Yes it lists it as a dependency and pbuilder should install it [13:32] does it? [13:33] You're probably building with a version of $somepackage that is too ancient in gutsy. [13:33] !info asterisk gutsy [13:33] asterisk (source: asterisk): Open Source Private Branch Exchange (PBX). In component universe, is optional. Version 1:1.4.10~dfsg-1 (gutsy), package size 2033 kB, installed size 5264 kB [13:40] it's strange, apt-get source gives me a 1.4.11 package...why could that be? [13:41] I'll have a look if pbuilder installs the h323-dev package [13:43] yes it installs libopenh323-dev [13:43] I installed the pbuilder environment like this: sudo pbuilder create --distribution gutsy --othermirror "deb http://archive.ubuntu.com/ubuntu gutsy main universe multiverse" "deb http://archive.ubuntu.com/ubuntu gutsy-updates main universe multiversion" [13:45] I tried it without the gutsy-updates before, but that didn't work either [13:52] How can I retrieve a specific version of a source package using apt-get? sudo apt-get asterisk=1.4.10~dfsg-1 does not work [13:53] flohack: well, one of your problems is that you're certainly not trying to add repositories that exist. [13:54] other problems would be that the first "othermirror" you've set is already done by default (certianly should be), and that you've not told it that the second url is also supposed to be an othermirror. [13:55] ok, I'll rebuild the pbuilder environment... [13:56] I prepended a --othermirror directive to the second deb string. Do I have to remove the first othermirror? [13:57] I just can't figure out why apt-get source asterisk=1.4.10 does not work. [13:58] Ah, it is not present in http://archive.ubuntu.com/ubuntu/dists/gutsy/universe/source/Sources.gz anymore. [14:01] flohack: In your case it may be easier to set up a normal gutsy pbuilder, then login with --save-after-login and make the changes to sources.list manually. [14:06] Ok, I modified the pbuilder environment (added main universe and multiverse for gutsy-backports) and installed all updates. I'll retry the build in a minute === slangase` is now known as slangasek [14:16] Hobbsee: Unfortunately I still get the same error. [14:26] Would someone else be willing to try to build the asterisk source package on gutsy? [14:31] hi folks === sistpoty|work changed the topic of #ubuntu-motu to: https://wiki.ubuntu.com/MOTU | Want to get involved with the MOTUs? https://wiki.ubuntu.com/MOTU/Contributing | Universe for hardy is frozen - enjoy the break :) [15:16] * Hobbsee feels cold. [15:16] * sistpoty|work feels he got a cold :/ [15:17] ffffrozen? [15:17] Aww [15:17] * Iulian hides [15:17] * jpatrick has a cold [15:29] * Hobbsee hides in the freezer [15:32] * jpatrick gets from fishfingers from freezer [15:32] someone's cooking? [15:33] I wish [15:55] * siretart shivers [15:59] yay, frozen [16:00] crimsun: so, lets see how long we can go before someone asks The Question [16:00] are we there yet? [16:00] oops, too late [16:01] Hobbsee: "can i have an exception?" [16:01] nope to both of you [16:01] is hardy released yet? [16:02] * Hobbsee beats apachelogger with a big stick [16:02] meh [16:03] Nightrose: did you see that? [16:03] unbelivable [16:04] ;-) [16:04] what relations.. [16:04] ok, 3 OpenWeek sessions left open - who wants to do one? [16:04] pass [16:04] boo [16:04] * apachelogger points at Nightrose [16:04] ohnoes [16:04] :P [16:05] 1600UTC, May 1 - you know you want it [16:05] * Nightrose really is too busy with SoC [16:05] Hobbsee: what happened to the LPS™? [16:05] sebner just told me in #ubuntu-at he wants to talk about something [16:06] apachelogger: mind the release schedule :P [16:06] * Hobbsee pokes geser with the Long Pointy Stick of DOOM!!!!!!!!!!!!!!!™ [16:06] omg, it is still alive [16:06] * apachelogger hides [16:06] it is! [16:06] but it's not in kubuntu anymore! [16:07] lol [16:07] ouch === tb1 is now known as tbf [16:07] Kubuntu is dead. long live ubuntu xD [16:07] * Nightrose blinks [16:07] * apachelogger smacks sebner [16:07] has anyone noticed firefox chewing cpu lately? [16:07] crimsun: yes [16:08] crimsun: yeas [16:08] what's up with firefox? [16:09] apachelogger: *ouch* === danielm_ is now known as danielm [16:43] what's the sh way to do ${var%.*} (tu use it in debian/rules)? [16:45] what do you want to achieve? [16:46] geser: uhm.. nevermind, I've just realized that I can use sed for that (as I know the extension) :) [16:59] http://paste.ubuntu.com/7861/ - any idea why this doesn't work? [17:00] oops [17:00] the $ [17:00] line 5 [17:00] ^$, I should say. [17:00] RainCT: Also, all statements must be on one line [17:01] crimsun: yes that's what the "oops" was for, but it still fails with the same error [17:01] mok0: uhm.. right [17:01] yeah, you have missing \ [17:01] I tell other people about this but forget it myself.. :( [17:01] RainCT: hehe [17:07] Hmm. Why does lintian still complain over distribution name hardy. [17:07] RainCT: and instead of the sed call you could use either basename $file .po or ${file%.po} [17:08] mok0: Which lintian? What is the revision number of your package? [17:08] persia: 1.23.46 [17:09] And the revision? [17:09] geser: thanks, didn't know that basename can do that. ${file%.po} is a bashism, afaik [17:10] persia: there isn't any [17:10] mok0: How can a package not have a revision? [17:10] now there's still the problem that "for file in po/*.po" isn't working for some reason... :S [17:10] persia: native package? [17:10] hmm.... someone should write a tiny gtk wrapper for dget and associate that file with .dsc files... [17:11] ....that way browsers would do the right thing, when you click a dsc link [17:11] persia: ii lintian 1.23.46 Debian package checker [17:12] mok0: That would be the problem then. Native packages require "ubuntu" in the version to not generate the warning, but it looks ugly to have it there, so it only belongs when it is native in debian (and I think we ought use $(debian)-0.0ubuntuX rather than $(debian)ubuntuX in those cases, so as not to cause problems with native NMUs). [17:13] The trick is that lintian can't tell if it's an ubuntu-native package or a native package derived from debian, and so doesn't really know whether the string "ubuntu" belongs in the package name. Unless my suggestion for native package version increments is adopted, we'll always have that bug for native packages. [17:14] Personally, I think the best solution is to abolish all native packages, but that is often considered excessive. === calc_ is now known as calc [17:14] persia: I agree, native packages tend to cause confusion [17:15] persia: ... and you still may want to update the packaging in which case the revision is relevant [17:15] mok0: Right. Especially for derivatives. [17:17] And if we are going to use native packages derived from other distros, we ought give them revision numbers that match NMU conventions: currently 1.1ubuntu1 > 1.1-0.1, which is unfortunate when looking at things that ought be merged, as it is probably important (or it wouldn't be an NMU), but it doesn't show up on the default lists. [17:17] (hence the 1.1-0.0ubuntu1 revision scheme previously mentioned) [17:18] tbf: except that one may not want clicking dsc links to invoke dget [17:20] any idea why "for file in po/*.po" doesn't work? (the same in a sh script works) [17:21] crimsun: you know them meaning of the word "wrapper"? [17:21] (http://paste.ubuntu.com/7865/) [17:22] RainCT: $ -> $$ [17:22] RainCT: you need to use $$file [17:22] crimsun: that wrapper would allow you to choose between "download related source code" and "download dsc file only" [17:23] crimsun: also that wrapper would have to ask you for the proper folder [17:23] RainCT: make is not shell. Try $(wildcard po/*.po) [17:23] crimsun: it's just boring to copy the link, open a terminal, and past the link there [17:23] sistpoty|work: where? [17:24] crimsun: also dget is not that easy to find [17:24] tbf: which is a pretty nasty UI workflow. [17:24] RainCT: everywhere, where you want a $ in shell code written in a makefile (having shell code in make is expensive, hence you need to give it more $$'s :P) [17:24] sistpoty|work: hah! [17:25] RainCT: Also, don't confuse your make variables ("$(foo)") with your shell variables ("$$bar") [17:25] crimsun: the current one, with no dsc support at all? yes! [17:27] ah, I see [17:28] great, works now. thanks :) [17:28] RainCT: ${file%.po} works also in dash (see the dash manpage) [17:29] Aren't dashisms just as frowned upon as bashisms? Anyway, even better if you can do it all in make. [17:29] that's POSIX [17:30] has dash any extensions beyond POSIX? [17:30] uhm.. true.. dunno how I tried it before then that it didn't work xD [17:39] geser: dash supports several XSI extensions, I believe [17:44] but nothing that is beyond what Debian/Ubuntu expects from /bin/sh? [17:45] geser: I think that's supposed to be the standard for what dash implements, yes [17:49] * sistpoty|work heads home [17:49] cya === Spec[x] is now known as Spec [18:08] Heya gang [18:22] Hi bddebian [18:22] 'lo bddebian [18:22] Hi geser, crimsun === kitterma is now known as ScottK2 === nixternal is now known as lanretxin === lanretxin is now known as nixternal [19:53] hola [19:57] norsetto: I'm cleaning up the motu-release bug queue and wondering if Bug #199190 not getting done means a piece of the gfortran transition got missed? [19:57] Launchpad bug 199190 in libitpp "[FFe] [Sync request] libitpp 4.0.3-1 from debian unstable" [Undecided,Confirmed] https://launchpad.net/bugs/199190 [19:58] * norsetto checks [19:58] Thanks. [19:58] i recently started experiencing massive slowdowns with firefox 3.0b5 [20:00] scottk: I thought that was fixed, is the bug that started the whole mess [20:02] scottk: apparently it was never sponsored since u-u-s was never subscribed [20:05] norsetto: I invalided that bug since it's to late to sync, but it sounds like an SRU is in order. [20:05] ScottK2: don't look at me, I will never do an sru again [20:07] OK. [20:07] norsetto: Would you beat on sistpoty to do it then? [20:08] norsetto: Is a rebuild enough? [20:08] scottk2: I can check, but I'm afraid not [20:08] If that's all it needs, then I'll handle it. Please check. [20:10] scottk2: sure, let me finish with quodlibet [20:13] Great. [20:16] Bugs related to MOTU Release Team: "There are currently no open bugs." \o/ [20:18] i'll prepare an SRU soon, so that might change :) [20:19] :D [20:22] laga: That's motu-sru. Not my problem any more. We have a different batch of bureacrats for that. [20:23] ah. too bad. maybe for intrepid then ;) [20:38] Bugs related to MOTU Release Team: "There are currently no open bugs."-> You guys (scottk, sistpoty) really deserve a HUGE THANKS for the wonderful work you did [20:38] :) [20:39] I think we did significantly better this time around than for Gutsy with each request at least getting considered and not being a block on progress. [20:40] Not idea, but better. [20:40] ideal/ideal [20:40] scottk2: much better (at least from what I could see on gutsy) [20:41] ScottK2: Is lighttpd security update going to make it into release? Or saving for post... [20:41] Did we have one we missed? [20:41] it's on the rcbugs list [20:41] emgent: Do you know? ^^^ [20:42] SSL vulnerability [20:42] I thought we got that one, but not sure. [20:42] ok [20:42] At this point it's all post release. [20:42] uhm just a moment [20:42] np, ty [20:43] i released sime week ago one fix for SSL bug [20:43] s/sime/some/ [20:43] Is the fix in the Hardy package? [20:43] yes [20:43] Coll. [20:43] Cool even. [20:43] milli: There you go. [20:44] emgent: Thanks. [20:44] thanks guys [20:44] should mark that on the rbugs page :) [20:44] ScottK2: https://edge.launchpad.net/ubuntu/+source/lighttpd [20:44] np :) [20:44] milli: Feel free. [20:44] CVE-2008-1531 [20:45] fixed in lighttpd 1.4.19-0ubuntu3 [20:45] (hardy) [20:46] Excellent. [20:46] norsetto: Had a chance to look into libitpp yet? [20:47] scottk2: I'm uploading a debdiff to bug 199190. Builds ok and seems to be using gfortran, but needs further checking. If you want, you can mentor the SRU, should be a good task for a new contributor [20:47] norsetto: OK. Please take invalid off the bug then. [20:48] ScottK2: no need, is a different task [20:48] Ah. [20:52] norsetto: Your task is for Gutsy. We need Hardy, don't we? [20:52] scottk2: ops [20:52] * norsetto tries to reset his internal release timer === DanikarPN is now known as Danikar [20:56] ScottK2: there you are [20:56] Thanks. [20:58] norsetto: Do we have to rename the binary packages? [20:58] ScottK2: yes [20:58] Urgh. [21:05] norsetto: RM has authorized upload to hardy-proposed. Please change the revision to ubuntu0.1 and the release to hardy-proposed and upload. [21:05] So much for paperwork. [21:06] scottk2: oh well [21:07] scottk2: I'm not following that up with the usual mess of testing request etc. etc. though [21:07] K. We'll make sistpoty do that. [21:08] I am the developer of unicap, a video capture SDK, and would love to see my software available via universe. Is there another way to get it packaged than to become the package maintainer by myself? [21:10] arf [21:11] arne_: what mok0 said :-) [21:11] arne, you can file a needs-packaging bug at launchpad.net, including all the information on where to pick up the sources etc. And then hope that someone picks it up [21:11] norsetto: :-) [21:11] arne_: bribes are happily accepted (mok0 has a penchant for rare beers for instance) [21:12] hehe [21:13] arne, in the meantime, pls take a look at this (incomplete) Upstream Guide: https://wiki.ubuntu.com/UpstreamGuide [21:13] should I take this like "the chances are not very high that someone will pick it up"? [21:14] that depends on how persistent you are :-) [21:15] Another piece of software where the author did the same got packaged within ~ 1 month [21:16] arne_: software in your particular field seems rather popular [21:17] My issue is that I am already quite swamped with work [21:17] currently I maintain my own repository [21:17] but since it has a lot of hits, I think it should rather go to universe [21:17] arne_: so, you have it packaged already? [21:18] yes but I do not know how well it is according to ubuntu guidelines [21:18] and I have too little time to read all the docs and check it by myseld [21:18] *myself [21:19] arne, upload it to revu, and it will get reviewed [21:19] arne_: if you have a source package already, you can also consider debian-mentors, we will get the package from them afterwards [21:21] ok, I will try with revu [21:21] thanks for your help! [21:23] arne_: but file a needs-packaging bug too. See bug 200783 as an example [21:23] Launchpad bug 200783 in ubuntu "[needs-packaging] ChibiTracker -- a portable impulse tracker" [Wishlist,Confirmed] https://launchpad.net/bugs/200783 [21:24] mok0: ok, will do [21:25] arne_: great! Thanks for your contribution [21:30] superm1: I see mythbuntu-control-centre (source) 0.28-0ubuntu1 in the unapproved queue. Do you need that? [21:32] ScottK2: yes. [21:32] that was the SRU i was going to file [21:33] laga: OK. Let me talk to the RM about it. [21:34] arne_: does it have python bindings? [21:34] no [21:35] :-( [21:35] unfortunately not [21:35] I once worked on them but did not have the time to finish it [21:35] arne_: also, check it with lintian (both the source and binary packages), this will give you quite a lot of hints already on things to be worked on [21:36] arne_:so which language does it support? [21:36] C [21:36] http://www.unicap-imaging.org [21:38] norsetto: thanks, I will check it out [21:38] arne_: ok,thans [21:38] arne_: if you have problems to understand what lintian is telling you (it can be cryptic sometime), just ask here [21:40] arne_: and a last tip, pay particular attention to debian/copyright; since you are upstream, if you have all sources with a clear header stating license and copyright, this will make things much smoother [21:41] norsetto: the license of all files should be clear [21:41] arne_: perfect then === __Czessi is now known as Czessi [21:43] scottk2: package uploaded to hardy-proposed, now forget about me .... [21:43] norsetto: Thanks. [21:48] Any news on when intrepid opens for development?? [21:48] can we finish releasing hardy first, please? :) [21:49] mok0: normally it takes 2 weeks IIRC [21:49] so we'll be stalled for 2 weeks? [21:50] mok0: bugs are always and everywhere ;) [21:50] sebner: yeah [21:51] mok0: but you are totally right. bleeding edge is the only true thing =) [21:51] sebner: ... a lot of the bugs will be fixed by syncs and merges, anyway [21:51] mok0: but that are not SRU [21:52] sebner: true [21:52] mok0: and yes. syncs and merges are my profession, my pleasure :D [21:53] sebner: :-) [21:54] mok0: You can review and comment on the backlog on REVU in the mean time. [21:54] ScottK2: good idea :-) [21:54] Gotta go. See you later. [21:54] See you! [21:55] hf [21:56] sebner: did you manage to get your merges uploaded for hardy? [21:56] mok0: define *my* merges [21:57] sebner: the ones you instigated :-) [21:57] mok0: in what timeframe ;) [21:57] last week [21:58] you had a whole bunch from the rc queue [21:58] mok0: hmm. I have 2 ones that are 3 weeks old ^^. Scott already added a comment for one "Please update for intrepid". Last few weeks I did a little little bit bugfixing but had to learn mostly for school [21:59] sebner: eeek [22:00] mok0: but also with your help I think all the RC things got uploaded :) [22:00] sebner: cool [22:01] mok0: and now I'm keen to merge and sync cool new stuff ^^ [22:01] cool: sry ^^ [22:01] s/cool/hot [22:03] sebner: if you really have nothing else to do, why don't you fix bug 221171? [22:03] Launchpad bug 221171 in xchat-xsys "xchat-xsys displays wrong distro" [Undecided,New] https://launchpad.net/bugs/221171 [22:03] norsetto: if nothing == school then yes ;) [22:03] norsetto: isn't that the one that depends on a non-existing package? [22:04] sebner: school teaches you nothing nowadays, yes [22:04] norsetto: or rather, a library not packaged [22:04] mok0: yes, it was [22:04] norsetto: xD. true. but I have final exams soon so I have to learn that stuff that also if it makes no sense ;) [22:05] sebner: final exams, in APRIL!? [22:05] norsetto: 2 weeks. and now we are learning and repeating the stuff from 4 years .. [22:05] sebner: well, so you will have a lot of time after april :-) [22:06] norsetto: hmm. 1 month to learn for my oral exams [22:06] in june .. [22:07] norsetto: but yes. I already thought about it to start uploading something on REVU during this month :) [22:09] sebner: if you really want to package new stuff, you might as well do something usefull, why not tackling bug 133784? [22:09] Launchpad bug 133784 in saods9 "saods9 is way out of date wrt upstream (4.0b7 vs 4.13)" [Wishlist,In progress] https://launchpad.net/bugs/133784 [22:09] sebner: there are 726 needs-packaging bugs... ;-) [22:10] mok0: yeah ^^ [22:10] norsetto: I'll file a bug in BTS. Merging is a lot more fun :P [22:11] sebner: don't bother with that, the package is orphaned [22:11] norsetto: then remove it ^^ [22:11] sebner: too bad, doing that package could also mean adopt it in debian and be on your way to be a dm [22:12] norsetto: we'll see. but for the next few weeks I don't feel like to become a dual maintainer ^^ [22:14] sebner: actually, what are you doing here instead of studying!? [22:14] norsetto: what procedure do you recommend for getting packages -> debian? upload to mentors? [22:14] norsetto: talk with some friends and the motu family to relax and calm down before going to bed ;) [22:15] mok0: yes, but it helps if you can also cc a team (if there is one which is relevant) [22:15] mok0: for instance, debichem for chemical packages, etc. [22:16] norsetto: I am involved with the debian-med team, but not all packages I maintain fall in that category [22:17] mok0: what I mean is that your chances are much higher if you find somebody which is interested in that package [22:17] norsetto: btw. becoming a maintainer wouldn't be the biggest problem but since I have no relevant programming skills at all I wouldn't be able to fix *any* bug. and always annoying upstream ... :( [22:17] sebner: I have no relevant programming skills too, but I'm good at annoying upstream :-) [22:17] sebner: learn some programming! [22:18] mok0: that's very easy to say ;) [22:18] Read Kernighan & Richies book on C programming, the best programming book ever written, and it's thin too [22:18] norsetto: not true. what about your conky fix [22:19] sebner: yes, what about it? you need programming skills for that!? [22:19] mok0: well. we'll see but to be honest. For *now* I'm happy with merging and syncing what doesn't that a lot of time [22:19] * mok0 sighs [22:20] norsetto: sure. I and a lot of other people wouldn't have had any clue how to solve .. [22:20] * norsetto shrugs [22:20] motu's are geeks [22:20] the ubuntu users are right xD [22:20] ^^ [22:21] Geeks have more fun [22:21] compared to? [22:22] ^^ [22:22] laga: non-Geeks :-) [22:22] lol [22:22] mok0: surprise surprise, geek is not in my collins [22:23] he [22:23] +h [22:24] sebner: bug fixing is problem solving. It has surprisingly little to do with programming when one considers. [22:25] crimsun: but it helps being able to see what the code does [22:25] crimsun: depends. for segfaults, it can matter. [22:25] mok0: absolutely. [22:25] crimsun: sure but if you have no knowledge in C for example you can't write that much new code or something similar [22:27] well, even if you can read code [22:27] it takes a lot of knowledge on how to fix it [22:27] pwnguin: true [22:27] worse, the code is often poorly documented, so you don't know nessecarily what the right fix would be [22:28] pwnguin: Go for the one you think is the best ... There might be a lot of best fixes :) [22:28] ie, in xkb, i found a function returning null in some cases, and a specific instance where it was not checked before dereferencing [22:28] even "worse", sometimes the "correct" fix constitutes a significant regression. Hooray for legacy support! [22:29] the saddest part of fixing bugs is having to juggle conflicting policies [22:29] most segfault bugs are due to null pointers, so if you can point out to upstream where and when that happens, it can be a great help [22:29] its not clear what returning null meant as the conditions were incredibly commplex, but apparently upstream decided the fix was "if null, return" [22:30] pwnguin: if the code attempts to dereference a null pointer, it is definitely a bug :-) [22:30] yes, but do you stop returning null, or do you simply ignore it in that case? apparently ignore it [22:31] another example would be liferea, where adding MIME to the parsers is simple, but making the program handle them might need six months to understand [22:32] pwnguin: if the pointer is being dereferenced, it is not supposed to be null _ever_, and so it must be a bug [22:32] fortunately, packaging doesn't involve writing new code [22:32] so that argument's out [22:32] pwnguin: shell code? [22:32] pwnguin: but bug fixing ;) [22:32] shell code? [22:32] oh [22:32] pwnguin: that's programming... [22:33] true enough [22:33] mok0: I don't even know that xD [22:34] sebner: neither do I lol [22:34] mok0: but you can programm [22:34] sebner: yeah [22:34] mok0: language? [22:35] fortunately, liferea decided to implement mime type support a few months after i filed a bug about it [22:35] My favourite ones are C, C++ & Python [22:35] Scientus: any news on mythbuntu-control-centre? [22:35] err, ScottK [22:35] sorry :) [22:36] laga: ScottK took off a while ago [22:36] mok0: I don't even have time to learn 1 language (what means a deeper knowledge of it) [22:37] if you dont have time to lern a language, where's the time to write packages coming from? [22:37] sebner: if u start messing around with code you would probably be suprised how fast you can know it. [22:37] sebner: just read K & R. Really. [22:37] does K & R use K&R synatx? [22:37] mok0: C. I hate C. The code is compared to python or C# so complicated ^^ [22:37] * pwnguin fails at typing today [22:38] Danikar: hmm [22:38] pwnguin: I think there's a new version that uses the ANSI syntax [22:39] pwnguin: but the K&R book teaches you everything you need to know about writing programs [22:39] sebner: I'd learn C or C++. C# and other higher level languages are nice but you really start to understand stuff with C and other languages liek htat. [22:39] K & R is a really good one if you have the basics down already. They go kind of fast. [22:39] Danikar: I started to learn c++ one or 2 years ago. after 1 month I gave up [22:40] sebner: C++ is stupidly complicated [22:40] ^^ [22:40] sebner: C is a lot easier than C++ imo. But you can't give up, it will be hard sometimes, just have to get through it. [22:40] But C++ is a superset of C, so learning C will help you learn C++ [22:40] Let's start a flameware [22:40] And learning Python will teach you about oo prgramming [22:41] mok0: Learning C++ was brutal, my C++ teacher just told me everything I did was wrong cause it was too C like. lol [22:41] mok0: i like printf too much. [22:41] lol [22:41] Danikar: heh [22:41] * sebner has to go to bed now. tomorrow is another challenge in school -.- [22:42] Danikar: learn OO programming from Python, then C++ becomes much easier. A lot of the STL stuff enables you to use C++ in a "Python-like" way [22:42] See you sebner [22:42] bye [22:42] mok0: I am writing some PHP5 stuff for work, so I am learning classes that way. They are pretty similar. I have never touched python, but I definetly want to look into it when I get the time. [22:43] mok0: i figure C++ is on its way out. STL is nice but i find it quite ugly and unintuitive at times [22:43] pwnguin: on it's way out? [22:43] yea. unless you think it's already out [22:43] pwnguin: what will replace it, C#? [22:44] My problem with programming is when I look at a real programs source code I am just lost. Don't know where to even start to find a bug. [22:44] c#, java, python, ML, etc [22:44] Danikar: usually one starts with a backtrace [22:44] C# is obknoxious Microsoft stuff [22:44] pwnguin: What is a backtrace? [22:45] hmm. a back trace is a bit like a path from the start of the program to how it crashed [22:46] its a dump of the stack, indicating which function called which etc that caused a serious error [22:46] " And learning Python will teach you about oo prgramming", I would rather say "teach you about how to make a program crash in about a zillion different ways" ;-) [22:46] pwnguin: So you manually go through it like that? [22:46] Danikar: you can use a debugger [22:47] gdb, for example [22:47] Danikar: you can load it up in gdb and if the program was compiled right, it'll tell you that info, including what lines of source code [22:47] at that point its pretty easy to see where the error happened, though determining what SHOULD happen instead is the hard part, like i said originally [22:47] ... and you can analyze the values of all variables at the time of the crash [22:48] well, most variables [22:48] pwnguin: Can you use that for fixing things that don't necessarily cause the program to crash? [22:48] right [22:48] depending on how bad the crash was ;) [22:48] Danikar: gdb is quite advanced, and will let you set "breakpoints" in code [22:48] I have never really looked at gdb. I just printf("Hello: MSG#123\n"); Till i locate the crash lol [22:48] Danikar: you can step through the program one instruction at a time [22:49] * pwnguin has yet to find a decent GUI frontend to GDB [22:49] Ill have to look into gdb then, well documented on how to use? [22:49] norsetto: Learning python (the right way) well teach you the joy of unittests :) [22:49] gdb is an awesome tool. There are also guis for it [22:49] does anyone know a nice perl debugging tool? i'd love to look at variable sometimes i step through my programs [22:50] doesnt perl have an interpreter? [22:50] pwnguin: that's just not the same ;) [22:50] surely ther's a perl onestep or something [22:50] RAOF: yes, the keyword is "the right way" [22:50] norsetto: Quite true. You don't have a compiler, so without unittests you don't have an automated stupidity checker :) [22:51] mok0: a gui for gdb!? horror! [22:51] mok0: what next, an IDE? (hehehe) [22:51] norsetto: hehehe [22:51] I think it's called emacs :P [22:51] pwnguin: looking at the man page for perl, there is indeed some nice stuff like "use diagnostics" which will print nicer error messages [22:52] laga: there's a package called perl-debug [22:53] laga: my only experience with perl was a bioinformatics class project where we borrowed a SNP analysis tool written in perl. after reading the code, i'm not sure i can safely eat tomatoes anymore ;) [22:54] pwnguin: yeah, perl can be quite read-only [22:54] err, write-only [22:54] heh, this was more than that [22:54] the termination condition? [22:54] Divide by zero [22:54] okay.. [22:55] anyways, I'm not sure how i feel about teaching biologists programming [22:56] Ok, so I can see why the programmer might want to write a SNP analysis tool in something that's not C, but doesn't it want to be really, really fast and parallel? [22:56] you're falling in to the trap of assumed competance [22:57] perl was clearly chosen because they new a bit about it [22:57] not because they were picking the right tool for the job [22:57] And it's not as totally rotten at string manipulation as C, presumably. [22:58] well, technically the best regexp engines are written in C [22:58] because perl regexps arent regular [22:58] Actually, Perl is written in C :-) [22:59] anyways, if watson can discover the structure of DNA without knowing chemistry, then i guess biologists can also write programs without knowing computation [23:01] pwnguin: it sounds so convincing... but somehow I have the feeling that something is wrong with your logic [23:03] pwnguin: we actually have "proper" bioinformatics degree programmes here [23:03] that's good [23:03] eg put them through programming 101, then try to teach the remaining 50% something more useful than standard ML ;) [23:04] we just have biology students who are just required to take some bioinformatics course. apparently the other course they offered was "harder" because it had lots of programming in perl [23:04] maybe there was some self selection at work there, but somehow I feel that biologists often choose the role because they hate math [23:07] welcome HARDY! [23:08] pwguing: funny, here biologists teach math in high school ... [23:08] well, good night [23:08] * mok0 wonders if the bittorrent trackers will be working [23:08] RainCT: g'night [23:09] norsetto: well, I live in Kansas [23:10] (an agrarian and highly conservative christian region of the US) [23:10] pwnguin: hmm, is that where they still teach creationism in high schools .... (no offense intended) [23:11] pwnguin: ok, I think that was the answer :-) [23:11] well, its a bit more complicated than that. its where some people would like to teach it to their children, but the courts have ruled it religious dogma and not science [23:11] for a second i thought you guys were still talking about perl [23:12] "but the courts have ruled it religious dogma" [23:12] laga: perl, religion, about the same stuff :-) [23:12] pwnguin: fortunately someone is still sane... [23:12] every ten years, once the agitators think the people's guards are down, they run for School Board, and try to pass their laws. then they all get voted out next election [23:12] Yes, computer topic and religon cross more often than u'd think =) [23:12] mok0: well, evolution is still required teaching, no matter [23:13] Mac Cults, Windows Crusade, The Light of Linux? [23:13] creationism is a good thing to be taught. it's part of the cultural background of many people. but it should be taught in religious ed class and not as a science [23:13] mok0: every time this happens, the university board of regents threatens to require evolutionary theory to enter university [23:14] the real problem i think is the order in which we present the sciences, and the emphasis we misplace in math [23:15] math should be more important than it is, and the order should be physics, chemistry and then biology. because there's a very real chemical component to biology that is often missed [23:15] pwnguin: I agree [23:16] * norsetto heads to bed [23:16] night everybody [23:16] pwnguin: I would also teach people the principles of science [23:16] pwnguin: math is important, if there's an application for it. i always disliked math, but now i have applications for that knowledge and i enjoy learning about it [23:17] laga: i think mathematicians are discovering that no matter how hard they try, they cant come up with a mathematical domain that has no applications [23:17] laga: yes, the traditional math teaching with proofs, etc. turn many people off [23:17] heh, i love proofs [23:17] people who can't do proofs turn me off [23:18] People who can't do math are usually the same people who are ok with being stupid. I hate those people. lol [23:18] Danikar: slam slam slam [23:18] my roommate's a computer engineer who got upset when he got a D in Programming Logic. apparently it his firm belief that logic begins and ends with NAND [23:18] Its not bad being stupid, it just bad to be ok with being stupid. =) [23:19] Danikar: hehe [23:19] mok0: The problem is that you _have_ to understard proofs to think mathematically :) [23:19] RAOF: I disagree, actually [23:19] You can learn a lot by studying applied math [23:19] pwnguin: I don't trust software written by computer engineers. :) [23:20] mok0: we call that science [23:20] pwnguin: sure. the difference is how it's presented in school. i spent thirteen years in school, and most of the math stuff there was boring because it was theory without any obvious practical application (of course there are lots of applications) [23:20] slangasek: well, my other roommate's an EE, so it gets pretty bad [23:20] mok0: Maybe I can restrict the domain of the statement 'think mathematically' to make it a tautology :) [23:20] laga: Just the fact of doing arbitrary problesm helps you just think more logically about other arbitrary real life situations. [23:21] laga: somehow, i cant see your argument going anywhere besides more word problems in Algebra class [23:21] I thought word problems were only on amd64? [23:21] sorry that was a bad pun [23:21] hah [23:21] ba-dump-ching [23:21] jdong: lol [23:21] pwnguin: ok, yes, EEs are worse [23:22] mok0: You certainly can learn a lot of stuff by messing around with applied maths; but to really _grok_ maths you need the kind of pedantic thinking that results in the 'black sheep' joke. [23:22] Danikar: i don't doubt the benefits. i'm trying to make the point that i found it boring as hell when it clearly doesn't have to be that [23:22] there's plenty of fun abstract math games [23:23] gplanarity is in universe i think [23:23] RAOF: I can use math without knowing the proof that it works. If I trust that someone has developed the theory consistently. [23:23] RAOF: The same as I don't need to know how to write a compiler to use one [23:23] mok0: Quite true. But I wasn't really talking about _using_ it, I was talking about 'thinking mathematically'. [23:24] RAOF: I think a small minority of students have an interest in that [23:24] mok0: but it certainly helps life to be able to write your own compiler [23:24] mok0: unless you intend to live life with only regexp [23:24] pwnguin: heh [23:25] note: regexp is bad for lots of problems [23:25] xml + regexp = fail [23:25] Life is grepping for the true regexp [23:26] so -mote is pretty slow during the freeze i guess [23:26] err, -motu [23:26] Eh; regular languages are boring. It's all finite-state-machines or semigroups :) [23:26] laga: Heh yeah I guess I can see how it could be boring. I really like logic/mathematics so I dont get that bored in it. [23:26] RAOF: semigroups? [23:27] Danikar: imagine being caught in a liberal arts class. ;) [23:27] They're like groups, but without the inverse. [23:27] laga: lol [23:27] i love wikipedia [23:27] RAOF: Ah ok [23:27] "in other words, a semigroup is an associative magma" [23:27] laga: I was a Philosophy major untill I relized that most philosophy is centralized on ethics. So I changed to Computer Science/Mathematics [23:28] Danikar: respect [23:28] Some might call that 'applied philosophy' :) [23:28] heh [23:28] (By that, I mean 'ethics', not maths) [23:28] laga: I just take philo calsses for my general ed, but they are very interesting / fun. [23:29] sure. [23:29] "Ethics" is the moral of those in power [23:29] so what of the ethics of open source? [23:29] RMS? [23:29] mok0: Heh Ethics is interesting, just not what I am interested in. Laws are the Ethics of those in power. =) [23:30] pwnguin: the moral of the leading figures in open source [23:30] i always like languages and all that liberal arts stuff better than math, but i also liked computers a lot, so i ended up in computational linguistics. :) [23:32] laga: that's a hot field [23:33] mok0: yes. very interesting, IMHO, and you get to play with some nifty stuff [23:35] Heh had to go look up on wiki what exactly that is. =) It does sound interesting. The one thing that I dislike about this world is that I wont have the time to learn everything. But damned if i wont try. [23:35] good point :) [23:36] * mok0 is trying to remember what famous software package is written by a linguist... [23:37] Is it Perl? [23:38] yeah, wikipedia tells me that larry wall used to study linguistics [23:39] Hooray, my brain still works :-) [23:40] /claps [23:43] I still think Larry Wall's most impressive creation is patch [23:43] patch is awesome [23:45] We wouldn't have Ubuntu without it :-) [23:46] unified diffs make me happy