17:01:45 #startmeeting 17:01:45 Meeting started Tue Oct 23 17:01:45 2012 UTC. The chair is zack. Information about MeetBot at http://wiki.debian.org/MeetBot. 17:01:45 Useful Commands: #action #agreed #help #info #idea #link #topic. 17:01:51 so, who's around? 17:01:56 * algernon is around 17:01:59 Diziet is a bit late but with us 17:02:04 lucas, algernon, zack here 17:02:20 h01ger: is sick (let's hug h01ger!) 17:02:32 not a good idea, if he is sick 17:02:33 Maulkin, moray: ? 17:02:36 yes, here 17:02:38 lucas: :) 17:03:10 no Maulkin atm, please ping us if/when you show up 17:03:32 at the URL in topic I've updated the agenda, with a brief report of relevant DPL stuff for the past 2 weeks 17:03:44 #topic agenda / 2 weeks status report 17:03:56 Hi. 17:03:59 I'm here now. 17:04:04 great! 17:04:16 going briefly trough it: 17:04:24 I need to apologise though because I have done precisely none of the things I said I would. 17:04:34 - outstanding delegation done (except -ctte, where I'm waiting for ctte input) 17:04:48 - relevant new topic is the big images for public clouds discussion on -private 17:05:16 which is on -private for no good reason, will be mentioned publicly soon, and we seem to have agreed to move it to a public place (so I don't worry too much about mentioning it here) 17:05:28 that has induced some political dealing with Amazon/M$ 17:05:41 and there's a next action of asking listmasters to create debian-cloud@lists.d.o 17:05:53 if anyone wants to pick it up, I'll be happy to suggest a description 17:06:05 #info volunteer needed to request creation of debian-cloud@lists.d.o and see how it goes 17:06:20 anyone interested in public cloud topics? 17:06:37 well, I'd be subscribed to the list once created... 17:06:47 I've also proposed a common political position on public clouds to FSF, but haven't heard back yet 17:06:54 o/ 17:07:05 Apologies for the delay) 17:07:10 Maulkin: hi, we'll need you for the important/urgent topic ;-) 17:07:16 zack: sounds a good idea, anyway 17:07:25 moray: thanks! 17:07:37 moving on, I've called for topics to raise at the next UDS, which I'll be attending 17:07:44 there's a link to my mail in titanpad 17:07:59 if you've topics, I'll be glad to add them to my UDS agenda, as we don't have that many... 17:08:13 (I can share the ones received thus far, if you're interested) 17:08:52 lucas: you've been an active Debian participant at UDS in the past, anything come to mind? 17:08:53 I'd be interested in hearing those, to not repeat them if already mentioned :) 17:08:58 I'm interested, especially if there is anything new in the list, rather than the same old topics :) 17:09:16 but I can probably only suggest the same old ones myself 17:09:26 - suggested/old topic: having Debian suites in PPA 17:09:39 I've already asked that in the past, and got a "no" answer from the company for "lack of resources"(!) 17:09:39 zack: not really, I'm a bit disconnected from ubuntu issues recently 17:09:52 I'll take care of asking, again, publicly this time 17:10:02 to avoid this gets asked again and again in the future 17:10:46 - another one is by pabs, and is well known too: avoid forking packages (think of: linux, mozilla, etc.) 17:10:54 (of course that's a "no" too) 17:11:09 s/of course/very likely/ 17:11:34 nothing else up to now 17:11:41 on topic of forking: 17:12:59 it might lessen the need to do that, if Debian maintainers were told in advance that ubuntu wants to do X. So instead of them filing bugs to "please do this, here's the patch", a day or two before, they could file a "we want to do this, can you help?" 17:13:06 (in parallel the only other point from recent activities is www.d.o relicensing, nothing to report beside progress and the fact I've invoked bkuhn's help, as he offered in the past 17:13:11 see https://lists.debian.org/debian-www/2012/10/msg00099.html ) 17:13:23 so, notify first, fork later vs fork first, patch later 17:13:26 zack: asking them straight off to avoid forking is probably a "no", but I guess you can try asking "what can we do to make things easier for you and reduce the number of cases you need to fork?" 17:13:50 algernon, moray: you seem to go in the same direction, but I guess it's mostly a matter of timing for them 17:13:51 moray: A good way to put it. It's not like they want to fork. 17:13:54 It's hard work... 17:13:59 which might include things like what algernon is saying, but it's more likely to be positive if it comes from their side 17:14:10 the "variance" in Debian maintainers response time is probably huge 17:14:14 but I agree 17:14:21 that's a nice way of putting it, and I can ask it that way 17:14:42 #action topic to raise at UDS: ask what Debian could do to avoid Ubuntu's need to fork packages 17:14:50 zack: have a policy of notifying debian maintainers where UDS proposals affect their packages would be good though 17:15:02 s/where/when 17:15:15 lucas: ah, that's a nice technical idea indeed! 17:15:41 lucas: that raises the problem of whether it should be opt-in/opt-out, though 17:15:52 I fear there will still be people complaining if the notification is opt-out... 17:15:55 they send patches to the BTS anyway. 17:16:06 most proposals would make valid bug reports, even without patches ready 17:16:09 I was thinking about manual emails, not something automated 17:16:11 algernon: yes, but that's not quite the same thing, is it? 17:16:12 notification can come in form of a wishlist bug 17:16:23 ok, manual sounds good 17:16:42 UDS proposals are usually more large-scale changes 17:16:57 (but I haven't been to UDS recently, so I'm not 100% sure it's still relevant) 17:17:07 lucas: can you please write a suitable #action line about that UDS topic for the logs? 17:17:34 ok 17:17:56 (if anything else shows up, feel free to follow-up on -derivatives of to me) 17:18:09 #topic important/urgent topics 17:18:37 I've briefly chatted with the release team and Maulkin this week 17:18:44 and I'm getting a bit worried about release communication 17:18:55 #action zack (at UDS) to ask Ubuntu contributors to send a mail to inform relevant Debian maintainers when a UDS proposal affect their packages, so they know about it/ can provide feedback before implementation starts 17:19:09 I've heard IRL various developers wondering "what's the status?" and I feel we're losing a bit of developer's attention 17:19:29 it looks crucial to me to avoid that at this point 17:19:40 so I wonder if we could help the release team in any way, in communicating with developers at large 17:19:45 * Maulkin nods 17:19:50 yeah, I'm personally unable to answer "is it going well?" 17:20:01 I've in the past helped drafting RT bits, but I don't want to step in RT shoes 17:20:08 Maulkin: what's your take on this? 17:20:13 do you need help on that front? 17:20:51 (of course I'll be happy to discuss this with RT at large, but given Maulkin's here, I see no reason not to ask) 17:21:05 Basically, at the moment there's far too many bugs, and the RT members are quite snowed under with the amount of unblock requests, which means when I ask for things to put in bits, it's a bit hard to get updates. 17:21:47 Something that may help is having people do things like fettle UDD or similar to get some nice stats about when things are fixed at BSPs 17:22:02 like Tolimar's stats, yes, I'm missing them too 17:22:07 Yeah. That. 17:22:24 I certainly don't notice much excitement/urgency from developers, more that people are just assuming the RT will get there 17:22:56 to be honest, I think that having a bits RT mail, even if it's rather empty, saying "we're not MIA" would be useful :) 17:23:04 and I've positive topics to suggest for it, like 17:23:09 - lots of BSPs are happening -> good 17:23:22 - we'll remove leaf packages every 1/2 weeks, see -devel \o/ 17:23:25 Maulkin: would unblock reviews done by DDs outside of RT help the RT? 17:23:31 If appropriate something like "we are snowed under with unblock requests. this is good. keep them coming" 17:23:39 or even: 17:23:39 - we'll hear from you every 2 weeks from now on :-) 17:23:46 Diziet: yeah, that too 17:24:01 doing a UDD cgi that generates a Tolimar-like email is quite easy 17:24:17 lucas: That would be useful, as I have little idea on how UDD does its magic. 17:24:22 lucas: or, better, a planet blog post 17:24:31 which I think is more useful when trying to generate "hype" 17:24:42 algernon: I think it may do. 17:24:50 lucas: would you be interested in doing that, or it's asking too much? :) 17:25:13 (yeah, I know, I'm a PITA) 17:25:16 One big thing could be to identify possibilities for removals too, that's kinda automated, but would help being ramped up IMO. 17:25:17 Maulkin: then maybe that could be emphasized in the bits mail too 17:25:18 zack: well a planet post is one copy/paste away once you have the email ;) 17:25:22 Re algernon's suggestion, this is perhaps a good way to help the RT scale. What would be a good way to start it off ? 17:25:56 zack: I'm a bit busy. I'll think about it. 17:26:03 lucas: fair enough 17:26:23 Diziet: a RC bug triage would help, so that the usertags can be applied. 17:26:45 #info suggestion for release process: UDD-based script to generate Tolimar-like bugs stats, and give evidence of BSP results and the like 17:26:58 (Though doing the release usertags for 500 RC bugs is quite a challenge...) 17:27:08 * algernon thinks a large "How can YOU help get wheezy out the door?" post/mail would be awesome, esp. when hardly any of it would actually be about *fixing* bugs. 17:27:09 Maulkin: OK, great, so I guess perhaps you should mention that in your bits mail with a link to the list of RT usertags ? 17:27:14 #info suggestion for the release process: mechanism to encourage "peer review" of unblock requests, RC bug triage, etc 17:27:48 Maulkin: so, regarding kicking out the "bits from RT" mail, I think that's *really* urgent now, like yesterday 17:27:56 how do you think we can help? 17:28:01 (few people are willing/able to fix RC bugs in unknown packages. triaging, helping with reviews, etc are much much easier to handle, imo) 17:28:14 * Maulkin had an idea that we shouldn't release with any O: packages, but it wasn't the most popular suggestion... 17:28:36 I'm available to both propose an outline, or to do the "think positive" rewording once a first draft is ready, or ... you name it 17:28:51 Maulkin: it's probably something to discuss at the start of a release process, not at the end, no? 17:28:54 zack: Chuck in an outline in titanpad? 17:29:04 Maulkin: so they can be removed early from testing? 17:29:05 Maulkin: ok, that I could do, thank mail RT? 17:29:06 lucas: I intend to do so after we've released :) 17:29:31 #action zack to draft a RT mail outline and pass it on to RT 17:29:56 Maulkin: is ntikyer's work on removal candidates automated at the moment or not? 17:30:05 zack: Semi-automated. 17:30:15 ok 17:30:41 have you decided on a cadence for it or it's "when niels has time" :) ? 17:30:46 (which would be reasonable, of course) 17:31:10 zack: the latter I think :) 17:31:17 argh! :) , but fair enough 17:31:39 If anyone wants to do patches for it to improve it, then I'm sure Niels would like it. 17:31:52 Maulkin: where's the code? 17:32:18 zack: mentioned in the removal mails I think, checking. 17:33:07 ah, yes, found it 17:33:28 #info patches welcome to improve niels workflow for identifying removals svn://svn.debian.org/svn/collab-qa/rc-buggy-leaf-packages 17:33:29 Just as mutt finally loads =debian/devel. 17:33:42 I mention it here not necessarily because some of us might want to do it 17:33:59 but because if we find ways to *motivate* others to do it, it'd be great 17:34:18 any other ideas on how to help the release team? 17:35:22 ok, next topic 17:35:27 #topic review action items from last meeting 17:35:43 That should probably do for now 17:35:44 I suggest that every owner goes through past items, reconfirming all still pending and/or commenting the others 17:35:53 Diziet: firsts are yours... 17:36:08 They are all still pending. I still intend to do them. 17:36:29 would you like to re tag them as #action for this meeting logs? or shall we leave them only in the old logs? 17:36:32 I apologise for not doing them yet. My lame excuse is our somewhat-overrunning building project at home, but it's very lame. 17:36:38 no worries! 17:36:51 I shall re-action them. 17:36:54 for mine, I'd prefer to have them in the most recents logs, but YMMV 17:37:01 #action Diziet start conversation on -project about dpl statement process 17:37:20 #action Diziet draft more formal statement re trademarks and discuss on -project, apropos of https://lists.debian.org/debian-project/2012/02/msg00073.html 17:37:35 (Re-actioning sounds a good policy to discourage the action lists from growing indefinitely.) 17:37:42 moray: Yes. 17:37:47 OK... 17:38:05 algernon: any news on the mentors.d.o/DMCA front? 17:38:14 (if not, please re- #action it) 17:38:29 I looked through it, and have a few questions / observations 17:38:48 should that be discussed here, or re-#action and do it in mail? 17:39:03 we've still some time, if you've short questions go ahead 17:39:08 otherwise re-action and mail please 17:39:29 lets go with re-action then. I'm not sure I can make the questions short enough. 17:39:33 k 17:39:44 #action zack to ask SFLC / other project re making DMCA policy draft public 17:39:53 #action algernon to review draft DMCA policy and adapt for the mentors.d.o use case 17:39:56 I've contacted SFLC for a couple of things recently, but forgot about that 17:40:16 in the meantime, I've got a "busy, but we'll get back to you" answer about the proper package description for libdvdpkgcss 17:40:30 that was in fact an #action of mine from last meeting, that didn't make the minutes 17:40:39 next one of mine is still pending to: 17:40:43 #action zack to write an outline of the -companies announcement and mail it to press@d.o 17:40:54 lucas, on the other hand, has progressed on the salvaging thread! 17:41:01 salvaging thread: revived the discussion a bit, waited some time, posted an updated proposal this morning. Probable outcome: consensus, with minor disagreements. NEXT: address new comments; then dev-ref patch. 17:41:16 yes, looks great 17:41:29 lucas: I guess the patch should also fix similar bug report by Bart Martens, right? 17:41:46 I haven't looked at this 17:41:56 lucas: he mentioned it in the thread, let me find the bug log 17:42:15 Michael Gilbert mentioned something, but... 17:42:28 ah, you're right http://bugs.debian.org/681833 17:42:35 I feel bad that noone answered to him 17:43:08 if it's related, it'd be nice to let him know "we're going to do it, but a bit differently, thanks for the suggestion" 17:43:53 noted 17:44:21 #action lucas to finalize salvaging thread and prepare devref patch 17:44:22 #action lucas wrap-up the salvaging/orphaning thread and submit dev-ref patch. also address #681833. 17:44:28 race condition!, sorry 17:44:40 next one is mine 17:44:43 and it's DONE 17:44:44 np, now I can be twice as efficient :-) 17:44:47 :-) 17:44:55 #action Diziet to bug zack re next meeting agenda 17:44:58 #info restricted logo has been renamed as such and is now live on the website 17:45:03 Diziet: indeed :) 17:45:05 When is the next meeting anyway ? 17:45:15 #topic next meeting 17:45:21 does 2 weeks from now work for everyone? 17:45:30 We have a timezone change. 17:45:44 erm, probably, yes 17:45:57 30th works for me. 17:46:04 can someone do the needed date -d trick? 17:46:09 algernon: that's next week 17:46:11 I suck at that :-/ 17:46:24 oh, right. two != one. 17:46:30 date -d @1352223966 17:46:34 ^ is what's proposed 17:46:35 #action zack to write an outline of the -companies announcement and mail it to press@d.o 17:46:38 (I forgot that one) 17:46:49 That would be an hour earlier for anyone in a European timezone with DST. 17:47:08 but do we want to switch an hour later, for the same apparent time for many of us? 17:47:15 I will be in Barcelona then. I have no idea if I will be able to make the meeting. 17:47:23 moray: that would be my preference, yes 17:47:38 Diziet: Do you mean: date -d @1352221200 17:47:41 +1 17:47:50 Maulkin: Probably, sorry. 17:47:59 Diziet: you won't be able to make it anyhow, or does it depend on the +/-1 hour we choose? 17:48:08 zack: I have no idea; disregard me :-) 17:48:32 ok :) 17:48:33 If no one is *against* one hour later, it's probably better to go that way. 17:48:36 for info, that's fine by me 17:48:41 (either is) 17:48:43 OK, so that would be date -d @1352224800 17:49:14 so, I propose same "apparent time" for most of us, meaning one hour earlier for Diziet 17:49:53 #info Next meeting Tue Nov 6 18:00:00 GMT 2012 date -d @1352224800 17:50:06 Diziet: yes, correct, thanks for clearing up the mess :) 17:50:13 #topic miscellanea 17:50:17 we've 10 minutes left 17:50:34 any specific topic you want to address? (I've one, but it's low priority) 17:51:14 so, here it is: 17:51:28 moray: did you give some further thought at the local group topic? 17:52:18 zack: not towards any concrete result. I was thinking about your point of combining this with local contact information etc. 17:52:47 moray: ok, no big deal 17:53:19 oh, no, wait, here's another important one! 17:53:40 zack: for the groups themselves I was still wondering if making some "template for a group" page then encouraging them at (roughly) city level would be the most likely to work 17:54:08 (and grandfathering in existing groups at that level, but really I don't think there are that many active ones) 17:54:19 that is probably what would work best to actually have local IRL activities, a-la meetup 17:54:51 yes -- which seems the most likely way for them to be useful to users or potential users 17:54:52 but then we will still need some national contact point (at least where language barriers matter, such as European countries) 17:55:19 moray: agreed, we do have two different use cases it seems 17:56:01 I agree contact points at higher levels are good, but I'm just wary of making it too easy for people to declare themselves the leader for country X rather than some kind of communications points 17:56:20 that's precisely the "governance" part I was weary about 17:56:37 I was pondering some DPL blessing in the beginning, and later on some "council" (as long as it has turnover) 17:56:51 anyway, before we run out of time, an important one 17:57:00 as per last meeting agreements 17:57:04 zack: has that discussion happened on a list already? 17:57:11 I'll mention the meetings in the next 'bits from the DPL' mail 17:57:18 if you think it's soon, shout 17:57:41 I personally think it'd be fine, in the hope that we'll have a more decent agenda, thanks to Diziet poking me :) 17:57:46 lucas: not that I know of, no 17:57:55 lucas: before going there, I think we need a more solid proposal 17:58:00 it seems fine to me, to do that before anyone starts the normal silly accusations of cabals 17:58:05 * algernon thinks it's fine, too. 17:58:18 what about mentioning there's now an active group, and that interested parties could still join? 17:58:59 ok, good, that was my default too 17:59:09 I kind-of fear that we will just get lots of lurkers 17:59:24 * algernon has to rush, will send the mentors.d.o/dmca mail as soon as he's back (couple of hours, tops) 17:59:33 that's quite probable, but it's not a problem 17:59:36 lucas: I suspect that too, but I don't see it as too harmful 17:59:39 algernon: bye, thanks for your time! 18:00:00 I dunno if you've followed tech-ctte meetings: it's quite the same, and sometime it's useful too 18:00:13 any other topics before closing the meeting? 18:00:36 ok, we can try, and switch to #debian-dpl-cabal if needed :P 18:00:42 bad lucas! 18:00:57 I hereby declare this meeting ... finished! 18:00:59 #endmeeting