| popey | hmmm | 00:44 |
|---|---|---|
| * popey taps the microphone | 00:44 | |
| popey | hi Yellow_Stevej | 00:48 |
| Yellow_Stevej | hello | 00:48 |
| Yellow_Stevej | well that seems to work! | 00:49 |
| === ubottu changed the topic of #ubuntu-meeting to: Calendar: http://fridge.ubuntu.com/event | Logs: https://wiki.ubuntu.com/MeetingLogs/ | 18 Sep 00:30: ubuntu-pa LoCo Team | 19 Sep 04:00: MOTU Team | 19 Sep 22:00: Platform Team | 22 Sep 04:00: Arizona LoCo IRC | 23 Sep 15:00: Server Team | 24 Sep 17:00: QA Team | ||
| === ubottu changed the topic of #ubuntu-meeting to: Current meeting: ubuntu-pa LoCo Team Calendar: http://fridge.ubuntu.com/event | Logs: https://wiki.ubuntu.com/MeetingLogs/ | 19 Sep 04:00: MOTU Team | 19 Sep 22:00: Platform Team | 22 Sep 04:00: Arizona LoCo IRC | 23 Sep 15:00: Server Team | 24 Sep 17:00: QA Team | ||
| === ubottu changed the topic of #ubuntu-meeting to: Calendar: http://fridge.ubuntu.com/event | Logs: https://wiki.ubuntu.com/MeetingLogs/ | 19 Sep 04:00: MOTU Team | 19 Sep 22:00: Platform Team | 22 Sep 04:00: Arizona LoCo IRC | 23 Sep 15:00: Server Team | 24 Sep 17:00: QA Team | 24 Sep 22:00: Platform Team | ||
| === Hobbsee` is now known as Hobbsee | ||
| freeflying | @schedule shanghai | 11:08 |
| ubottu | Schedule for Asia/Shanghai: 19 Sep 12:00: MOTU Team | 20 Sep 06:00: Platform Team | 22 Sep 12:00: Arizona LoCo IRC | 23 Sep 23:00: Server Team | 25 Sep 01:00: QA Team | 25 Sep 06:00: Platform Team | 11:08 |
| persia | #startmeeting | 13:01 |
| MootBot | Meeting started at 07:00. The chair is persia. | 13:01 |
| MootBot | Commands Available: [TOPIC], [IDEA], [ACTION], [AGREED], [LINK], [VOTE] | 13:01 |
| persia | [LINK] https://wiki.ubuntu.com/MobileTeam/Meeting/2008/20080918 | 13:01 |
| MootBot | LINK received: https://wiki.ubuntu.com/MobileTeam/Meeting/2008/20080918 | 13:01 |
| persia | Welcome to the Mobile Team meeting. The agenday is as listed at the above URL. | 13:01 |
| persia | First item: Roll Call. Who's here? | 13:01 |
| persia | OK. Moving on then. | 13:03 |
| persia | [TOPIC] Action Items | 13:03 |
| MootBot | New Topic: Action Items | 13:03 |
| StevenK | What was that bit? | 13:03 |
| persia | Last bit was Roll Call. Asking who was here. | 13:03 |
| cgregan | cgregan | 13:04 |
| lool | Hi | 13:04 |
| persia | So, first up: ogra: status of Q1 touchscreen support | 13:04 |
| lool | ogra is logging in | 13:05 |
| lool | <ogra> I'll do it next week | 13:05 |
| persia | OK. | 13:05 |
| persia | Next persia to register spec for installer | 13:05 |
| persia | See https://blueprints.launchpad.net/ubuntu/+spec/ubuntu-mid-live-installer | 13:05 |
| persia | Next: choice of mail client: modest vs. thunderbird. | 13:06 |
| * lool made no progress whatsover on modest | 13:06 | |
| lool | At this point we should just keep thunderbird and revisit modest next cycle | 13:07 |
| lool | It's too much WIP tright now | 13:07 |
| persia | OK. Shall we just agree on thunderbird for intrepid then, and look at modest for next cycle? | 13:07 |
| persia | Any objections? | 13:07 |
| cgregan | +1 | 13:07 |
| persia | [AGREED] Ubuntu MID to use Thunderbird as the default mail client for intrepid | 13:08 |
| MootBot | AGREED received: Ubuntu MID to use Thunderbird as the default mail client for intrepid | 13:08 |
| persia | [TOPIC] Current Items | 13:08 |
| MootBot | New Topic: Current Items | 13:08 |
| persia | There's a few things listed here. | 13:08 |
| persia | amitk: anything you have to say about kernel issues? | 13:08 |
| persia | Hmm. I think perhaps he's not here. | 13:09 |
| persia | Next: cgregan: Anything on the QA front? | 13:09 |
| cgregan | Just case work | 13:10 |
| cgregan | continuing | 13:10 |
| persia | Are you blocked by anything? | 13:10 |
| cgregan | need a spec for installer | 13:10 |
| cgregan | :-) | 13:10 |
| lool | haha :) | 13:10 |
| persia | cgregan: How much detail do you need? | 13:11 |
| cgregan | persia: Similar to StevenK's | 13:11 |
| persia | OK. | 13:11 |
| cgregan | Also, I might be mostly offline next week | 13:12 |
| persia | [ACTION] persia to flesh out ubuntu-mid-live-installer | 13:12 |
| MootBot | ACTION received: persia to flesh out ubuntu-mid-live-installer | 13:12 |
| cgregan | potentially getting sent out your way persia | 13:12 |
| * ogra waves | 13:12 | |
| persia | OK. Next up: StevenK (but without any specific items for discussion: very confusing agenda) | 13:13 |
| StevenK | Oh, I'm on the agenda? | 13:13 |
| persia | StevenK: Yep. You're listed under "Current Items". Do you have any "Current Items"? | 13:14 |
| StevenK | What about me? :-) | 13:14 |
| Hobbsee | i'm sure people can suggest what should be on the agenda w.r.t. StevenK | 13:14 |
| persia | heh | 13:14 |
| Hobbsee | maybe that he has to fix all the bugs by tomorrow, or something. | 13:14 |
| ogra | show off your screenshots :) | 13:14 |
| ogra | !! | 13:14 |
| StevenK | I did | 13:14 |
| * lool didn't get them | 13:15 | |
| * persia points at https://wiki.ubuntu.com/MobileTeam/Meeting/2008/20080918 again in case people didn't see the LINK earler. | 13:15 | |
| StevenK | I reimplemented a new launcher in one day, replacing mobile-basic-flash | 13:15 |
| StevenK | [LINK] http://people.ubuntu.com/~stevenk/IMG_2095.JPG | 13:15 |
| MootBot | LINK received: http://people.ubuntu.com/~stevenk/IMG_2095.JPG | 13:15 |
| persia | Woo Hoo! | 13:15 |
| lool | So you got python working? | 13:15 |
| StevenK | [LINK] http://people.ubuntu.com/~stevenk/IMG_2096.JPG | 13:15 |
| MootBot | LINK received: http://people.ubuntu.com/~stevenk/IMG_2096.JPG | 13:15 |
| persia | How much NEW do you need? | 13:15 |
| StevenK | So we go from 1,800 lines of C to about 70 lines of Python | 13:15 |
| StevenK | persia: Two source NEW, 3 or so binary NEW | 13:16 |
| lool | StevenK: what was the issue with pythonhildon? | 13:16 |
| === user__ is now known as davidm | ||
| StevenK | lool: It builds two binary packages, and only one of them was installed. | 13:17 |
| lool | StevenK: Do you provide new .menu files? | 13:17 |
| lool | I think it would be nice to define All, Home, and what we care about for menu-filters in our own menu files if we have a serious xdg menus parser | 13:18 |
| persia | Definitely. I'm up for generating .menu files if nobody else wants it. | 13:19 |
| lool | Generating? | 13:19 |
| persia | Yes, generating. We'd have to construct them, no? | 13:19 |
| lool | I'm speaking of /etx/xdg/menus/*.menu files | 13:20 |
| lool | LIke /etc/xdg/menus/applications.menu | 13:20 |
| persia | Yes, I also. | 13:20 |
| StevenK | I think the first step is to get the thing into the archive. | 13:20 |
| persia | That works. | 13:20 |
| persia | StevenK: Do you want an action? | 13:20 |
| StevenK | I was defeated by hildon-desktop today since mobile-basic-flash is hard-coded in a few places. | 13:20 |
| StevenK | I'll go through it next meeting, an action probably means I do it sooner | 13:21 |
| lool | StevenK: The h-d config files use mbf | 13:21 |
| lool | So yes, we need to fix them | 13:21 |
| persia | [ACTION] StevenK to get new xdg-compliant launcher into the archives | 13:21 |
| MootBot | ACTION received: StevenK to get new xdg-compliant launcher into the archives | 13:21 |
| lool | I think we need to fix the hd build in general | 13:21 |
| StevenK | lool: Right. And we can't rebuild hildon-desktop | 13:22 |
| lool | The hack of providing mbf etc. is plain ugly and doesn't solve the real issue that hd doesn't build | 13:22 |
| lool | StevenK: You can grep on HildonFile and hildon-file to see what hd is using from hildonfm | 13:22 |
| lool | Hopefully not too much | 13:22 |
| StevenK | I have two quick hacks in my launcher that provide mobile-basic-flash files, so it should work | 13:22 |
| lool | FIlechoosers are easy to replace, it's compatible with the gtk ones | 13:23 |
| lool | StevenK: We need to fix the hd desktop build, think of security updates for instance | 13:23 |
| persia | Can we even distribute it if we can't build from source? | 13:23 |
| StevenK | I guess | 13:23 |
| lool | Err we should just fix the build, not look at removing it; we use it | 13:23 |
| StevenK | lool: I'm not sure how to do that. I can post my build log | 13:23 |
| lool | StevenK: grep has this -r flag which allows for grepping "recursively" so to speak | 13:24 |
| lool | it will descend in directories and grep all files for the pattern | 13:24 |
| lool | So it will basically grep the whole source | 13:24 |
| lool | And so you can tell which files "match the pattern" | 13:24 |
| lool | so to speak | 13:24 |
| StevenK | I know that | 13:25 |
| lool | StevenK: So what would be nice is adding a configure.in flag to build hd without hildonfm | 13:26 |
| lool | StevenK: And adding some ifdefs for this | 13:26 |
| persia | I think we're getting deep into specifics. Maybe the two of you could discuss this towards an implementation for the next meeting? | 13:27 |
| lool | StevenK: Could you try doing this in the next week and throw technical questions at me if you get stuck? | 13:28 |
| StevenK | Sure | 13:29 |
| persia | OK. StevenK: anything else? Does anything block you? Do you need anything from someone? | 13:30 |
| StevenK | Nope, I'm good | 13:30 |
| persia | OK. Next is me talking about the installer. | 13:30 |
| * lool pumps up the volume | 13:31 | |
| lool | I heard we have working installer in images since monday? | 13:31 |
| persia | Bug #182004 was milestoned for beta, so there's a very good chance we'll be able to have the preseeding working properly for beta images. | 13:31 |
| ubottu | Launchpad bug 182004 in ubiquity "partitioner fails if partially preseeded due to seen flag madness" [Medium,In progress] https://launchpad.net/bugs/182004 | 13:31 |
| * davidm listens with interest | 13:31 | |
| persia | There was a report (I don't have the bug number) about issues with installing grub. I've made some changes to the platform.intrepid seeds that ought fix that, but am still testing the implications thereof, and trying to understand exactly what isn't working. | 13:32 |
| lool | So is it correct that we have a working installer included in the dailies? | 13:32 |
| persia | I'm currently blocked by the state of the linux-lpia and linux-meta-lpia packages, but have been working on them to try to get something that meets our needs. | 13:33 |
| lool | If it's not preseeded, I guess it just ask more questions? | 13:33 |
| persia | lool: Yep. | 13:33 |
| lool | Cool \o/ | 13:33 |
| persia | The problem is that it enforces a password for the "ubuntu" user, which is contrary to the behaviour we wanted. Only with working preseeding can this be skipped. | 13:33 |
| lool | Can you just remove the partman preseeded Qs? | 13:34 |
| persia | There's also the chance that someone could change the username, which would break the way we launch X, and we don't honor the GDM auto-login flag that ubiquity can set. | 13:34 |
| persia | No. partman fails under --automatic even when you don't preseed it. IT has to do with the way that partman and ubiquity interact. | 13:34 |
| persia | I don't actually want to preseed partman, but we need 182004 fixed in order to preseed other stuff in --automatic mode, even when we ask the partitioning questions. | 13:35 |
| cjwatson | yeah, I'll be trying to sort out 182004. I have some ideas | 13:35 |
| persia | cjwatson: Oh. I thought Evan was working on it. If you're both chasing it, that's even better news. Thanks. | 13:36 |
| cjwatson | actually partman will succeed if you *do* preseed it | 13:36 |
| cjwatson | it's when you don't and use --automatic that you have a problem | 13:36 |
| cjwatson | if the blocker is simply knowing how to preseed it properly, I can help with that | 13:36 |
| persia | Ah. So if we forced formatting the entire drive (as was done for hardy), it would work with --automatic today? | 13:36 |
| cjwatson | yes | 13:36 |
| persia | OK. I'll try that for a few days, and we can see if we like it. I'll work with the installer team to make sure I have the right hints. | 13:37 |
| persia | Thanks cjwatson. | 13:37 |
| lool | cjwatson: Do you think this will be fixed for intrepid's release, or shall be best switch to preseeding the partionning right now? | 13:37 |
| cjwatson | you should be able to remove the password preseeding without trouble | 13:37 |
| cjwatson | lool: I'm reasonably hopeful of the former | 13:37 |
| lool | Cool | 13:37 |
| cjwatson | we might not get it perfect but we'll at least work around the immediate problems | 13:37 |
| lool | ogra: I hope you'll pull the installer in the mobile images as well :-) | 13:38 |
| persia | Actually, there aren't any mobile images, for any architectures, but let's get to that in the next items. | 13:38 |
| persia | I'm done. | 13:38 |
| ogra | lool, absolutely :) | 13:38 |
| persia | Next up is lool, who is also confusingly listed in the agenda without any specific items. | 13:38 |
| lool | I think it's just the status reports which we wanted to move on top of the agenda | 13:39 |
| ogra | but please note that my image build wont happen before monday, i simply wont find the time before sunday to work on the patches for livecd-rootfs and for StevenK's wrapper | 13:39 |
| lool | Anyway, status is I spent a lot of time on elisa which is looking great | 13:39 |
| lool | I'd love if we could use it, but there's a catch | 13:39 |
| lool | it requires GL | 13:39 |
| persia | That's less than ideal for a number of devices :) | 13:39 |
| lool | So it's going to be problematic for poulsbo and virtual images | 13:40 |
| lool | I guess it's a non issue for intel chips though | 13:40 |
| ogra | lool, perfect for ubuntu-mobile then :) | 13:40 |
| lool | I also spent time on misc hardy stuff and am at OSIM and maemo summit this week | 13:40 |
| ogra | the launcher currently requires GL anyway | 13:40 |
| lool | So not much is going to be listed in my AR | 13:40 |
| lool | I have many high priority stuff to fix for hardy and intrepid | 13:41 |
| ogra | so we should probably provide elisa as option ;) | 13:41 |
| lool | ogra: Indeed; you need to make sure they interact nicely though | 13:41 |
| persia | Well, anything in the repos is an option. | 13:41 |
| lool | ogra: I know elisa works badly with compiz right now | 13:41 |
| ogra | lool, well, the image is a first shot thing anyway ... | 13:41 |
| ogra | the launcher too ;) | 13:41 |
| lool | The other options for media playback would be totem and moblin-media | 13:42 |
| ogra | and maximus is a metacity extension | 13:42 |
| lool | I don't consider vlc, mplayer or xine integrated enough, but they work fine for me | 13:42 |
| ogra | cant use compiz on ubuntu-mobile atm | 13:42 |
| persia | Given the current state of moblin-media, shall we look at getting totem into the images and tested? | 13:42 |
| lool | That sounds good | 13:42 |
| lool | StevenK: Does moblin-media still fail to start after my python-hildon upload? | 13:42 |
| lool | Anyway, not much more to report; meeting interesting people here | 13:43 |
| StevenK | I think it's fixed | 13:43 |
| lool | Ok; so moblin-media remains an option | 13:43 |
| lool | It's UI is quite simplistic, but totem is decent | 13:43 |
| persia | The missing part of moblin-media is the photo app. Do we want fspot? gthumb? | 13:43 |
| lool | I'm happy either way | 13:43 |
| lool | Is antyhing else pulling mono? | 13:44 |
| persia | moblin-media doesn't have any icons, last I looked. Do we have icons? | 13:44 |
| persia | Nothing else is pulling mono | 13:44 |
| lool | I'm not sure fspot isn't a bit too rich for MIDs; perhaps for netbooks | 13:44 |
| persia | -mobile uses fspot. | 13:44 |
| lool | No strong opinion here either | 13:44 |
| lool | I like fspot, gthumb doesn't really offer the same features, but then pulling mono is expensive | 13:45 |
| persia | I'd prefer gthumb for -mid just for the reduced depedencies, but I'm on a likely useless quest to fit it on a 512MB stick. | 13:45 |
| persia | cgregan: Would you have time to test and compare the two, and make a recommendation? | 13:46 |
| cgregan | sure | 13:46 |
| persia | [ACTION] cgregan to compare F-Spot and GThumb on MID and suggest which to use as a replacement for Photos. | 13:46 |
| MootBot | ACTION received: cgregan to compare F-Spot and GThumb on MID and suggest which to use as a replacement for Photos. | 13:46 |
| persia | lool: Anything else? Anything blocking you? Do you need anything from anyone? | 13:46 |
| lool | No, as I was saying earlier I'm done reporting; nothing blocking me | 13:47 |
| persia | OK. Next is ogra, also without specifics. | 13:47 |
| lool | Except my TODO list keeps getting longer and it's not going to improve this week :-/ | 13:47 |
| StevenK | Hah | 13:47 |
| ogra | well, behind on touchscreens, but done with the Q1 keyboard remapping | 13:47 |
| * lool would like to report that ogra has been smoking; it's very bad for his health | 13:48 | |
| * ogra still hopes anyone can test that apart from himself | 13:48 | |
| ogra | two cigarettes today ! | 13:48 |
| ogra | come on | 13:48 |
| lool | ogra: You might mention this critical text rendering bug you have with -mobile | 13:48 |
| persia | ogra: I think we need images. Any objections if I generate -mobile live images? | 13:48 |
| lool | ogra: Do you have a bug id? | 13:48 |
| ogra | yeah, the mobile desktop is largely done, still needs a patch i only have locally for the netbook-launche and has issues with font rendering atm | 13:49 |
| lool | persia: You have cdimage access? | 13:49 |
| persia | lool: No, but I have been reading all the relevant code closely recently, and so can probably put together the set of changes required fairly quickly. | 13:49 |
| persia | Turning code that generates images into cdimage images is a simpler matter. | 13:49 |
| ogra | lool, no, but there is one open, i have to look it up | 13:50 |
| lool | ogra: Ok, good that it's file | 13:50 |
| lool | *filed | 13:50 |
| ogra | yeah | 13:50 |
| ogra | njpatel is aware of it and ties to fix it before release | 13:50 |
| ogra | i'll bug him next week again | 13:50 |
| lool | persia: Ok; ogra has a lot of time between cigarettes and has cdimage access and knowledge (per cmpc work) | 13:50 |
| ogra | ppppft | 13:50 |
| * ogra decides to go for a smoke after the meeting ... all these mentioning of cigarettes ... | 13:51 | |
| persia | OK. I won't generate -mobile images then :) | 13:51 |
| lool | persia: I'd be tempted to see you work on critical installer issues prior to mobile images, but that's probably just me | 13:51 |
| lool | It's not like we in the team feel like we can help on installer issues much; at least I feel like it would take me some hours or a day to grasp where you're stuck and start working on fixes | 13:52 |
| persia | I think that's just you. The only "critical" issue is beyond me, and I'll prepare a workaround in the next day or so. | 13:52 |
| ogra | the mobile images are trivial to get, its really only copying the lines of -mid in livecd-rootfs | 13:52 |
| persia | Yeah. Just didn't know where you were for time. | 13:52 |
| lool | Where who was for what? | 13:52 |
| ogra | what ? | 13:53 |
| lool | persia: Could you rephrase your last line? | 13:54 |
| persia | ogra: I agree that the image generation change is relatively trivial, I just wasn't sure if you would have time to complete it in the next day or two. | 13:54 |
| lool | 6 minutes warning | 13:54 |
| persia | Right. | 13:54 |
| persia | ogra: Anything else? Anything blocking you? Do you need anything from anyone? | 13:54 |
| ogra | persia, i will prepare it on sunday | 13:55 |
| ogra | nothing from the top of my head ... | 13:55 |
| * lool just realized we don't have a status report for davidm | 13:55 | |
| ogra | the conference is slightly distracting from work here | 13:55 |
| lool | We should take some time to ask him how things are going like for the rest of us | 13:55 |
| persia | lool: Add it to the agenda next time :) | 13:55 |
| ogra | yeah | 13:55 |
| lool | I will | 13:55 |
| persia | Anyway, moving on | 13:55 |
| ogra | make him type on the n800 :) | 13:55 |
| persia | [TOPIC] Everything else | 13:55 |
| lool | *10 | 13:55 |
| MootBot | New Topic: Everything else | 13:55 |
| persia | Anyone have anything else they want to discussi? | 13:56 |
| lool | Germans are crazy | 13:56 |
| * ogra sees davidm starting to type | 13:56 | |
| persia | That's not really on-topic for this meeting :) | 13:56 |
| lool | ogra: It will take a while with the tiny keyboard :-P | 13:56 |
| ogra | not as crazy as the french :) | 13:56 |
| MacSlow | lool, maybe ... but I always try not to be ;) | 13:56 |
| ogra | yeah | 13:56 |
| lool | persia: "everything else" sounded pretty largely inclusive | 13:56 |
| persia | lool: Well, yeah. It's still supposed to be constrained by "Mobile Team Meeting". | 13:57 |
| lool | Ah right | 13:57 |
| lool | But we're almost in the desktop team meeting, we need some kind of transition | 13:57 |
| lool | See MacSlow picked it up | 13:57 |
| persia | ian_brasil mentioned some ideas for theme changing in IRC, but I've not heard of a specific patch yet. | 13:57 |
| ogra | well, lool could put that comment under "conference report" :) | 13:57 |
| lool | Anyway, thanks for chairing, persia: | 13:57 |
| lool | s/:/! | 13:57 |
| davidm | the show is going well 1/2 of the people are interested in arm ubuntu which we don't currently have. | 13:57 |
| persia | Right. Ending the meeting then. | 13:57 |
| davidm | #endmeeting | 13:58 |
| persia | #endmeeting | 13:58 |
| MootBot | Meeting finished at 07:58. | 13:58 |
| ogra | not so fast, david needs a minute per sentence :) | 13:58 |
| lool | I wonder what ogra will do in the 2 extra minutes | 13:58 |
| lool | haha | 13:58 |
| * lool disappears again & | 13:58 | |
| persia | ogra: The trick to using a MID to participate in an IRC meeting is to have a large text buffer from which one can cut & paste. | 13:58 |
| ogra | hehe | 13:59 |
| ogra | i usually do that | 13:59 |
| MacSlow | re | 14:00 |
| === nxvl_ is now known as nxvl | ||
| davidm | #endmeeting | 14:00 |
| persia | davidm: I already did, and you can't because you're not chairing | 14:02 |
| ogra | persia, he's playing with the up arrow key :) | 14:03 |
| persia | Ah! | 14:03 |
| StevenK | Heh | 14:04 |
| davidm | :-) | 14:04 |
| davidm | the N810 is very nice | 14:05 |
| davidm | but hard to type on, no touch typing! | 14:05 |
| davidm | thumbs are even hard to use | 14:06 |
| === Rafik__ is now known as Rafik | ||
| persia | OK. Who's here for the Java meeting? | 15:01 |
| Koon | o/ | 15:01 |
| persia | Hmm. Let's wait a bit then and see if anyone else shows up. | 15:01 |
| persia | Well, apparently not. | 15:05 |
| persia | OK. Getting started then. | 15:05 |
| persia | Roadmap review: | 15:05 |
| persia | robilad and slytherin don't appear to be present, and Koon has a special agenda item about maven, so I think there's nothing special here. | 15:06 |
| persia | Next up: Koon, with Discuss KISS alternative design from MavenSupportSpec | 15:06 |
| Koon | hmm. that agenda item might date back from last week's meeting | 15:06 |
| persia | That's what I thought. I remember discussing it last week. | 15:06 |
| Koon | anyway, I had an action point about it, start a bootstrap analysis | 15:07 |
| Koon | https://wiki.ubuntu.com/JavaTeam/MavenKissBootstrap | 15:07 |
| Koon | I started it, this is the very first steps | 15:07 |
| * Koon reposts url for slytherin | 15:07 | |
| Koon | https://wiki.ubuntu.com/JavaTeam/MavenKissBootstrap | 15:07 |
| Koon | that doesn't look like so many packages to do | 15:08 |
| Koon | but there is still some bootstrap ordering to do. Like is it necessary to have maven-compiler-plugin to build maven-compiler-plugin | 15:08 |
| persia | What do the colors mean? | 15:08 |
| Koon | green means in the repositories | 15:08 |
| Koon | red means not in | 15:08 |
| Koon | on the version number, green means probably ok | 15:09 |
| Koon | yellow means potentially too different from required version | 15:09 |
| Koon | (and yes, running a simple compile makes maven download three different version of plexus-utils) | 15:09 |
| Koon | I mean... 4 different versions :) | 15:10 |
| persia | RIght. 4 different versions of plexus-utils is right out :) | 15:10 |
| Koon | so it's a finite number, but I'm not sure you can bootstrap it from source | 15:11 |
| Koon | I don't really get how to build maven-compiler-plugin | 15:11 |
| Koon | I need to push that analysis a little further | 15:11 |
| persia | So it's about 30 packages we need, and once we have them, we may need some bootstrapping to get them all built against themselves? | 15:11 |
| Koon | persia: i'm not exactly sure what's acceptable in that area (circular build deps) | 15:12 |
| persia | Circular build deps are bad, and should be avoided. | 15:12 |
| persia | When it's not possible to work around them, one generates a special bootstrap version of a package that "builds" in a clean environment, and uses that to build enough to be able to build a clean package, and then uses that to rebuild the deps, etc. until everything built is clean. | 15:13 |
| Koon | persia: you should still avoid to build A with a A build dep, I suppose | 15:14 |
| persia | Most critically, one wants to minimise the number of "bootstrapping" packages so that the majority is actually built cleanly. As such, the special bootstrapping package may not closely resemble the final form of that package. | 15:14 |
| persia | Koon: Building A with A as a build-dep is undesireable, but sometimes hard to work around (consider ghc6 or gcc). | 15:15 |
| Koon | anyway, more noext week | 15:15 |
| persia | Having A build-dep on B and C and B build-dep on A and C and C build-dep on A and B is considerably worse. | 15:16 |
| persia | For 30 packages, we'll probably need to find some strategy to get them working: maybe hack the build systems all to use ant for a first run, and then go back to upstream build systems and rebuild, or some such. | 15:16 |
| Koon | i'm pretty sure we can emulate what maven does for those | 15:17 |
| persia | Personally, given the number already present, I'm fairly confident we can find a way to build most of it with the existing tools, and then use this stack to build other things. | 15:17 |
| Koon | yes | 15:17 |
| persia | That avoids recursive bootstrapping. | 15:17 |
| persia | OK. slytherin: we skpped MoveToUniverse earlier. I notice there are still 10 bugs listed on the wiki page. Are any of them critical for beta, or are we basically done for intrepid? | 15:19 |
| slytherin | persia: done as in I am not going to log any more bugs. I am just continuing work on existing ones. | 15:21 |
| slytherin | I will probably convert few FFE bugs to simply "compile with openjdk' bugs. | 15:21 |
| persia | OK. Are any of the existing bugs stuck, or are they all going fairly smoothly? | 15:21 |
| slytherin | going fine. | 15:21 |
| persia | OK great. | 15:22 |
| persia | Anyone have anything else to raise, or shall we close the meeting? | 15:22 |
| Koon | just a thing | 15:23 |
| persia | Yes? | 15:24 |
| Koon | openjdk in alpha6 has different versions for amd64 and i386 | 15:24 |
| Koon | amd64 failed to build for some reason | 15:24 |
| Koon | https://launchpad.net/ubuntu/intrepid/+source/openjdk-6/6b12~pre1-0ubuntu3 | 15:24 |
| persia | Failed on sparc as well, which is somewhat unexpected. | 15:25 |
| Koon | I'm not sure what the safeguards are to avoid this in a more... definitive release | 15:25 |
| Koon | because on testing I get different results depending on arch | 15:25 |
| Koon | (the depends for -ubuntu2 are much heavier than for -ubuntu3 | 15:26 |
| Koon | ans that's... undesirable :) | 15:26 |
| Koon | an amd64 failed build doesn't prevent the i386 package to go to the archive and then on the CD ? | 15:27 |
| persia | http://people.ubuntu.com/~ubuntu-archive/testing-ports/intrepid_outdate.html and http://people.ubuntu.com/~ubuntu-archive/testing/intrepid_outdate.html are the current safeguards. | 15:27 |
| persia | These lists need to be reviewed and cleared. | 15:27 |
| persia | Mind you, those are only for main. Production of similar lists for universe and multiverse is being investigated, but may not be ready for intrepid release. | 15:28 |
| Koon | ok, was just a little worried when I saw that :) | 15:28 |
| persia | No, there's no such block: it's expected that all such situations like that can be fixed. | 15:28 |
| persia | segfault during documentation construction! My, that's annoying. | 15:29 |
| persia | Has anyone tried a local build to see if it's transient? | 15:29 |
| slytherin | local build of openjdk? you are kidding right? :-) | 15:29 |
| Koon | hmm. building openjdk is not... light :) | 15:29 |
| Koon | my VMs don't have enough RAM, tbh | 15:30 |
| Koon | they start swapping and hell breaks | 15:31 |
| persia | I wasn't kidding. I'd also recommend building in a chroot, rather than in a VM. | 15:32 |
| Koon | (my VM runs a pbuilder chroot) | 15:32 |
| slytherin | I don't have amd64 machine or I would have tried it. | 15:32 |
| persia | And I'm typically running too many VMs to also run that big a compile. I suppose we'll hope doko can fix it then. | 15:33 |
| persia | OK. Anything else? | 15:33 |
| Koon | nope | 15:33 |
| slytherin | I have one thing to say | 15:33 |
| slytherin | do we plan to test any frequently used java apps with openjdk. ex. azureus | 15:34 |
| persia | We've not identified any testing plans as a team. Do you think we ought organise some as we near release? | 15:35 |
| slytherin | yes, probably after beta. But I am not sure what all apps we should test. | 15:36 |
| persia | Do we have anything other than the browser plugin that is destined for any of the CDs? | 15:36 |
| persia | Koon: Did tomcat make a CD? | 15:36 |
| Koon | There is tomcat-server tasksel on the server Cd | 15:37 |
| slytherin | cool | 15:37 |
| Koon | I'm in the process of plugging a test case into iso.qa.ubuntu.com | 15:37 |
| * persia checks the livecd manifest again | 15:38 | |
| Koon | (that's why I was concerned about extra deps on the amd64 cd ;) | 15:38 |
| persia | Hmm. LiveCD doesn't even ship the browser plugin. | 15:39 |
| persia | OK. I claim testing tomcat is the priority. azureus might be a good second test. Any other packages that would be release-critical if they didn't work? | 15:40 |
| slytherin | persia: what is size of Live CD? browser plugin will need jre also so don't think tehy will fit on CD. | 15:41 |
| * slytherin has to go for meeting. | 15:41 | |
| persia | slytherin: Indeed. | 15:41 |
| persia | OK. Anything else to raise? | 15:42 |
| persia | OK. Ending the meeting then. | 15:43 |
| persia | Koon: please share how we can test tomcat for the next meeting | 15:44 |
| persia | slytherin: Please organise some test cases for azureus so we can get it tested. | 15:44 |
| persia | Have a good week everyone. | 15:44 |
| Koon | ok | 15:44 |
| === asac_ is now known as asac | ||
| === mc44_ is now known as mc44 | ||
| === MacSlow_ is now known as MacSlow | ||
| Arc | has anyone heard anything about the EMEA membership meeting reschedule? | 20:37 |
| stgraber | Arc: no planed date yet, sorry. | 20:57 |
| Arc | so I guess it's not going to be this week after all | 21:00 |
| leoquant | @schedule | 21:18 |
| ubottu | Schedule for Etc/UTC: 19 Sep 04:00: MOTU Team | 19 Sep 22:00: Platform Team | 22 Sep 04:00: Arizona LoCo IRC | 23 Sep 15:00: Server Team | 24 Sep 17:00: QA Team | 24 Sep 22:00: Platform Team | 21:18 |
Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!
