[00:00] <cjwatson> we need to get round to transferring mvo from desktop so that he can answer this sort of question. :)
[00:01] <slangasek> :-)
[00:01] <cjwatson> asac: I suggest mailing Michael and/or ubuntu-devel
[00:01] <cjwatson> ok, we're at time
[00:02] <cjwatson> only thing I forgot is to remind folks that I'll be away next week on a music course
[00:02] <james_w> have fun cjwatson
[00:02] <james_w> asac: I've got it now, I killed update-notifier and restarted then waited a while
[00:02] <cjwatson> have fun releasing alpha 4 without me :)
[00:02] <slangasek> and I'm at DebConf, so pitti is the designated release nagger^W manager for alpha-4
[00:02] <james_w> I'm away for a long weekend, back on Wednesday
[00:02]  * evand notes that he will be away from tonight until next sunday
[00:03] <slangasek> at DebConf next week, that is; I'll still be around to nag people this week :)
[00:03] <asac> oops. noone left for alpha 4 ;)
[00:03] <cjwatson> yes, we'll be very lightly staffed in general next week
[00:03]  * asac runs ;)
[00:03] <james_w> slangasek: I'm sure you could manage to nag remotely as well
[00:03] <cjwatson> there's aKademy too
[00:03] <slangasek> james_w: from experience, I don't expect to have the spare attention circuits :)
[00:04] <james_w> slangasek: just nagging in spirit then?
[00:04] <bryce> slangasek needs a nagbot
[00:04] <TheMuso> heh
[00:04] <calc> will be bumping all the OOo a4 bugs back a bit unless OOo 3.0rc1 is released magically on Aug 8
[00:04] <slangasek> james_w: yes :-)
[00:05] <bryce> maybe we could make slangasek obsolete with a tiny bash script?
[00:05] <cjwatson> oh, adjourned, in case it wasn't obvious
[00:05] <slangasek> thanks, all :)
[00:05] <james_w> thanks all
[00:05] <evand> thanks!
[00:05] <ArneGoetje> thanks
[00:05] <TheMuso> heh thanks all.
[00:05] <calc> thanks
[00:05] <liw> thanks
[00:05] <bryce> thanks
[00:05] <james_w> though I think we should continue mocking slangasek unofficially
[00:05]  * TheMuso heads to the store to get some more milk...
[00:05] <bryce> james_w, agreed
[00:05] <cjwatson> surely slangasek would need to be replaced with a POSIX sh script rather than bash
[00:06] <cjwatson> wouldn't do to be standards-uncompliant
[00:06] <asac> thanks!
[00:06] <slangasek> heh
[00:06] <doko> so I never will replace slangasek =)
[00:07] <liw> hmm, my 121 line nagger script is written for bash, so it's not useful for this, too bad :)
[00:08]  * liw nags himself to bed, good night
[12:51] <cody-somerville> Mobile Team Meeting time?
[12:51] <persia> In about 10 minutes.
[12:58] <davidm> Indeed, about 3 minutes now :-)
[13:00] <davidm> #startmeeting
[13:00] <MootBot> Meeting started at 07:04. The chair is davidm.
[13:00] <MootBot> Commands Available: [TOPIC], [IDEA], [ACTION], [AGREED], [LINK], [VOTE]
[13:00] <davidm> Hello everyone
[13:01] <persia> Good morning
[13:01] <davidm> The ubuntu mobile meeting is started
[13:01] <lool> Hey
[13:02] <davidm> We have no action items listed from last week, persia since you chaired the meeting is there anything I'm not aware of?
[13:02]  * ogra waves
[13:03] <persia> davidm: Oughtn't be.  The only big item on the agenda was moving the meeting time.
[13:03] <davidm> OK, just checking
[13:03] <davidm> I just pinged StevenK to join us.
[13:06] <davidm> There is no new business on the wiki but I had a heck of a time getting the pages up they switched the wiki over to openid yesterday
[13:06] <davidm> So I'm not surprised that there is nothing there.
[13:07] <davidm> persia, StevenK lool any issues we should talk about?
[13:07] <davidm> ogra, ^^
[13:07] <lool> Not that I'm aware of
[13:07] <lool> (yet :-)
[13:07] <persia> One thing that came up a couple days ago was perhaps doing a review of the seeds to see if we have the right package mix for intrepid.
[13:07] <lool> I'm happy to report I'll be attending the Maemo Summit in Berlin, mid-September
[13:07] <StevenK> Oh yes. Are people happy with my seed changes?
[13:08] <davidm> lool, great
[13:08] <persia> I fiddled with a chroot, and things seem to be good, but it might need some hints to determine if that's what we want to be testing against come FeatureFreeze at the end of the month.
[13:08]  * ogra left his atom classmate idling the last two days ... by the looks of it in idle state generic is taking similar amounts of battery than lpia both survive about 6h
[13:08] <ogra> just fyi
[13:09] <davidm> ogra, thanks for that info.
[13:09] <lool> persia: The mobile or the mid seed?
[13:09] <ogra> not sure what happens during actual usage
[13:09] <persia> lool: Yes.
[13:10] <lool> persia: So both, ok
[13:10] <lool> Since we have some time in this meeting, I'd love it if we could take some minutes to check progress of intrepid tasks: merge of ppa, installer, images etc.
[13:10] <persia> ogra: "generic" == i386?
[13:10] <ogra> no -generic
[13:10] <ogra> == 586
[13:11] <persia> Yeah, but dpkg-arch-wise
[13:11] <davidm> lool, good idea
[13:11] <persia> Unless I've missed something (entirely possible), the non-hildon stuff should all be merged from the PPA.
[13:11] <davidm> [topic] status of  intrepid tasks: merge of ppa, installer, images etc.
[13:11] <MootBot> New Topic:  status of  intrepid tasks: merge of ppa, installer, images etc.
[13:12] <lool> persia: Could you list what remains to be done on the merge front?
[13:12]  * ogra would like to actualy get some assigned beyond "jump in where gaps are" :)
[13:12] <lool> I understand MIC, perhaps xulrunner ?
[13:13] <lool> ogra: Oh you're bored?  *cliketeeclick* I'll find you something   :-P
[13:13] <persia> Ah, well, there's the ones I didn't touch: xulrunner, libdrm, and similar.  I don't currently have access to the tracking data I was using for merges previously, but may be able to recover it if a full list is desired.
[13:13] <ogra> classmate has the release candidate out and i only have compcache left to implementation in the platform team ... so unless any weird additional classmate stuff comes u i'm free
[13:13] <ogra> *to implement
[13:13] <persia> ogra: Can you make -mobile not be a copy of -desktop, look nice, and work?
[13:14] <lool> persia: Could we go into more details?  I'm pretty sure we wont merge psb and libdrm: psb needs the special libdrm and intel needs to provide updated drivers first
[13:14] <ogra> persia, sure :) if someone dfines the corner points of "look nice"
[13:14] <lool> We need updated drivers for the kernel, an update on upstream libdrm status as to rebase needed libdrm patches, and then we can push -psb to intrepid
[13:14] <lool> persia: Concerning xulrunner, we should discuss this with asac
[13:15] <lool> If necessary, we should drop the gconf backend if that can't be merged
[13:15] <persia> lool: Erm.  Has TTM or GEM landed in upstream libdrm?  I hadn't heard of any clear plans to do so within the intrepid timeframe.
[13:15] <StevenK> bryce would know.
[13:15] <lool> persia: I don't know what the plans are either; I don't think psb is using any of the two though
[13:16] <persia> I thought psb was using HEAD, which was TTM in late May/early June.
[13:16]  * persia looks through old mail
[13:16] <lool> I think it's using HEAD too, but I don't know whether it's relying on any memory manager stuff or doing modesettings
[13:17] <lool> Anyway, not our problem; Intel needs to rebase on upstream and tell us what they rely upon for 2.6.26 stuff IMO; the ball is on their side for psb driver support AIUI
[13:17] <persia> If it's using HEAD, it would be using the new memory manager stuff, as I understand it, the old way and the new way of handling VRAM are different semantics.
[13:18] <lool> Ok; well, in both cases it's to Intel to tell us what they are using; we don't have to sort out what their upstream drivers should be or were or will be using
[13:18] <persia> So, that leaves in the NOT MERGING status language-pack-gnome-* which ought be useless and derive from rosetta, grub, which patch is reported as "unsafe".
[13:19] <persia> moko, which is obsolete, and keyring, which doesn't mean anything without a special archive.
[13:19] <lool> The grub patch I'm afraid we will drop; it will remain in the librarian for posterity, but it's unlikely to be merged in the current grub upstream and it wont work with grub2
[13:19] <ogra> lool, the prob is that pulling in an uncommon drm module will break all other drm drivers
[13:19] <lool> Concerning the langpacks, I'm not sure these are obsolete; I suspect we need some integration work to get them integrated in intrepid
[13:20] <lool> ogra: We wont be pulling the psb drm anymore
[13:20] <ogra> so we should be intrested what they base on indeed :)
[13:20] <ogra> ah, k
[13:20] <persia> Note that nouveau waits on HEAD as well.  It's just a matter of porting all the rest of the drivers
[13:20] <lool> Until they rebase it
[13:20]  * ogra missed that
[13:20] <lool> ogra: I mean, as long as we don't get any drivers information from intel, we will remain with the upstream kernel bits
[13:20] <ogra> good
[13:21] <ogra> well, basing everything on HEAD would indeed be preferable ...
[13:21] <lool> I'll doublecheck with ChickenCutClass
[13:21] <ogra> but i guess that extends the intrepid tieframe if we have to do it ourselves
[13:21] <ogra> *time
[13:22] <lool> persia: So on the merge front status is: xulrunner gconf backend needs merging (lool to discuss with asac); langpacks (lool to discuss with asac); MIC (may I assign this to you?)
[13:22] <lool> moko, keyring, grub, to be dropped
[13:22] <persia> lool: MIC was uploaded 80 minutes ago.  langpacks should be useless if there's not a separate PPA: the standard infrastructure should handle them.
[13:22] <lool> As well as psb/libdrm/kernel bits
[13:23] <lool> persia: I'm not sure the standard infrastructure currently ships the midbrowser bits
[13:23] <persia> I'm not sure about xulrunner, and I'll try to get my melting system up to see if there's anything else interesting not merged.
[13:23] <lool> The /usr/lib/midbrowser xulrunner extension
[13:23] <lool> Ok; anything else on the topic of merging ppa to intrepid?
[13:23] <persia> lool: In the langpacks?
[13:23] <lool> persia: In the langpacks
[13:24] <persia> I'll look again.  It ought.
[13:24] <lool> Ok; let's move to kernel, I see amitk joined us
[13:25] <persia> Note that langpacks only work for main: we may need a few MIRs if something is missing.
[13:25] <lool> amitk: I understand linux-lpia is now in the hands of Michael
[13:25] <amitk> lool: that's right
[13:25] <lool> I'm not sure whether the Ubuntu kernel team worked on armel support already; I don't think we have any requirement to do so yet, so it might be a task with a lesser priority?
[13:26] <davidm> lool, no requirement as yet.
[13:27] <amitk> lool: I will have a armel git tree available in the next 2 weeks. I am about to use debian bootstraps to see if it works.
[13:27] <lool> amitk: Great
[13:28] <lool> amitk: I'm not sure whether there's anything else on the pure-kernel topics, I know I wanted to support lpia udebs and that required kernel udebs, but I don't think this was on your plate
[13:28] <lool> amitk: Perhaps we shoudl discuss drivers shortly
[13:28] <lool> amitk: Any drivers you're working on for mid/mobile devices for intrepid or hardy?
[13:29] <amitk> lool: I have added support in the intrepid tree for udebs before I passed it on to Michael
[13:29] <ogra> cool
[13:29] <lool> amitk: Excellen
[13:29] <lool> t
[13:30] <amitk> http://kernel.ubuntu.com/git?p=amitk/ubuntu-intrepid-lpia.git;a=summary
[13:30] <MootBot> LINK received:  http://kernel.ubuntu.com/git?p=amitk/ubuntu-intrepid-lpia.git;a=summary
[13:30] <amitk> ^ for anyone interest in playing with intrepid lpia
[13:31] <persia> amitk: Cool!
[13:32] <lool> amitk: Thanks for the updates
[13:32] <amitk> I'll publish the armel tree as soon as I have it booting on NSLU
[13:32] <lool> StevenK: How are images going?  Did infinity help setup dailies for intrepid?
[13:32] <cjwatson> oh, are lpia udebs in the archive yet?
[13:32] <lool> cjwatson: Geez I can't believe you beep on "udeb"
[13:32] <lool> :-P
[13:33] <cjwatson> I don't :-)
[13:33] <StevenK> lool: I'm working on it, I keep getting distracted by Hardy
[13:33]  * persia is looking forward to udebs: current work to make an installer is happening on i386, which isn't ideal from the stanpoint of seeing the end results.
[13:34] <lool> StevenK: What's the current status?  Do you think it will require a lot of work to get dailies for intrepid?
[13:34] <cjwatson> oh, yes, they seem to be in the archive
[13:34] <cjwatson> I'll build debian-installer for lpia once I get half a chance then
[13:34] <amitk> cjwatson: they built -> https://edge.launchpad.net/ubuntu/+source/linux-lpia/2.6.26-1.1/+build/669803, not sure if they are in the archive though
[13:34] <persia> cjwatson: Thank you.
[13:35] <amitk> aah good
[13:36] <StevenK> lool: Not a great deal, no. Just requires I have some time :-)
[13:37] <lool> StevenK: Okay
[13:37] <lool> persia: Would you be tempted to talk about installer work?
[13:38] <persia> lool: I've been fiddling with glade to get the desktop installer to work at 640x480 (just in case anyone wants to install on a very small screen).
[13:38] <lool> Great!  Scalable UI is a hot topic on Gtk+ channels these days :-)
[13:39] <persia> That's nearly done, and I ought have something for evan to review shortly (although I'll not block on review).
[13:39] <persia> Err, not exactly scalable, just smaller :)
[13:39] <lool> Okay; anything else on installer?
[13:39] <ogra> a different UI or just the existing one adjusted ?
[13:40] <persia> I've put off looking at the 2G/squashfs use case for now: I want to get something working for the larger class of devices first.
[13:40] <ogra> i could look into producing a udeb
[13:40] <persia> For 2G/squashfs/reset capability there is a need for an additional d-i module, and I'm still wrapping my head around the d-i code.
[13:40] <ogra> for the size detection etc
[13:40] <persia> ogra: If you want to chase that, it would be a great help.  Thanks.
[13:40] <ogra> ok, on my TODO then
[13:41] <persia> Once it exists, we can just feed it as part of the ubiquity preseed, and we ought be OK.
[13:41] <lool> Ok; I'm sceptical on the time it would take on such devices to build the squashfs at install time, but that's not the place or time to discuss this
[13:41] <ogra> i did some udeb stuff before with ltsp, so shouldnt be a biggie ... and cjwatson didnt seem reluctant to the idea anymore :)
[13:41] <persia> lool: Not build the squashfs at install time.  I looked at that, but it's just infeasible.
[13:41] <lool> Last topic which comes to my mind ATM would be on ogra's activities for intrepid: what are you up to?  :-)
[13:41] <persia> It's a matter of using the existing squashfs.
[13:41] <cjwatson> what would this new udeb do?
[13:41] <cjwatson> and it's unlikely that you'd want to use that udeb in ubiquity
[13:42] <ogra> if content > capacity -> copy squashfs, set up aufs and warn the user about upgradeability
[13:42] <ogra> if content < capacity -> copy content in place and configure
[13:42] <cjwatson> it already has its own perfectly good squashfs handling code
[13:42] <cjwatson> this is all in ubiquity?
[13:42] <persia> In the case where we detected larger devices, we would have to remove the squashfs hints from the CD suashfs.
[13:42] <ogra> this was supposed to be a udeb
[13:42] <cjwatson> if d-i isn't involved, you should just do this in ubiquity/scripts/install.py, not a udeb
[13:42] <ogra> ubiquity would need a variable to skip the content copying i can preseed for that
[13:43] <cjwatson> or it could figure it out for itself ...
[13:43] <ogra> or that
[13:43] <lool> ogra: Why not copy only the differing contents after setting up the unionfs on the target?
[13:43] <ogra> but that smells like more code changes to ubiquity
[13:43] <persia> cjwatson: Evan advised me that ubiquity worked cleanest as a d-i front-end, and that we ought think of it as an advanced preseeding system, plus the possible facility to copy a livefs to reduce installation effort.
[13:43] <cjwatson> ogra: smells better, though
[13:43] <cjwatson> persia: this is true, but the copying part of ubiquity is very much not geared around using a udeb
[13:43] <ogra> i was planning to change ubiquity itself to the least amount i can ... but if you think thats better i can do it
[13:44] <lool> :c
[13:44] <cjwatson> we use d-i components for all the configuration parts where possible
[13:44] <persia> cjwatson: Ah, so if the copying was to be replaced, it shouldn't be done with a udeb, right?
[13:44] <cjwatson> the copying stage, though, is just done inline
[13:44] <cjwatson> persia: that would certainly not be my inclination
[13:44] <ogra> lool, copy content above would mean standard ubiquity
[13:44] <persia> OK.  That's easier then :)
[13:44] <ogra> lool, only in case where the content doesnt fit it shuld do a different setup
[13:44] <lool> ogra: Ok; didn't know it could copy contents after setting up an unionfs on the target
[13:45] <persia> lool: You can't.
[13:45] <ogra> no that must hapen before
[13:45] <cjwatson> I don't think you can support the 2G/squashfs case particularly sanely in d-i
[13:45] <persia> lool: I'll explain the theoretical mechanism in #ubuntu-mobile after the meeting.
[13:45] <cjwatson> I mean, I know people have tried, but it's really geared towards building up the system from scratch
[13:45] <ogra> cjwatson, the advantage of a udeb is that oem's could just use it with d-i
[13:45] <lool> persia: Ok, let's do this
[13:46] <ogra> where ubiquity might not be wanted
[13:46] <ogra> i.e. mass installs etc
[13:46] <cjwatson> ogra: but it would be much, much, much harder than just creating a udeb to make d-i work in this type of scenario
[13:46] <cjwatson> trust me, I've tried
[13:46]  * lool <15 minutes EOM warning>
[13:46] <ogra> ok
[13:47] <cjwatson> lots of bits of d-i would become essentially non-functional; for instance how do you do package selection when you have a squashfs?
[13:47] <cjwatson> there's really not a lot of point
[13:47] <persia> Right, so anyway, with the udebs in place, and cjwatson's generous offer to do an lpia d-i build, I should have a working installer for -mid next week, for the >4G case.
[13:47] <ogra> right, no package selection in that case indeed
[13:47] <persia> For the smaller case, we'll have to fiddle more, but at least there will be something to test (assuming we can build images)
[13:47] <cjwatson> and FWIW OEMs are already using ubiquity for mass installations
[13:47] <ogra> ok
[13:47] <cjwatson> hardy has a noninteractive mode
[13:47] <ogra> i always hought d-i would be easier for that :)
[13:48] <persia> ubiquity noninteractive seems a lot easier than d-i+preseed for OEM use.
[13:48] <ogra> (i would take d-i if i were OEM ... :) )
[13:48] <ogra> yeah, squashfs is likely faster than having package selection being installed
[13:48] <ogra> indeed
[13:48] <cjwatson> yeah, d-i just doesn't really work with a pre-existing filesystem
[13:48] <persia> ogra: That makes it just a file copy, rather than repeating all the postinsts, etc.
[13:49] <cjwatson> so easier or not, ubiquity is what you've got ;-)
[13:49] <ogra> ok
[13:49] <ogra> sad :) i like shell more than python ... but will find my way around :)
[13:49] <davidm> [action] lool the investigation of xulrunner, langpacks and linux-lpia bits
[13:49] <MootBot> ACTION received:  lool the investigation of xulrunner, langpacks and linux-lpia bits
[13:49] <lool> Thanks davidm
[13:50] <davidm> lool happy to do so.
[13:50] <lool> If we are done with the installer, I'd love to use the remaining time to discuss ogra's work in intrepid
[13:50] <davidm> 10 minute warning
[13:50]  * ogra listens up 
[13:50]  * persia also, especially any ideas for the -mobile seed
[13:51] <lool> ogra: Well actually I was expecting you would give us (mostly me) an update on what your plans are, what you're currently working on for intrepid etc.
[13:52] <lool> I don't have anything to throw at you at this minute, but I've noted that you have free cycles and will come up with tasks for you later on  ;-)
[13:52] <ogra> well, i was tied in by classmate work and as i said have one task left for platform (compcache) which should be done mid next week
[13:52] <lool> Okay; thanks for the update
[13:53] <lool> Are you also merging classmate stuff into intrepid?
[13:53] <cjwatson> s/platform/foundations/ ;-)
[13:53] <lool> I know you had an issue with merging the last kernel patches
[13:53] <ogra> as i understood persia he wants me to clean up the -mobile seed ... if that isnt supposed to be a copy of -desktop with different themeing for smaller sizes i need input on the apps that differ
[13:53] <ogra> cjwatson, oh, right :)
[13:54] <ogra> lool, the kernel team agreed to make my patches SRUable for the standard hardy kernel which should take the load off my sholders
[13:54] <ogra> intrepid has all these fixes in already
[13:54] <lool> ogra: I think describing the new use of the mobile seed as "different theming for smaller sizes" is quite a good target; I guess we can remove applications which aren't useful in the subnotebook case or replace them with more netbook-friendly apps
[13:54] <persia> ogra: Most of the apps seem to work with -desktop on my Kohjinsha, but menus are awkward, and some stuff doesn't work at 600 veritical pixels.
[13:54] <lool> (no particular example in mind, but I thought we should go through the seed line by line with these criterions)
[13:55] <ogra> lool, well, ther are patches for apps that dont work (i.e. i'd like to look at evo and gdm patches)
[13:55] <lool> Ok
[13:55]  * lool <5 minutes til EOM>
[13:55] <ogra> the question is if we can get them in in a sane way to not affect normal desktop usage
[13:56] <ogra> that will need some review
[13:56] <persia> ogra: Probably the scalable UI stuff, really.
[13:56] <ogra> you mean the panning modes ?
[13:57] <ogra> that requires special X setup ... not sure i like to have it in a general setup
[13:57] <lool> Ok; thanks everybody for the heads up on your work -- did I miss anybody, anything?
[13:57] <persia> Either that, or something that doesn't assume 96 DPI and >10" screen so that one could use the information one has from the X server to drive app behaviour.
[13:57] <ogra> and also not sure how it will work with the new Xorg ... thats moving away from virtual desktops to be used for panning upstream
[13:57] <lool> persia: You want to read the discussions on Gtk+'s em support
[13:58] <pitti> hi
[13:58] <persia> lool: Yes I do: thanks for mentioning it earlier.
[13:58] <lool> Ok; any other last minute topic?
[13:58] <MacSlow> hey pitti
[13:58] <MacSlow> hey seb128, mpt, ogra, lool
[13:58] <seb128> hello
[13:58] <davidm> looks like it's about time to end the meeting going once.......................
[13:59] <davidm> looks like it's about time to end the meeting going twice..........
[14:00] <mpt> ooh, GTK em support?
[14:00] <davidm> #endmeeting
[14:00] <MootBot> Meeting finished at 08:04.
[14:00] <lool> Thanks everybody!  Also thanks davidm for chairing!
[14:00] <mpt> lool, where could I find those discussions?
[14:00] <davidm> I like this time better :-)
[14:00] <lool> mpt: Mentionned on Planet GNOME; grabbing the mail.gnome.org URL
[14:00] <Keybuk> davidm: you have to finish on time though ;)
[14:01] <lool> http://mid.gmane.org/1218056830.3623.64.camel@x61.fubar.dk
[14:01] <persia> Keybuk: Mootbot is 4 minutes ahead!
[14:01] <lool> mpt, persia ^
[14:01] <pedro_> hello everybody
[14:01] <davidm> Keybuk, yes, I have to see why my time and MootBot differ
[14:01] <MacSlow> hey pedro_
[14:01]  * mvo waves
[14:01] <mpt> thanks lool
[14:01] <davidm> I thought I was using ntp on my laptop :-/
[14:02]  * lool wishes next meeting attendees a good meeting and leaves the stage
[14:02] <pedro_> hi MacSlow, mvo
[14:02] <Keybuk> davidm: I read it as 1301Z
[14:02] <Keybuk> bwahaha
[14:02] <Keybuk> I can so ensure you finish your meeting on time ;)
[14:02] <Keybuk> annnnyway
[14:03] <Riddell> hi
[14:03] <MacSlow> hey Riddell
[14:04] <Keybuk> https://wiki.ubuntu.com/DesktopTeam/Meeting/2008-08-07
[14:04] <Keybuk> after much battling with the wiki, I almost have an agenda
[14:05] <tedg1> mpt: Did you see the em video in that message?  Very cool.
[14:05] <Keybuk> so first thing on the agenda
[14:05] <Keybuk> mpt: Welcome Aboard!
[14:06] <Keybuk> mpt is now a full-time member of the team as our UI designer
[14:06] <mpt> \o/
[14:06] <mvo> mpt++
[14:06] <mvo> welcome!
[14:06]  * pitti hugs mpt
[14:06] <Keybuk> he is available for all of your UI needs
[14:06] <MacSlow> what video?
[14:06] <pedro_> welcome mpt!
[14:06] <Keybuk> and will no doubt be coming to you with sketches and ideas
[14:06] <tedg1> MacSlow: http://people.freedesktop.org/~david/gtk-resolution-indepence.ogg
[14:07] <seb128> mpt: welcome!
[14:07] <Keybuk> tedg1: that's doing scaling of controls by point size?
[14:08] <seb128> and we already are off-topic ;-)
[14:08] <tedg1> Keybuk: Yes, and scaling of spacing.
[14:08] <Keybuk> shiny
[14:08] <tedg1> Keybuk: Basically all "values" become EMs -- which covert to pixels based on point size.
[14:08] <Keybuk> pitti: thanks for doing last week's meeting while I was down with Turkish bugs
[14:09] <Keybuk> I see one action outstanding:
[14:09] <Keybuk>  * Mirko to clarify gdm goals for face browser for Intrepid/Intrepid+1.
[14:09] <Keybuk> (I assume MacSlow? :p)
[14:09] <MacSlow> tedg1, very nice
[14:09] <pitti> Keybuk: you're welcome
[14:09] <pitti> Keybuk: I tend to use real names on !IRC...
[14:10] <Keybuk> pitti: no matter how badly spelled? :)
[14:10] <pitti> oh, it's Mirco?
[14:10]  * pitti hugs MacSlow, sorry
[14:10] <Keybuk> MacSlow: any update on the gdm goals?
[14:10] <MacSlow> Mirco is correct
[14:11] <MacSlow> Well, I don't think I can meet the feature-freeze deadline for the login-experience.
[14:11] <Keybuk> I saw your mail this morning
[14:11] <Keybuk> and I saw your mail last week outlining the missing bits
[14:11] <Keybuk> so I think we can consider that done?
[14:12] <MacSlow> The amount of "boiler-plate" code for a graphical-greeter is far more than I anticipated... partly also due to the lack of experience with GObject... never was forced to seriously use it before.
[14:12] <Keybuk> ok
[14:12] <Keybuk> mpt: also you were attempting to get in touch with the user admin people, any joy there so far?
[14:12] <MacSlow> no... it's not done
[14:13] <seb128> Keybuk: user admin?
[14:13] <mvo> who are the "user admin" people?
[14:13] <mpt> Keybuk, yes, they'd like my input. So it's up to you to put that it in the appropriate position on my priority list. :-)
[14:13] <Keybuk> MacSlow: I mean is the goals being clarified done?  It looks to me like they are
[14:13] <mpt> mvo, Fedora developers at the moment
[14:13] <seb128> context?
[14:13]  * mvo nods
[14:13] <seb128> users-admin in something fedora doesn't use
[14:14] <seb128> s/in/is
[14:14] <mvo> so a modern replacement for the one that gnome-system-tools is providing now?
[14:14] <pitti> my +1 for getting a replacement for users-admin soon :)
[14:14] <MacSlow> Keybuk, you mean with upstream?
[14:14] <mpt> specifically Jon McCann and Matthias Clasen
[14:14] <seele> mpt: you have design changes for the fedora ui?  are they on the wiki somewhere?
[14:14] <seb128> mpt: can we get some context about what tool you are speaking about?
[14:14] <mpt> seele, seb128: http://www.redhat.com/archives/fedora-desktop-list/2008-May/msg00006.html
[14:15] <seb128> mpt: ah, thanks!
[14:15] <seb128> ok, that's something new, didn't know about that
[14:15] <seele> mpt: right, i've seen that.  it sounded like you had changes
[14:15] <seb128> pitti reviewed their tool some weeks ago and that was not something to use
[14:15] <pitti> seb128: oh, that was system-config-user
[14:15] <pitti> seb128: but they are planning a new one
[14:15] <mpt> seb128, James Westby raised it at this meeting three weeks ago.
[14:15] <seb128> pitti: right, I didn't know that ;-)
[14:16] <Keybuk> mpt: I think at this point new user admin is jazzy jaguar, so something to work on in the background
[14:16] <MacSlow> Keybuk, ah... ok read my report from last week again... now I know what you refered to... yes I know what's needed. The amount of work needed is not doable until feature-freeze.
[14:16] <Keybuk> ok
[14:16] <Keybuk> other agenda items
[14:16] <Keybuk> Riddell: libzip MIR
[14:17] <Keybuk> I saw a nag in your activity report
[14:17] <mpt> seele, not yet. :-)
[14:17] <Keybuk> slangasek: around? is there much of a backlog in MIRs at the moment?
[14:17] <Riddell> I keep trying to nag jdstrand and kees, no result so far
[14:17] <seele> mpt: right on, let me know if youre looking for feedback.  i was thinking of doing the same for KDE once akademy is over
[14:18] <seb128> Keybuk: MIR are pitti and doko
[14:18] <seb128> no?
[14:18] <pitti> Riddell: if it's otherwise ok, we can turn it upside down, promote it now, and turn the bug into an intrepid milestoned bug for security
[14:18] <pitti> seb128: it is marked as 'needs security review'
[14:18] <seb128> what is the issue? duplication and having yet another code to do that?
[14:19] <jdstrand> it's on our list of things todo-- we've had some rather high priority stuff come up lately
[14:19] <pitti> that's of course always a general issue
[14:19] <pitti> unzip itself has had a number of issues, and it's certainly much better reviewed than libzip
[14:19] <Keybuk> ok, as long as it's not been dropped
[14:19] <Riddell> pitti: that would help me
[14:20] <Keybuk> it sounds like this is on the TODO and will be done as soon as it can be
[14:20] <Riddell> Keybuk: has been for two months..
[14:20] <pitti> Riddell: can you please follow up on the MIR bug, or send mail?
[14:20] <Keybuk> pitti: if you have no objection to that fast track with your RM hat on, go for it
[14:20] <pitti> (just as a reminder)
[14:20] <Riddell> pitti: ok
[14:20] <pitti> I tend to forget IRC pings, but I'm eager to get inbox zero and thus have to do stuff for that :)
[14:21] <Keybuk> ok
[14:21] <Keybuk> pitti: Team structure, member shuffling
[14:22] <pitti> that's actually a question
[14:22]  * tedg suffered atheros based netsplit, sorry
[14:22] <pitti> a while ago it was announced that some people will change teams
[14:22] <pitti> so I'm just curious about that, since it partly affects me as well (whom to ping for reports, which specs to watch, etc.)
[14:23] <Keybuk> ah, ok
[14:23] <Keybuk> it's probably been bounced around in various forums
[14:23] <pitti> I haven't seen any official announcement so far, I think
[14:23] <Keybuk> I'll attempt to outline the changes as best I can ;)
[14:23] <Keybuk> The first change is that Colin and I are stepping down as team managers
[14:23] <Keybuk> Colin will be taking the role of Technical Lead for the platform team
[14:24] <Keybuk> while I will be taking some currently untitled role in the platform team (proposed titles include Senior Engineer, Lead Developer, etc.) - basically a technical lead level position without the specific responsbility
[14:24] <Keybuk> we will be adding Technical Leads to the other teams
[14:24] <Keybuk> as of the sprint a few weeks back, the Desktop Team Technical Lead is Martin Pitt
[14:25] <Keybuk> we're hiring for both the new desktop and platform team managers
[14:25] <Keybuk> so that's the top shuffle
[14:26] <Keybuk> since everybody will end up changing line managers anyway, we figured it was a good opportunity to fix some of the strange historical reporting
[14:26] <Keybuk> so moving from the desktop to the platform team will be mvo
[14:26] <pitti> so that will not happen until we actually get new managers?
[14:26] <pitti> mvo: bye bye compiz? :)
[14:26] <Keybuk> and moving from the platform to the desktop team will be asac, ArneGoetje, bryce, calc and maybe TheMuso
[14:27] <Keybuk> right, the plan is not to make this change until the new managers are hired
[14:27] <Keybuk> since it saves having too much upheaval across too long a time
[14:27] <pitti> ok, that was my primary concern
[14:27] <Keybuk> makes more sense to have a musical chairs flag day
[14:27] <mvo> pitti: not necessarily, but it will probably be less
[14:27] <pitti> I never quite knew which people to ask for activity reports, etc.
[14:27] <Keybuk> so today, the teams are as they were
[14:27] <Keybuk> desktop experience:
[14:28] <Keybuk> a new desktop experience team is being created, and a manager hired who will report to Mark (I think, it also may be ChrisKenyon)
[14:28] <Keybuk> until that manager is hired, Mark is directly leading the effort
[14:28] <Keybuk> and MacSlow and tedg are seconded to that team, whilst remaining reporting to the desktop team
[14:28] <Keybuk> (so they still attend this meeting and send me or you activity reports, but Mark manages their priority list)
[14:28] <MacSlow> until the new manager for that is in place I assume
[14:29] <Keybuk> right, once the new manager for that team is hired, we'll make some reporting changes
[14:29] <Keybuk> marketing:
[14:29] <Keybuk> the marketing team is growing a director, a manager, a visual designer, etc.
[14:30] <Keybuk> it may or may not be that artwork will report to that team
[14:30] <Keybuk> or it may remain reporting to the desktop team
[14:30] <Keybuk> that is not quite clear
[14:30] <tedg> Or Desktop Experience... kwwii isn't here, so I think we should tell him to report to all three ;)
[14:30] <MacSlow> :)
[14:30] <Keybuk> heh
[14:31] <Keybuk> the reason it's not clear, is that kwwii has been asked what he wants to do and given some options
[14:31] <Keybuk> and he hid :p
[14:31] <Keybuk> key points are that nothing changes until managers are hired for any of these teams
[14:31] <Keybuk> and that at that point, there will be formal announcements
[14:31] <Keybuk> only changes so far:
[14:31] <Keybuk>  - mpt joining the distro team \o/
[14:32] <pitti> ok, thank you for the summary
[14:32] <Keybuk>  - pitti promoted to technical lead
[14:32] <tedg> Do you know how the progress on hiring those managers is going?  Is it something likely to be this week, this month, this year?
[14:32] <Keybuk> tedg: for the desktop experience team, Mark has somebody in mind, but he plans to blog about the team for a month or so, and see if he can get interesting replies from interesting people that may be more suited
[14:33] <Keybuk> for the distro team managers, a few candidates have been interviewed so far, and some show promise
[14:33] <Keybuk> I don't know much about the marketing team's hiring
[14:34] <Keybuk> good time for questions if anyone's got any :-)
[14:35] <Riddell> when you say moving to platform team, that should presumably now read moving to foundations team
[14:35] <Keybuk> err?
[14:35] <Keybuk> no?
[14:35] <pitti> will take a while for us to adapt :)
[14:35] <Keybuk> did that get renamed?
[14:35] <pitti> Keybuk: yes, I think
[14:35] <Riddell> it seems to be renamed
[14:35] <Keybuk> when did that happen?  I missed a mail
[14:35] <Riddell> we're all platform now
[14:35] <mvo> oh? name change?
[14:35] <MacSlow> I thought that discussion form the sprint was more of a joke
[14:35] <pitti> Keybuk: this week's pla^foundation team report
[14:36] <pitti> Riddell: oh, platform has a new meaning now?
[14:36] <Keybuk> ahhh
[14:36] <mpt> Not to be confused with Launchpad's Foundations team
[14:36] <Keybuk> mpt: I suspect the naming is intended to match
[14:36] <Keybuk> so the distro team is now the ubuntu platform team?
[14:36] <Keybuk> and the ubuntu platform team is now the ubuntu foundations team?
[14:36] <Riddell> Canonical Distro team now Ubuntu Platform team I believe
[14:36] <pitti> argh
[14:37] <pitti> that could just be "ubuntu" team
[14:37] <Keybuk> shouldn't the desktop team now be the ubuntu architecture team? :)
[14:37]  * pitti ponders introducing SQL-like ID numbers, should they change again in 3 months
[14:37] <tedg> Desktop Experience is going to be Ubuntu Facade Team ;)
[14:37] <seb128> could we get moving, there is quite some points still on the agenda ;-)
[14:37] <MacSlow> tedg, I prefer "Bling Brigade" ;)
[14:38] <Keybuk> tedg: see, people read suggestions like that, and run with them
[14:38] <seb128> (and I might add some if we don't overrun)
[14:38] <Keybuk> yes
[14:38] <Keybuk> let's get moving
[14:38] <Keybuk> sorry, slightly distracted by things moving under me
[14:38] <Keybuk> mvo raised a concern about the length of his own activity report
[14:38] <Keybuk> which seemed to be a good item, how long to people think they should be?
[14:38] <mvo> well, that was not really meant as a agenda item
[14:38] <Keybuk> what content do you like to read in them?
[14:38] <Keybuk> (do you read other people's? :p)
[14:39] <mvo> but I'm curious what people think, I noticed that my reports are kind of verbose
[14:39] <seb128> mvo: they are nice to read ;-)
[14:39] <pitti> mvo: I read them, too, but I think you should drop some details, about every bug and so
[14:39] <MacSlow> mvo, is avoids to have to ask questions :)
[14:39] <pitti> spec progress is great, and I'd keep that
[14:40] <pitti> mvo: btw, I have packagekit commit access, so I can rely some commits, especially if they are specific to apt
[14:40] <pitti> s/rely/relay/
[14:41] <Keybuk> from my POV. I find single paragraph summaries for each major item (spec or ongoing responsibility) works quite well
[14:41] <pitti> hm, I specifically don't mention ongoing responsibility
[14:41] <mvo> pitti: I have access too, I just need to get around commiting it and making git happy (that is in itself a challange)
[14:41] <mvo> but thanks :)
[14:41] <pitti> since it would be quite dull for writing/reading "email, archive maintenance, MIR, SRU" every week?
[14:41] <Keybuk> pitti: by that, I mean things like jockey
[14:42] <pitti> mvo: good luck, haven't fought with branches yet, just checked them out
[14:42] <Keybuk> where you may not have a spec, but you might do some large improvement
[14:42] <pitti> Keybuk: ah, ok
[14:42] <Keybuk> for mvo it'd be things like synaptic
[14:42] <mvo> pitti: thanks, I will need it :)
[14:42] <pitti> Keybuk: I usually mention that, yes
[14:42] <pitti> since it's non-"chores"
[14:43] <Keybuk> ok
[14:43] <Keybuk> roadmap update
[14:43] <mvo> Keybuk: interessting point, I will consider that
[14:43] <Keybuk> https://wiki.canonical.com/DistroTeam/Desktop/8.10
[14:43] <Keybuk> could everybody over the next day or so add a paragraph or so to each of their items there with an update on progress
[14:43] <Keybuk> and in particular, what's done and not done from a Feature Freeze point of view
[14:43] <Keybuk> err
[14:43] <Keybuk> paragraph _in_the_spec_ :p
[14:44] <MacSlow> Keybuk, I assume I should add a section about the state of the login-experience there?!
[14:44] <pitti> so, not on the /8.10 wiki page
[14:44] <pitti> IMHO current status belongs into the whiteboard
[14:44] <Keybuk> MacSlow: no, it's not on my roadmap - it's on Mark's ;)
[14:44] <seb128> pitti: I do nothing out of bugs, updating GNOME and archive admin tasks, do I still need to write activity reports? ;-)
[14:44] <Keybuk> pitti: whiteboard is fine too, not on the /8.10 wiki page
[14:44] <MacSlow> Keybuk, does that have a page?
[14:45] <Keybuk> seb128: interesting things in new GNOME releases, interesting bugs
[14:45] <pitti> MacSlow: it should have a blueprint?
[14:45] <MacSlow> pitti, it has nad a spec
[14:45] <Keybuk> MacSlow: no, but update the blueprint
[14:45] <MacSlow> ok
[14:45] <pitti> MacSlow: "nad" == "had" or "not"?
[14:45] <Keybuk> seb128: you had some additional items?
[14:46] <MacSlow> pitti, it has a blueprint and spec
[14:46] <pitti> ahm, "nad" == "and"
[14:46] <seb128> Keybuk: some topics to raise for discussion
[14:46]  * pitti shuold train more letter shuffling
[14:46] <MacSlow> :)
[14:46] <Keybuk> seb128: go for it
[14:46] <Keybuk> we have 13 minutes
[14:47] <seb128> one being empathy against pidgin
[14:47] <seb128> since GNOME accepted empathy in their desktop seed, should we give it a try?
[14:47] <seb128> we are not in feature freeze yet
[14:47] <seb128> maybe everybody could give it a try this week and comment next week on it?
[14:48] <Keybuk> seb128: with which set of telepathy backends?
[14:48] <Keybuk> I use empathy for day-to-day jabber and salut, but not MSN
[14:48] <seb128> Keybuk: what is recommended, gabble and salut
[14:48] <seb128> we should test msn too if possible, I expect that's what users run the most nowadays
[14:48] <mvo> so empathy would only replace parts of pidgin? some protocols?
[14:49] <seb128> mvo: it can use libpurple so do everything pidgin is doing
[14:49] <seb128> jabber should be better, it does video for example
[14:49] <Keybuk> seb128: I think we should try it - maybe post instructions to ubuntu-devel calling for a trial?
[14:49] <mpt> Should I write up a usability comparison?
[14:49] <seb128> Keybuk: alright
[14:49] <seb128> mpt: yes please
[14:49] <pitti> seb128: I'll test ICQ and jabber, sure; apt-get install empathy?
[14:50] <seb128> pitti: yes
[14:50] <pitti> seb128: does empathy have video chat capabilities already?
[14:50] <pitti> IOW, could it already replace ekiga?
[14:50] <pitti> well, I'll find out
[14:51] <seb128> pitti: yes, the jabber provider should do that
[14:51] <seb128> ok, next questions is gnome-session
[14:51] <pitti> seb128: no ICQ in the protocol list...
[14:52] <pitti> telepathy-haze - A telepathy connection manager that use libpurple
[14:52] <Keybuk> seb128: yup?
[14:52] <pitti> ah
[14:52] <seb128> pitti: no, you need to use the libpurple integration for that, let's discuss this out of the meeting
[14:52] <seb128> new gnome-session sucks at the moment
[14:52] <seb128> and doesn't seem to be moving fast
[14:52] <seb128> I'm pondering if we should revert to 2.20, though that's still a bit early
[14:52] <pitti> indeed, totally broken session management, etc.
[14:53] <pitti> (or, rather, absent)
[14:53] <seb128> one thing is that we need to decide what to do for the logout dialog
[14:53] <pitti> and still no compiz by default, right?
[14:53] <pitti> oh
[14:53] <pitti> mpt: designing a logout dialog that doesn't suck is a huge UI task, I believe
[14:53] <seb128> if we keep the new version we need to take a decision, because the old patch doesn't apply and the dialog needs to be rewritten
[14:53] <seb128> and with all I've to do I doubt I'll be able to do that soon
[14:53] <tedg> I think with the "new" FUSA applet we'll have a set of menu items for the various logouts.
[14:53] <mpt> pitti, I completed that a couple of years ago. The hard part is implementing it. :-)
[14:54] <tedg> So we shouldn't need a dialog per se.
[14:54] <Keybuk> tedg: what is the status of that wrt FF?
[14:54] <pitti> tedg: we shouldn't make fusa mandatory, though
[14:54] <seb128> new fusa applet?
[14:54] <tedg> Keybuk: My plan is to have it done.
[14:54] <tedg> seb128: Patch to to IM status and session management into the applet.
[14:55] <tedg> Not "new", really "patched".
[14:55] <Keybuk> seb128: I guess we need a list of items we consider essential for session
[14:55] <seb128> ah, I don't use the applet and I expect quite some users don't
[14:55] <Keybuk> to compare new/old gnome session against
[14:55] <Keybuk> I know that Matt keeps asking where the old logout dialog went, and when is it coming back?
[14:55] <seb128> Keybuk: well, the real question is "do we want one dialog having all the actions, or should we consider the 2 dialogs upstream way"
[14:56] <seb128> if we want one dialog, do we want the old one we had or something based on what upstream will get (what opensuse has now)
[14:56] <mvo> what is that?
[14:56] <seb128> and then, who would be available to write the dialog ;-)
[14:56] <mvo> two dialog? so first logout, then shutdown?
[14:56] <seb128> mvo: no, what you are in intrepid now
[14:57] <seb128> s/are/have
[14:57] <Keybuk> we've had one dialog with all of the options until now
[14:57] <seb128> mvo: system, logout and system, shutdown
[14:57] <Keybuk> we should keep that one dialog until we make a decision not to
[14:57] <Keybuk> and that decision should be on usability, etc. grounds
[14:57] <pitti> Keybuk: TBH, the old dialog sucked as much as the two dull current ones, it just looked a little better
[14:57]  * Keybuk looks at mpt
[14:57]  * pitti votes for two dialog
[14:57] <seb128> Keybuk: there is nothing to keep, the code which was patched doesn't exist in the rewrite
[14:57] <pitti> I never quite liked the "Do something to my computer" with 7 options thingy
[14:58] <seb128> so we need to rewrite a dialog
[14:58] <seb128> I don't fancy to rewrite the old one first, then think and decide to drop what we should rewrotte
[14:58] <tedg> I guess, are we going to have two ways to logout?  If we assume that some people aren't running the FUSA applet.
[14:58] <mpt> Keybuk, I've done the spec for this. If anyone comes to me and says "We can't do part X because of Y", then I can revise it.
[14:58] <seb128> s/should/just
[14:58] <Keybuk> sounds like we should take that to the mailing list
[14:58] <Keybuk> since there's lots of pieces of decision
[14:59] <seb128> tedg: I don't use fusa, I'm the only user on this laptop
[14:59] <seb128> Keybuk: ubuntu-devel or ubuntu-desktop list?
[14:59] <Keybuk> seb128: ubuntu-devel has more subscribers
[14:59] <tedg> seb128: Would you if it was your IM status also?
[14:59]  * pitti doesn't use fusa either, waste of precious panel space
[14:59] <seb128> tedg: I know my IM status, it's in the notification area, I don't need an applet for that ;-)
[14:59] <tedg> Keybuk: It might be a good discussion to make -desktop more relevant though :)
[14:59] <Keybuk> tedg: also true
[15:00] <tedg> seb128: Good point, need to patch that notification mess out ;)
[15:00] <Keybuk> ok
[15:00] <Keybuk> we're out of time now
[15:00]  * persia knows the meeting started a bit late, but hopes it doesn't run over too much
[15:00] <Keybuk> anybody have any last minute urgent items?
[15:00] <seb128> persia: it's done
[15:00] <seb128> persia: and we still have 1 minute ;-)
[15:00] <persia> seb128: You've at least a minute left
[15:01] <Keybuk> no? good! ok, thanks everyone!
[15:01] <seb128> thanks
[15:01] <mvo> thanks
[15:01] <tedg> thanks
[15:01] <pitti> thanks everyone
[15:02] <persia> OK.  Who's here for the Java meeting today?
[15:02] <MacSlow> thanks
[15:03] <robilad> me!
[15:03] <persia> Hurrah!
[15:03] <persia> Koon: ?  slytherin?  doko?
[15:03] <Koon> o/
[15:04] <persia> OK.  We've a *very* short agenda at https://wiki.ubuntu.com/JavaTeam/Meeting
[15:04] <persia> Basically, just a rundown of the Roadmap.
[15:05] <persia> So, first up: robilad: How goes the effort to breakdown the tasks involved in integrating Java into the server stack?
[15:06] <robilad> i didn't get much done on that, unfortunately, after we agreed that maven is a req
[15:07] <persia> robilad: Is there anything other than maven that you want?
[15:07] <persia> application servers, specific tools, missing libraries, etc?
[15:07] <Koon> robilad: one other area of concern is headless running, so that we can have Java server programs that do not pull all of X in
[15:07] <robilad> i think maven is the largest hurdle atm - regardless of the app server.
[15:07] <Koon> we need to fix the depends of most java libraries
[15:07] <robilad> yeah, that's something koon's started poking at in the bug database
[15:08] <persia> maven feels like a tool to build stuff, but I think we want to get a list of what gets built with maven so we can begin to prioritise.
[15:08] <Koon> robilad: is there a definitive way of saying a library doesn't require a full JRE ?
[15:08] <robilad> i'll poke the glassfish team for their own dep list -
[15:08] <persia> There's not much time until FeatureFreeze, and we don't want to have a delay in moving forward once we have maven.
[15:08] <robilad> though if you read their dev list, they are stuggling with maven daily, too. :/
[15:09] <robilad> feature freeze is on?
[15:09] <persia> 28th August.
[15:09] <persia> https://wiki.ubuntu.com/IntrepidReleaseSchedule
[15:10] <robilad> ok - i'll dig into analyzing some deps, then.
[15:10] <persia> robilad: Cool.  Do you think you'll have time to craft a task list for the next meeting?
[15:10] <robilad> there is no definitive way of pinning down libratry deps
[15:10] <robilad> since trhrough reflection any java code could in theory call any other code
[15:11] <persia> (or at least a draft task list)
[15:11] <slytherin> Can anyone give tell us in brief what is current state of either tomcat or glassfish?
[15:11] <robilad> if the code doesn't use reflection, you can run if through javap, dump it, and scan it for references to class it uses, though, and build a transitive closure that way.
[15:12] <robilad> persia: yes.
[15:12] <persia> That's probably a good start, and enough to let us make a guess as to whether we can reach intrepid.
[15:12] <persia> Thanks.
[15:12] <robilad> i'm back in hamburg after two weeks of travel, so ... phew. ;)
[15:12] <persia> OK.  Next up: slytherin: how is the progress of getting more of the stack into universe?
[15:13] <robilad> Koon: but for code that does not use reflection (java.lang.reflect & Class.forName, etc.): yes.
[15:13] <slytherin> persia: jboss had kept me busy. But I guess I will move on to 'MoveToUniverse' task over weekend.
[15:14] <persia> slytherin: jboss is in, or still FTBFS?
[15:14] <slytherin> persia: still FTBFS. Haven't got time for last 2-3 days to debug.
[15:15] <slytherin> I think I will upload libjboss-cache1-java to latest upstream version sometime next week and see if it builds.
[15:15] <persia> slytherin: OK.  For the MoveToUniverse, even if you can't make much progress, if you could get together a task list, maybe some of the rest of us could help as well.
[15:16] <slytherin> persia: Sure. That will be done by tomorrow.
[15:16] <persia> (I'd at least like to see us catch up to Debian)
[15:16] <persia> Excellent.
[15:16] <Koon> what version of jboss is that ?
[15:16] <slytherin> persia: I will add the packages  that I think should be evaluated on the MoveToUniverse page.
[15:18] <persia> slytherin: OK.  I'll take a look at that page on Saturday and see what I can push.
[15:19] <persia> Moving on unless there is more JBoss discussion (as there seems to be no response).
[15:20] <persia> Koon: Where are we with maven?  Still with the draft implementation needing review and assistance with the .pom files?
[15:20] <Koon> I drafted most of the spec, it's waiting for comments
[15:21] <persia> robilad: Now that you're back from your travel marathon, would you have time to review/comment the spec?
[15:21] <Koon> on the subject of implementation, unfortunately I cannot commit any more time at that atm, so we might need someone to pick it up
[15:22] <Koon> I still have to write the email to ubuntu-java to explain the work done so far
[15:22] <robilad> persia: yeah, looked over it, like it
[15:22] <Koon> it's mostly the jpp implementation + a few things to improve to be more Debian-like
[15:23] <persia> OK.  Does anyone have any issues with the spec, or shall we go ahead with it, pending availablility of someone to implement?
[15:23] <Koon> The patched maven2 is ready for someone to test it and bring it to the next step :)
[15:24] <robilad> nice - ppa?
[15:24] <Koon> yep - https://launchpad.net/~tcarrez/+archive
[15:24] <persia> Koon: Is there a bug against maven, or does it need REVU?
[15:25] <Koon> persia: it's not complete yet -- still has those funnt looking Fedira directories
[15:25] <Koon> Fedora
[15:25] <persia> Ah.  Still needs cleanup then :(
[15:25] <Koon> it's ready for some fake testing
[15:25] <Koon> set up the directories, populate them with artificial POMs
[15:25] <Koon> get a basic maven project and try to have it build from source
[15:26] <persia> OK, so at this point we mostly need some maven users to check it out and make sure it mostly works?
[15:26] <Koon> yes, to validate that the patches are even working.. because the only thing I did is to integrate them in a package
[15:26] <Koon> it's clearly not ready for any REVU :)
[15:27] <Koon> it would be nice to have some maven guy to set up the basic helloworld project
[15:27] <robilad> any simple maven project we could try it with?
[15:27] <Koon> I don't know so much about maven
[15:28] <slytherin> how about testing it with struts 1.3
[15:28] <persia> slytherin: Does that use maven?
[15:28] <Koon> just remember that you should artificially deploy the poms and depmap fragments that would have been installed with your JARs
[15:28] <slytherin> Koon: Ubuntu doesn't have jpp right? So how does your patched maven work exactly?
[15:29] <slytherin> persia: yes, the main reason why it has not landed in Debian/Ubuntu
[15:29] <Koon> slytherin: it integrates the jpp patchset, which is just telling maven to llok into specific directories and depmap on your system, if called with a specific -D arg
[15:30] <robilad> Koon: i poked paulcager about pinging you on #ubuntu.java about it
[15:30] <slytherin> Koon: so it means it will simply look in /usr/share/java and if jar files are not found then exit with error right?
[15:30] <robilad> he's the debian maven2 guy, afaict.
[15:30] <Koon> so you can install it from the PPA, create by hand the expected directories (like the dirty /etc/maven/depmap-fragment) and put into it what would have been installed by the library if it was maven2-jpp-compliant
[15:31] <Koon> slytherin: it's a bit more complicated than that
[15:31] <Koon> slytherin: did you read https://wiki.ubuntu.com/JavaTeam/Specs/MavenSupportSpec ?
[15:32] <slytherin> Koon: yes, but didn't fully understand it.
[15:32] <Koon> slytherin: I don't explain it so well. Deepak explanation was way better, if you can retrieve it from the logs
[15:32] <slytherin> Koon: Ok. I will check again.
[15:33] <Koon> it sounds complicated (and it is) but it's probably the only way around the maven problem
[15:33] <Koon> looks like a pile of hack-ish workarounds to make maven2 behave
[15:36] <persia> http://irclogs.ubuntu.com/2008/07/24/%23ubuntu-meeting.html has the discussion from dbhole
[15:36] <robilad> something we have to go through, I am afraid.
[15:37] <Koon> robilad: or convince all maven-using upstream to drop it :)
[15:37] <Koon> robilad: so far almost everyone I contacted in large Java projects seem to question the use of it
[15:37] <slytherin> Koon: so does your package works in complete offline mode? I mean 'new-go-online-no-matter-what'? And can you paste your patch for review somewhere?
[15:37] <robilad> aeh - an FSF campaign about the maven trap? ;)
[15:38] <robilad> i'd be happier with convicing the maven upstream to get such patches in and improve their tool.
[15:39] <lenards> amen
[15:39] <robilad> i haven't spoken with brettporter in a while, though.
[15:39] <Koon> slytherin: yes. As for the patch, it's just the Fedora patches, fixed to apply on our maven2. You can review it in aforementioned PPA
[15:39] <robilad> and i'm kind of hopiing that jvanzyl's move to eclipse with his company around maven will give the project a bit more ... focus.
[15:39] <Koon> I repeat, the package in that PPA is not a good package, it's a PoC
[15:39] <Koon> just to help someone getting started with testing
[15:40] <Koon> I could not find time so far to bring it to the next necessary step
[15:40] <robilad> there's even been talk about maving getting signatures for jars recently on the maven lists. ;)
[15:40] <robilad> and security, and waht not.
[15:40] <persia> While that's a nifty idea, it makes it even harder to properly support using maven as a build tool within a distribution.
[15:41] <robilad> anyway - my experience is similar to Koon's, talking about maven is a bonding experience of rants ;)
[15:41]  * lenards chortles
[15:43] <robilad> persia: i know - i keep telling them that security and recycling their current archive don't mix, but I don't think I'm getting through. anyway.
[15:43] <persia> Right.  On that note, let's move on.
[15:43] <persia> robilad: It's an education process :)
[15:43] <robilad> been at is since 2003 ;)
[15:43] <persia> So, anyone have any other items they want to add to the agenda at the last moment?
[15:46] <persia> No?  OK.  We've had poor luck with this recently, but who would be willing to write up the minutes?
[15:46] <robilad> me
[15:47] <robilad> now that I'm back here again. ;)
[15:48] <persia> robilad: Thank you :)  We've missed your excellence in that regard.
[15:48] <persia> OK.  See you all next week.  Quick action item review:
[15:49] <persia> robilad will prepare some information on next steps post-maven for the next meeting
[15:49] <persia> slytherin will update the MoveToUniverse page tomorrow with current candidates
[15:49] <persia> persia will push MoveToUniverse candidates on Saturday
[15:50] <persia> Koon will post a summary of the PoC maven-jpp stuff to the mailing list this week
[15:50] <persia> DId I miss anything?
[15:50] <robilad> i don't think so  - thanks for leading the meeting, persia.
[15:51] <persia> Thanks for attending.
[15:51] <Koon> Thanks persia !
[16:55] <emgent> @schedule rome
[17:12] <slangasek> Keybuk: I'm not on the MIR team, but I'm not aware of a backlog?
[17:36] <VdA> hi
[20:10] <Mez> @now
[21:23] <nellery> @schedule Vancouver
[21:38] <nizarus> @schedule
[22:16] <pwnguin> @schedule Chicago
[22:16] <pwnguin> is that right?
[22:18] <nellery> http://fridge.ubuntu.com/node/1614
[22:18] <nellery> if my conversions are correct, it should be
[22:19] <pwnguin> it want last time =/
[22:55] <keffie_jayx> @schedule caracas
[22:56] <nizarus> @schedule Loco
[22:56] <nizarus> @schedule Tunis
[23:41] <j_ack> @schedule Berlin
[23:56] <nizarus> hi boredandblogging, can i ask for an advice ?