#title #ubuntu-meeting Meeting Meeting started by wendar at 18:06:28 UTC. The full logs are available at http://ubottu.com/meetingology/logs/ubuntu-meeting/2011/ubuntu-meeting.2011-10-28-18.06.log.html . == Meeting summary == *Review action items ''LINK:'' http://wiki.ubuntu.com/AppReviewBoard/Agenda (wendar, 18:06:43) ''LINK:'' http://people.debian.org/~bap/dfsg-faq.html (wendar, 18:21:19) ''LINK:'' http://opensource.org/docs/osd (wendar, 18:22:03) ''LINK:'' https://bugs.launchpad.net/ubuntu-app-review-board/+bug/881515/comments/4 (ajmitch, 18:47:03) ''LINK:'' https://blueprints.launchpad.net/ubuntu/+spec/community-p-app-review-board (wendar, 18:56:52) Meeting ended at 19:03:59 UTC. == Votes == == Action items == * (none) == People present (lines said) == * wendar (128) * ajmitch (65) * coolbhavi (22) * meetingology (3) * ubottu (1) == Full Log == 18:06:28 #startmeeting 18:06:28 Meeting started Fri Oct 28 18:06:28 2011 UTC. The chair is wendar. Information about MeetBot at http://wiki.ubuntu.com/AlanBell/mootbot. 18:06:28 18:06:28 Available commands: #accept #accepted #action #agree #agreed #chair #commands #endmeeting #endvote #halp #help #idea #info #link #lurk #meetingname #meetingtopic #nick #progress #rejected #replay #restrictlogs #save #startmeeting #subtopic #topic #unchair #undo #unlurk #vote #voters #votesrequired 18:06:43 [TOPIC] Review action items 18:06:43 [LINK] http://wiki.ubuntu.com/AppReviewBoard/Agenda 18:06:56 * ajmitch has not found any that are really ready to be voted on yet 18:07:37 We notified all old applications of the Oneiric update, so we can mark that done. 18:08:08 * coolbhavi is a bit new to this so getting into grips with it now 18:08:36 IIRC, the notice about old archive contents being copied over was taken care of, but stgraber was going to double-check on the longer-term implications 18:09:23 that is, right now, our process is that we don't put anything into the archive of the development release, so it's always empty when the new archive for the next development release is created 18:10:07 which is why there's no precise directory on extras.ubuntu.com/ubuntu/dists/ 18:11:16 hmm 18:12:21 so we've got a lot of new submissions for oneiric now, not all of them are great 18:12:35 yes 18:13:25 you've done some scanning this week 18:13:31 do we want to run through the list here, or just talk about general problems with them? 18:13:57 if we have any ready to go (I think not?) we can review here 18:14:16 but, general problems is probably more useful now 18:14:32 I noted a few of them in the reviewer instruction documents 18:14:50 source tarballs that can't be built 18:15:01 and binary packages instead of source packages 18:15:20 closest that I have ready to go are the askubuntu-lens again, blocked on a SRU which was uploaded to -proposed this week, and maybe harmonySEQ if the permission problems with it can be sorted (/dev/snd/seq) 18:15:43 the common problem with java submissions is submitting a tarball full of .jar files 18:16:12 one of those which I saw is even in a PPA like that, where the source package has multiple .jar files & debian/install to copy them into place 18:16:32 yeah 18:16:51 the short term solution is to individually guide developers to what we're looking for 18:17:03 but, longer term, we need documentation on these pieces 18:17:08 that problem is solvable, but they bundle a *lot* of upstream .jar files with differing licenses, not all of which are compatible 18:17:16 so, the developers can reach what we need independently 18:17:40 I have a small input here 18:17:43 educating developers on license compatibility is probably trickier than some technical documentation 18:17:58 coolbhavi: yes, that's great, go ahead 18:18:50 coolbhavi: new ideas are much needed and greatly appreciated :) 18:19:00 yes how about a elaborate explanation of dfsg and free licenses on a wiki page such that devs understand in a simple language? 18:19:44 it'd be useful to have something like that on developer.ubuntu.com 18:20:04 we have longer explanations in the Ubuntu and Debian documentation 18:20:18 but, if we can find a way to teach it more simply, that'd be great 18:20:27 you'd probably want to link to something like http://www.gnu.org/licenses/licenses.html 18:21:02 yes might be too trivial but I guess covering main parts of licenses so that it gets in a simplified language to help devs 18:21:19 http://people.debian.org/~bap/dfsg-faq.html 18:22:03 http://opensource.org/docs/osd 18:22:04 would be hepful... something similar or a link on developer.ubuntu.com might help 18:22:39 I think a combination of a clear and short explanation, with links to "more reading" at the bottom might be good 18:22:54 +1 here 18:22:58 that reminds me of the point I added on the arb blueprint - looking over developer.ubuntu.com & making sure that documentation is clear, and that it matches the processes 18:23:17 coolbhavi: you'll be here at UDS next week? 18:23:29 yes m flying tomorrow 18:23:37 ll be there :) 18:23:54 coolbhavi: great, I can connect you with David Planella, this is something we could get done this cycle 18:24:06 sure wendar 18:24:17 ajmitch: yes, an overall review of developer.ubuntu.com is needed 18:24:35 ajmitch: I've found a few spots (like the description of the ARB) that need some updates 18:24:55 ajmitch: not huge updates, but bits and pieces 18:25:20 yeah, things like app submission assuming that everything should have a price 18:25:31 yup 18:26:36 we'll need to sort out, probably in a dicussion at UDSm how much hand-holding & work we're going to do 18:26:38 I'll add those two as action items, and as topics for the session next week 18:26:48 yes, that too 18:27:11 Or why not make a check in the portal itself if its a free license or not ... If not the portal shouldnt allow someone to submit apps 18:27:33 coolbhavi: I think it already does that 18:27:58 it checks whether they say it's a free license 18:28:03 because the reviewing process is used for both commercial/proprietary & ARB apps 18:28:16 the ARB doesn't even look at it unless it's marked as a free software license 18:28:30 but, developers don't always understand what's needed 18:28:42 or, may have license problems internally 18:28:50 (that they didn't realize they had) 18:28:55 Oh okay I still do not have proper access to the portal so I havent checked 18:28:58 right, they can put a copy of the GPL in their tarball, but not submit source along with it 18:29:13 nor do I, I'm just going from the public parts that I can see :) 18:29:19 coolbhavi: yes, we'll get access set up next week 18:29:35 the new ARB queue is set up now 18:29:38 great 18:29:41 it has nothing in it now 18:29:41 will they still be copied into LP bugs? 18:29:50 but, we'll start moving things into it 18:30:07 ajmitch: we should be able to stop copying to LP bugs now 18:30:19 ajmitch: which will help a lot :) 18:30:33 ok, then is there a way that I could get a list of the queue withouth screenscraping developer.ubuntu.com? :) 18:30:51 but, I haven't run an app all the way through the new process yet 18:31:06 ajmitch: a list of the queue? 18:31:09 * ajmitch is using lplib at the moment to pull the list of bugs & make sure that there are notes locally or at least an entry about them on my local system 18:31:31 just something I hacked up very quickly to make sure that I can keep on top of them 18:31:35 ajmitch, something like sponsors queue? 18:31:54 what MyApps has now is a first approximation of our process 18:32:09 so, we'll need to try it out, and let the developers know if/where we need changes 18:32:33 wendar: right, I've subscribed to the developerportal bugs, I'll file them there 18:32:55 ajmitch: ah, cool, sounds like a useful tool 18:33:25 just looking at the list, there are a couple there that were up for review on the debian-mentors list, but didn't get anywhere 18:33:32 ajmitch: so, I'm sitting in a session right now on how we get good reports on how the queues are doing, and keep on top of them 18:33:40 wendar: ok :) 18:33:58 :) 18:34:03 ajmitch: so, it's very llikely we can get what you need from that tool built into the web interface 18:34:11 that would be nice 18:35:02 ajmitch: yeah, I want to make sure we don't fall back into REVU, where submissions sit for months or years without progress :( 18:35:39 it's a real risk, we're already falling behind the rate of submission :( 18:36:09 wendar: the one I'm looking at, openssn, even has packaging in the debian pkg-games repository which is closer to what we need than what was submitted to the ARB 18:36:34 interesting 18:36:47 I wonder if we should try & work with some of these to get them into debian 18:36:50 so, maybe the best recommendation is to help him get it into debian? 18:36:53 aye 18:36:59 I think that's quite valuable 18:37:11 help them find contacts or sponsors in Debian 18:37:27 grub-customizer fails the 'no system apps' part of the checklist, was also on REVU & on mentors.debian.net 18:37:28 also, I'm a DM and lfaraone is a DD 18:37:58 * ajmitch is a DD, not sure about the others 18:38:14 ajmitch: yeah, it modifies a pretty fundamental part of the system 18:38:41 we can ask cjwatson about it, and see if it's likely to have any chance in Debian 18:38:43 apart from that, it'd be nice to get in, get it synced to precise & then into -backports 18:39:18 yeah, if the code is good, I can see it being a valuable tool 18:39:44 it needs a grub-knowledgable sponsor 18:39:49 we also had listed somewherewe had listed somewhere a general guideline for size of applications, is that still around? 18:39:59 it's not a hard rule 18:40:07 sorry if I'm a bit all over the place this morning :) 18:40:10 but ~10,000 lines 18:40:30 ajmitch: np, that's what the IRC sessions are for, gathering "other business" 18:40:50 rstudio will need to go into the distro then, sloccount said it was about 100k 18:41:17 it's a mix of C++ & java, currently has no packaging 18:41:28 ah, that would be too bad, it's one of the best apps we've gotten :( 18:41:36 yeah 18:41:52 I was just surprised at the size of it 18:41:56 if it's clean code, and non risky, it could still be fine 18:42:43 (like, it's not modifying core systems, making network connections, etc) 18:42:45 it could be, but someone would need to step up & do the packaging - the java part is for a GWT web frontend 18:42:50 yeah, I'm surprized 18:43:29 iamfuzz said he'd be willing to do the packaging for it 18:43:35 great 18:44:32 though I still feel that something of that size & quality may be a better fit for the distro, rather than something to copy over every 6 months :) 18:45:20 yeah, it seems like the kind of thing we should mentor to move up into Debian 18:45:32 I think the developer would be open to that 18:45:47 but, getting it into extras for Oneiric is a good easy introduction 18:45:53 rstudio was the one where the submitter asked about building against newer versions of boost & qt - both of which aren't available yet in the distro 18:46:28 ah, I hadn't read the follow-up on that one 18:46:37 so, the versions in Oneiric weren't sufficient? 18:46:44 (I asked, didn't see the answer yet) 18:46:56 they're sufficiant for now 18:47:03 https://bugs.launchpad.net/ubuntu-app-review-board/+bug/881515/comments/4 18:47:04 looking at his README, it looked like the versions in Oneiric were what he needed 18:47:04 Launchpad bug 881515 in Ubuntu Application Review Board "Application Review Request: rstudio-v0.94" [Undecided,New] 18:47:32 ah, got it 18:47:40 so, the answer to that is "no, it's not possible" 18:47:52 I would not want to ever allow apps to bundle newer versions of libraries like that 18:48:02 so, we'd have to hold off updating in the distribution until the library is updated 18:48:39 ajmitch: want me to take that one? 18:48:45 sure 18:50:01 will do 18:50:01 it's going to be hard to keep track of who's looking at what if we split them up between us 18:50:29 the shepherd idea we used before helped with that 18:50:37 but, it also meant some submissions got no attention 18:50:45 if the shepherd ran out of time 18:50:55 something to talk about next week 18:51:20 (and, I'm not sure how that'll work out in the new interface) 18:51:28 I had a thought this week 18:51:35 depending on when the uds session is, I may not be able to attend remotely 18:51:40 do you know how the patch pilot's program works? 18:52:05 a roster of people who look at the sponsoring queue & pick items off it 18:52:09 instead of assigning responsibility by submission, we do it by time 18:52:11 yeah 18:52:23 a person agrees to a particular 4-hour period once a month 18:52:33 and, in that time they're available to answer questions 18:52:34 we'd probably want to track when something was last touched 18:52:45 and work through as much of the queue as possible 18:53:12 yup, we'd have to make sure there was available info on what had been done 18:53:19 so, the next pilot could pick it up 18:53:23 with LP we can set bugs to incomplete if we're waiting for info, and it lists the last changed on the overview page 18:53:25 why dont have a wiki page with all the TODO's and updates and ownerships to be updated every week maybe and the responsibility of updation is owned up by someone so we can have a spoc maintainability of tracker? 18:53:38 ajmitch: the new system has that too 18:53:45 wendar: great, was about to ask that 18:53:57 it gets hidden from the list needing review 18:54:03 until the developer responds 18:54:43 coolbhavi: that's a good idea, especially the interface we get isn't quite good enough to start with 18:54:43 coolbhavi: it may work, just depends on how awkward it is to keep in sync with new submissions 18:55:02 we've obviously seen quite a few new submissions since the new site went up 18:55:02 coolbhavi: it can be a pain to manually update, so hopefully we can automate that quickly 18:55:32 ideally, it'd be a report page in the interface, with no manual maintenance work 18:55:49 the simple tool that I hacked up is dead simple it just reads the statuses & bug links out of a .yaml file :) 18:56:07 ajmitch: cool 18:56:11 yaml just because it's close to freeform text 18:56:12 yes thats another part as we get more active, submissions will increase I guess. Agreed. 18:56:16 I haven't been updating statuses 18:56:39 (I have that on our list to talk about next week too, although we might not need it if we're switching away from LP) 18:56:47 I've mostly just been keepign my own statuses, such as ones I haven't looked at, those that need packaging done, etc 18:56:52 https://blueprints.launchpad.net/ubuntu/+spec/community-p-app-review-board 18:57:12 ajmitch: it'd be nice to get that somewhere we share it 18:57:23 ajmitch: is it crontab-able? 18:57:58 wendar: not particularly, I'll need to spend more than 30 minutes on it 18:58:15 I do feel the pain, I totally missed a response from one of the developers on a package I was shepherding :( 18:59:06 all I wanted initially was for something to try & make sure I wrote notes about every submission 18:59:18 I can work on it some more if you think it'll be useful 18:59:23 ajmitch makes sense 18:59:42 it'd be cool to take a look at it 19:00:16 I don't know, if we can get the features in the web interface quickly, I don't want to suck up your time 19:00:28 I guess we'll see 19:00:35 I'll probably still use something like it for my private notes :) 19:00:42 yup, makes sense 19:00:54 if you do keep working on it, share what's useful 19:00:57 some of them aren't really publishable once I got to the 3rd or 4th submission with .jar files & license violations 19:01:10 heh :) 19:01:29 curse silently, teach kindly :) 19:01:36 right :) 19:01:46 alright guys I am sorry that I have to leave in the middle as its way past midnight n m very sleepy now ... See you at the UDS next week :) 19:01:59 coolbhavi: thanks for stopping in, see you round :) 19:02:11 coolbhavi: we're pretty much ready to wrap up 19:02:19 OK, I guess we've got some points that can be discussed at UDS anyway 19:02:23 take a look at the blueprint and let me know if you want to add anything 19:02:40 one final thing, chair for next meeting 19:02:53 ajmitch: do you want to take the next rotation? 19:02:57 sure 19:03:05 cool 19:03:11 :) 19:03:15 I'll take the minutes and distribute them 19:03:20 thanks 19:03:21 and update the agenda for next time 19:03:31 thanks all! 19:03:31 thanks so much 19:03:45 see you at UDS 19:03:51 :) 19:03:58 see you there :) 19:03:59 #endmeeting Generated by MeetBot 0.1.5 (http://wiki.ubuntu.com/AlanBell/mootbot)