21:01:04 #startmeeting Spins SIG 21:01:05 Meeting started Mon Feb 15 21:01:04 2010 UTC. The chair is kanarip. Information about MeetBot at http://wiki.debian.org/MeetBot. 21:01:06 Useful Commands: #action #agreed #halp #info #idea #link #topic. 21:01:13 #topic Roll Call 21:01:16 * kanarip here 21:01:16 * brunowolff is here 21:01:22 * nirik is sorta around, but working on work stuff too. 21:01:26 biertie, ping 21:01:41 right 21:01:47 * biertie is here 21:02:15 alright 21:02:19 i guess maxamillion is here too ;-) 21:02:26 hi hi 21:02:29 no, he's faking 21:02:54 #topic Agenda 21:03:04 Moblin merge? 21:03:11 Games spin status 21:03:13 * sdziallas hollers 21:03:15 lzma update 21:03:20 mini ks? 21:03:27 * sdziallas shuts up, listens 21:03:47 and Makefile, autofoo 21:04:12 #topic Moblin Merge 21:04:46 (side note: was that as way out of left field as I thought it was or did someone know something and I'm just out of the loop?) 21:05:15 I saw it happened and was wondering if it was going to have any significant affect on the spin? 21:05:32 Do we need to change the name or worry about trademarks again? 21:05:38 probably no 21:05:39 not 21:05:53 has it been approved by the board carrying the moblin name? 21:05:53 should we ask legal about it? 21:06:03 yes maxamillion 21:06:09 I haven't talked to Peter for some days lately. Not exactly sure what his plans are... 21:06:11 but moblin 2 stays moblin 21:06:42 so since meego isn't really released yet, and we don't use it yet 21:06:46 I don't think there is a problem 21:07:04 I suppose we might also want to check and see if there has been any code changes and see if there are more bits that need packaging, or if there is a new build needed, or possibly $other 21:07:20 well yeah, that's a good point 21:07:29 meego will be qt based 21:07:36 while moblin is gtk based, right? 21:07:47 afaik, yes 21:07:51 biertie: yes 21:08:03 but i'm probably the least familiar with moblin/meego here, too ;-) 21:08:04 so its going to be a complete rewrite or fork 21:08:32 so, this requires no action or intervention then, for Fedora 13? 21:09:21 I don't think moblin 2 will change, no 21:09:28 ok 21:09:30 #agreed No action pending on the Moblin Spin (merge) 21:09:31 they didn't announced that anyway 21:09:40 #topic Games Spin status 21:09:45 yeah, I can't see moblin2 changing at all since it is already released 21:09:57 brunowolff, vous avez le donner de la parole 21:10:08 It's building again and I have been running size tests in the nackground yesterday and today. 21:10:26 It looks like I will need to drop a big game for F13. Probably alien arena, 21:10:34 I'll commit something tonight. 21:11:15 I may put openoffice back to match desktop or I might stick with abiword. 21:11:22 which will probably leave the games spin in very good shape for F13, as far as the size is concerned, no? 21:11:35 dropping a big game like alien arena, that is? 21:11:44 Yes, alien arena is over 250 mb. 21:12:08 anything more on the games spin? 21:12:21 Not directly move on to lzma update. 21:12:37 #topic lzma update 21:12:44 Good news there. 21:12:50 brunowolff, vous avez le donner de la parole (encore) ;-) 21:13:00 Full details are in https://fedoraproject.org/wiki/Features/LZMA_for_Live_Images 21:13:20 But in short Lougher added an xz wrapper that works with fedora. 21:13:35 My test showed a 10% reduction in size of the games spin. 21:13:58 I am still looking at F14 or an update for F13 not release for this. 21:14:08 that's sweet! (10 %) 21:14:12 maybe also mention on the LZMA feature page, the current size of the games spin 21:14:39 just to give people an idea of the importance of trimming the squashfs image by 10% 21:15:14 OK. At the time it was from about 4.3 to 3.9 GB. 21:15:42 I can use the rough numbers now and fill in some more accurate ones later. 21:16:37 I need to hear back from Kyle, but squashfs-tools should go into rawhide very shortly after the branch. 21:17:12 It will be about a month for the 2.6.34 kernel, depending on where fedora kernel development goes. 21:17:37 They might concentrate on 2.6.33 since that will probably be used for the F13 release. 21:17:51 i think that's the case, yes 21:17:58 I'll make up an enhancement proposal for livecd-creator. 21:18:11 we encountered some exotic issues with the 2.6.3x series in our respins as well 21:18:24 Basicly I want the default to be lzma with an option to change it. 21:18:55 That's the current status. 21:19:00 thanks brunowolff 21:19:10 good work on this brunowolff 21:19:20 #topic Mini kickstart 21:19:28 huff, are you around? 21:19:42 * kanarip has not seen the mini ks in spins-kickstart yet 21:20:00 * sdziallas neither, but would love to - is there anything we can do to get this done soon-ish? 21:20:33 yes of course, just include the mini.ks 21:21:19 as in: take the one from pbrobinson and put it in for now? 21:21:47 sdziallas, yes 21:22:13 kanarip: awesome, I can take care of this and adjust the spins in question accordingly. 21:22:27 sdziallas, also update the Makefile.am when you do so 21:22:34 kanarip: actually... I guess we should probably import the moblin spin, too, if we want it for alpha / in time, right? 21:22:42 kanarip: okay, I'll do that. 21:23:04 sdziallas, yes 21:23:21 #agreed kanarip to update the Makefile.am / configure.ac (done already, pushed) 21:24:02 #agreed sdziallas to pull in the fedora-live-mini.ks and update Makefile.am accordingly, bumping the Z in the X.Y.Z versioning in configure.ac 21:24:24 #agreed sdziallas to pull in the fedora-livecd-moblin.ks as well (again updating the Makefile.am and so forth) 21:24:45 * sdziallas will do so immediately post-meeting. 21:24:50 sdziallas, thanks 21:25:10 merci ;) 21:25:11 sdziallas, test your changes with a simple "autoreconf -v && ./configure && make rpm" 21:25:32 * kanarip has one more topic 21:25:40 #topic Board SWG: Spins 21:26:04 kanarip: im here now 21:26:10 * huff reads up 21:26:16 #link http://lists.fedoraproject.org/pipermail/spins/2010-February/000996.html 21:26:30 huff, concerning the mini.ks shuffle/merge 21:27:07 kanarip: what I would like to see happen is AOS ks becomes mini.ks 21:27:20 and get modified to make smaller and more generic 21:27:39 sdziallas, is that sustainable for you guys? 21:28:11 kanarip: well, I'd be happy with it as long as we can get either a mini-x thing on top of it or are allowed to put the required packages into the spins ks. 21:28:30 kanarip: so generally yes - we'd need to figure out whether the livesys stuff from %post would go, though. 21:28:46 sdziallas, does the aos.ks have sufficient of base.ks's scripting to make sure the spin behaves properly? 21:28:58 huff: if I understand you correct then this is currently not part of AOS, is it? 21:29:06 sdziallas, that's my point exactly 21:29:07 not as a livecd 21:29:13 aos is basicly just a min package set 21:29:38 huff, the need for a mini.ks is the minimal package set + the scripts that set the spin up as a spin 21:30:47 ok do you want me to add that? 21:31:32 or... 21:31:44 cwickert: welcome 21:31:49 sorry to be late 21:32:04 kanarip: we could revive the scriptlet attempts from ~ F9... but it's probably a bit of late notice for that 21:32:08 fedora-live-mini.ks contains the scripts and the fedora-live-aos.ks for the packages 21:32:28 fedora-live-base.ks contains fedora-live-mini.ks (for the scripts and the minimal set of packages) and the extra packages 21:32:45 sdziallas, yeah, either of these changes seem a little late to me anyway 21:33:19 i'd rather make the rigorous changes post-f13 21:33:19 * sdziallas mhms. 21:33:35 or rather, after we branch off master to F-13 21:35:11 so how do we proceed here? just import the old mini.ks and do minor mods if needed - or put in the aos.ks adding a bunch of stuff to it. 21:35:36 sdziallas, i'd favor pulling in mini.ks and leaving the rest as-is 21:36:07 shall i call a vote? 21:36:51 sounds good. 21:37:12 vote: pull in mini.ks as-is and use that for soas and moblin (+1) or use the aos ks and copy the scripts from base.ks (+0) 21:37:14 +1 21:37:51 +1 21:37:56 +2 21:38:13 Weak +1 21:38:27 * nirik is fine with whatever the affected spins want to do... but weak +1 here I guess. 21:38:45 * sdziallas is +1, but would like to see the AOS thing done afterwards. 21:38:59 I agree with sdziallas 21:39:54 as in "afterwards" = "make sure we get it done altogether in time" :) 21:40:11 biertie, your vote is invalid 21:40:20 ;) 21:40:28 I'm fat, so I count for 2 21:40:35 perhaps after we branch that could land in the head branch? 21:40:36 but we have sufficient +1's for a majority anyway 21:40:37 huff: I'd be happy to help with testing / nuking randomly annoying package dependencies, but you guys did already the work and it just needs to be integrated in the repo at some point. 21:41:05 biertie, no you're not fat, and no you do not count for 2 even if you were 21:41:21 nirik: +1 21:41:22 :( 21:41:27 #agreed mini.ks to be pulled in, soas and moblin to use mini.ks 21:41:51 nirik, we can branch for F-14-devel now and make the changes, have them fall back on master post-F-13 release? 21:42:18 dunno. whatever works I guess. 21:43:00 well, it'd facilitate the work gets started now, but not interrupt the nightly spins that are composed from master iirc, right? 21:44:00 yes. 21:44:40 ok, that's a wrap on the mini.ks then 21:44:55 i'll branch off F-14-devel to facilitate work being done right now 21:45:04 #action kanarip to branch off master to F-14-devel 21:45:29 #topic Board SWG: Fedora Target Audience wrt. Spins (attempt #2) 21:46:12 so, the board apparently really wants this target audience thing to be defined, in terms of what Fedora Project's target audience is, and they're wondering how such target audiences relate to spins 21:46:52 #link http://lists.fedoraproject.org/pipermail/spins/2010-February/000996.html 21:47:10 I liked the option 2 type solution. I want the board to set a priority for the project to help with resource allocation and to help 21:47:39 * nirik also favors option 2, but hasnt had time to reply to the email yet. 21:47:43 make design decisions. But I want spins able to target other groups where there isn't an unresolvable conflict. 21:47:47 I honestly don't see why this questions needs answering, the spins exist because there is a target audience that the desktop spin doesn't satisfy ... its kind of self explanitory 21:48:34 i favor option 3, where each spin defines it's target audience 21:48:44 I would hope the Fedora target is either general enough to include all spins audiences, or at least not limiting us to it. 21:49:51 are we to decide about the solutions? 21:49:54 well to me the Fedora target is something completely different then what the board is coming up with anyway 21:50:02 The issue with 3 is it opens up a can of worms. 21:50:03 kanarip: I'm with you, I think that's kinda what we are doing now and it seems to be working 21:50:07 cwickert, we're to indicate our most favorable 21:50:10 brunowolff: how so? 21:50:14 IMO they just asked us for our problems 21:50:36 and IMHO the problems are more serious than all this discussion 21:50:48 What if you need two different versions of the same program to satisfy the requirements of two spins? 21:51:02 brunowolff, example? 21:51:06 brunowolff: I don't follow 21:51:25 brunowolff: spins work together to either use the same version, or provide parallel installable ones? 21:51:25 brunowolff, technically non-sustainable, because those two versions would not be in Fedora 21:51:29 I don't think it is likely. But I think that is one of the differences between option 2 and 3. 21:51:57 brunowolff, take a look at ubuntu vs. xubuntu. xubuntu ships a stripped down version of gnome-session because they need it for gdm 21:52:14 I really think the conflicts thing is overblown. Sure, there are conflicts between groups... but we are all sane adults, we can try and solve them via techinical means, if not we can always ask fesco/board to decide. 21:52:17 but IMO the problem should not be patched or adressed on a packaging level but upstream 21:52:21 With 2 spins need to work with in the guidelines. If there is a conflict the target wins. 21:52:36 nirik: +1 21:52:47 brunowolff, that's not making any sense to me 21:52:56 With option 3 I think people need to at least be thinking about updated guidelines where there are conflicting packaging 21:53:05 requirements between different spins. 21:53:05 nirik, +1, I cannot really think of a conflict 21:53:10 brunowolff, there can't be conflicting packaging 21:53:16 I mean: are kde and gnome conflicting? 21:53:20 brunowolff, that's not allowed in fedora 21:53:33 ell, there are things like gnome folks moving to polkit, and kde needing PolicyKit still for a cycle... 21:53:34 no spin is going to fit into the "target" they are basically just formally putting together a plan to say "we endorse the Desktop Spin and the audience they target" ... more or less 21:53:58 the whole point of the spin is to satisfy an audience not satisfied by the target 21:54:00 nirik, ok, but this can be done within the spin 21:54:07 I think that is sort of the point of the question. How far are we willing to go to accomadate spins that want to do something different/ 21:54:29 cwickert: right, I am saying thats an example where there was a conflict and it was solved by allowing kde folks to keep the older interface around until they didn't need it. 21:54:39 IMHO we should make sure that we all build from and upon fedora 21:54:51 if we started like *buntu we are damned 21:55:08 I think that is more in line with solution 2 than solution 3. 21:55:25 of course there are limitations in the way we customize our spins, but we need to solve this technically 21:55:30 cwickert: if we were anything like *buntu, I wouldn't be here ... I left that mess a long time ago 21:55:40 maxamillion, +1 21:55:49 can we please cut off the comparisons with *buntu? 21:55:52 you make me feel sick 21:55:55 kanarip: +1 21:56:00 * maxamillion was just commenting 21:56:05 look 21:56:09 it's all in fedora 21:56:11 I am however tired of the *buntu comparisons on the whole 21:56:14 there can't be many conflicts 21:56:20 kanarip: +1 21:56:26 and where there are, the spins sig is not the place to solve those problems 21:56:31 cvs.fedoraproject.org is 21:57:08 now 21:57:11 for example if we want another config or another branding for program foo, then we need to make subpackages. gnome-panel-branding foo and gnome-panel-branding-bar. suse alreay does this and they do it pretty clever 21:57:44 if we let the board define target audiences to create spins for, no matter how broad, how well do you think this addresses spins with a niche target audience such as our current AOS spin? 21:58:03 cwickert, that's Re-Mix, not Spins 21:58:14 kanarip, not necessarily 21:58:26 think of the mini spin that wants another panel layout 21:58:38 this has nothing to do with branding, it's both fedora 21:58:43 if the means and packages are in fedora already, then again the spins sig is not where the battle is fought 21:58:53 right 21:59:25 I agree with that. 21:59:29 but the spins sig should kick others to acress the technical problems 21:59:43 then we don't need to worry anout conflicts and target audiences 21:59:44 i'm very reluctant to go into the realm of matching target audiences for our spins with target audiences for the project as a whole 21:59:51 hence i'm in favor of option 3 22:00:18 +1 22:00:19 The email message proposed Spins SIG -> FESCO -> Board, but I think just FESCO -> Board seems more appropriate. 22:00:36 but IMO this is all too theoretical 22:00:44 cwickert: +1 22:00:53 +1 22:01:04 -1 22:01:42 I mean, if we stick with the "working" target audience, I don't care if the people are working on/with Gnome/KDE/LXDE/whatever 22:01:46 * nirik was +1ing cwickert's saying IMO this is too theoretical, not voting for option 3 22:01:50 I believe the questions are about how to go about resolving conflicts when they come up. 22:02:17 and if we say we are not for "working" but for "consuming" people, this is nothing we can influence with the spins 22:02:44 here's a question: 22:02:53 We can certainly add we aren't seeing unresolveable conflicts now if that is how spin owners are seeing things. 22:02:58 why are we adding this burocracy(sp?) on top of the way we do things now? 22:03:56 FYI, currently we have: "Someone who (1) is voluntarily switching to Linux, (2) is familiar with computers, but is not necessarily a hacker or developer, (3) is likely to collaborate in some fashion when something's wrong with Fedora, and (4) wants to use Fedora for general productivity, either using desktop applications or a Web browser." 22:04:14 so, some of the spins don't match that target. 22:04:23 which ones? 22:04:26 but I think conflicts are very rare 22:04:32 I don't see added bureaucracy for the cases where there aren't conflicts, which for most stuff there isn't. 22:04:47 say the electronics lab targets developers/hackers? 22:04:52 nirik, AOS Spin 22:05:04 i for one think all of 4 just stink 22:05:10 nirik: wtf? .... Xfce Spin target audience = users who want to use Xfce ... why do we need the break down about types of users? 22:05:16 design-suite doesn't go for 'general productivty' ? 22:06:15 My view on why we are getting asked is that the Board is looking at a way to resolve conflicts and determine priorities using 22:06:42 a targert audience and asking us (the Spins SIG) if that is going to cause problems. 22:06:46 maxamillion, I think these definitions are on different levels. active/inactive/working/consuming is on a completely different level than Xfce, KDE, Games, Design etc 22:07:07 If that's the question, I think the answer is that it isn't going to cause problems. 22:07:20 brunowolff, +1 22:07:38 I think we all can subscribe to the current definition that nirik just quoted 22:07:42 cwickert: right, but the Xfce, KDE, Games, Design, etc. is the level at which the Spins are sitting, why do we need to be involved in the decisions of the active/inactive/working/consuming/etc. level? 22:08:24 maxamillion, I don't want to get involved. I'm interested in technical problems and solutions but not in hypothetical questions. 22:08:26 * sdziallas sorry, back. kanarip: everything should be in shape now. 22:08:38 that's why I'm in FESCO, not the board 22:08:53 cwickert: right, but if I understand this correctly ... that's what the Board is asking of us, is it not? 22:08:54 sdziallas, ok 22:08:58 * nirik goes to get more coffee. 22:09:15 maxamillion, yes, but I think I cannot comment on that 22:09:16 I don't see why any of what the Board is trying to do is our problem or how/why it even effects us? 22:09:30 that's why is skipped the first part in my reply completely 22:09:34 rgr 22:09:40 * cwickert is working on a reply atm 22:09:50 * maxamillion hates politics 22:10:00 :) 22:11:30 so do we agree to say "not our buisness"? 22:11:50 or should we give a recommendation for something 22:12:00 +1 at saying "not our business" 22:12:05 * kanarip is working on a reply too 22:12:06 which would preferralby 3 IMHO 22:12:10 More or less. In addition, I think it is useful for the project to have a target audience. 22:12:21 * maxamillion facepalms 22:12:36 mine will be from an individual perspective, not the Spins SIG representative or anything like that 22:12:57 i feel we would need to reach consensus here first, before we can send out any message on behalf of the Spins SIG 22:12:57 I'm going to stay out of the rest of this part ... you guys just hash it out and tell me what the result is. I will adjust my actions/activities/involvement accordingly 22:13:12 * cwickert will write from an Xfce/LXDE perspective 22:13:27 You might also point them to this log as the discussion might be of interest to get a better idea of how we interpreted their questions. 22:13:32 or a "non-gnome" perspective 22:14:30 For the Spins perspective I can live with whatever. I am here more for techincal than vision reasons and the impact on any 22:14:48 of the soltuions on the games spin is negligible. 22:15:15 I think spins defining their target audience could be nice... if only to help in marketing... 22:15:51 but yeah, I don't think we as a sig should reply with anything unless their is some consensus. 22:16:24 To some extent we do and document it on the summary sections of our spin pages. 22:16:24 nirik: I think each spin already does 22:16:27 * abadger1999 thinks the Board asked the SPin SIG a bunch of questions that are better targeted at the Spin owners themselves. 22:16:38 abadger1999: +1 22:16:41 * nirik nods. 22:17:07 Yeah, the other Kevin probably has a lot different opinions on this than the rest of us. 22:18:51 * nirik thinks we are way over time. ;) 22:19:17 nirik: +1 22:23:03 #endmeeting