=== wedgwood_away is now known as wedgwood === wedgwood is now known as wedgwood_away === Sweetsha1k is now known as Sweetshark [08:46] hi all === marxjohn1on is now known as marxjohnson === udsbotu changed the topic of #ubuntu-uds-community-1 to: Track: Community | IRC team | Url: http://summit.ubuntu.com/uds-1303/meeting/21674/community-1303-irc-team/ [14:01] hi all [14:01] no hangout yet . . . [14:03] jono is on the case [14:05] hi [14:06] yes [14:06] hi [14:09] jono: can you add "Jussi “Tm T” Kekkonen" [14:09] anyone else got a google plus name to add? [14:09] jono: IdleOne / Giovanni Chiazzese [14:10] stupid google not allowing Tm_T as nickname [14:10] indeed [14:10] I'll just lurk and watch and annoy you via IRC [14:10] Myrtti ♥ [14:10] ok, so lets chat on IRC too and you can all watch me type [14:10] I see you're at the office AlanBell [14:10] nice business gear, button up shirt and tie and all [14:11] yeah, we keep it formal [14:11] nice [14:12] hi [14:12] hi [14:14] sehr gut!!! http://www.spiegel.de/wirtschaft/unternehmen/olg-duesseldorf-haelt-netzentgelte-befreiung-fuer-verfassungswidrig-a-887209.html [14:15] sorry wrong channel :( [14:16] * tsimpson_ pokes his head in [14:17] in case it hasn't been obvious from my moanings of late, I'll probably move over to Debian on my personal systems. Will stick around in some of the channels, but dropping some channels out and probably my Ubuntu membership as well. [14:17] Heya, tsimpson_. [14:17] and of course still a freenode staff member. [14:20] Good morning, sorry I am late [14:23] #ubuntu-de is good [14:26] sorry, have to go... [14:32] wouldn't hurt asking "what kind of communication you would need?" in time to time (: [14:33] jono: s/the mistake we have made/among the many mistakes we have made/ [14:33] o/ popey [14:33] o/ [14:33] popey, yup [14:34] i am unconviced "lack of info for the community" is the big issue at hand. [14:38] there is no officel wiki for ubuntu in germay we have ubuntuusers.de instead [14:39] * IdleOne checks [14:39] Yup, human. [14:40] apparently we have ~100 members https://launchpad.net/~ubuntu-ops [14:40] it's not because we particularly want to help, but it's easier to help with things going right on the first try than try to put off the fires when they've been done the wrong way and there's a great hurry to fix them, because X Y and Z start in 5 minutes [14:41] yeah I rather help when it causes less work and damage (: [14:42] and also I rather help than watch things go wrong just because [14:44] jono: you have to excuse us for thinking it's already been decided when a lot of things have seem to be told to people on a 5 day notice without much discussion. It kinda sets people's teeth on edge [14:44] but I'm sure this has been said by many people already, so w/e [14:45] there was at one time some mirroring of the bots [14:46] when the responsibility of running them moved from Seveas to others [14:47] and jussi's was just one of the mirrored ones, I had one mirrored one as well ready to be launched if something goes sour [14:47] and while we are talking about bots, we do have an issue with ubot2 [14:47] yup [14:47] the issue being that it's horribly out of date and not maintained [14:48] I'd like to have the bots decentralized, but that seems like something that requires much more planning and coordination that I can do on my own [14:49] tsimpson_: I think people would be happy with functioning backups for a start [14:49] and then there's the long awaited open-sourcing of the floodbots [14:51] * Unit193 more or less considers his a backup. [14:51] bringup later or? [14:51] yeah, the bans database is probably the main thing that isn't very redundant [14:51] someone (smart) needs to come up with a way to integrate LP and ChanServ (access lists) [14:51] QUESTION: where are the etherpad notes? :) [14:51] tsimpson_: I did get some way on that [14:51] Mirv: http://pad.ubuntu.com/uds-1303-community-1303-irc-team [14:51] hi Mirv I haven't been making any as such [14:52] outrage! [14:52] AlanBell: ok, I was just joining after other session to get some sort of summary [14:52] Mirv: about to start making some [14:52] tsimpson_: reading that instead now [14:52] Mirv: basicly "we want better communication and usage of our services" (: [14:52] Mirv: in summary: Canonical are evil. [14:52] tsimpson_: I mean the IRC log [14:52] nothing much to read atm [14:52] popey: awww [14:52] lol popey [14:52] popey, where is evilpopey today? [14:52] popey: NoTheyArent [14:52] (tm) [14:53] logs are on ubottu.com and (please use) irclogs.ubuntu.com [14:53] boo [14:53] o/ evilpopey [14:53] i must left [14:53] by === wedgwood_away is now known as wedgwood [14:54] * ahayzen runs to next session [14:55] hello again X) === udsbotu changed the topic of #ubuntu-uds-community-1 to: Track: Community | Quality in a rolling release world | Url: http://summit.ubuntu.com/uds-1303/meeting/21670/community-1303-quality-rolling/ [14:55] hello :-) [14:55] just speak up if you'd like to be in the hangout [14:55] balloons: o/ [14:56] * stgraber waves [14:56] * gema_ shouts [14:56] * cjohnston pops balloons [14:56] ohh look [14:56] balloons: can you send me the link to the hangout? [14:56] my missing notes [14:56] yes.. one sec to all those who pinged [14:56] lmap [14:56] lmao [14:57] balloons: You will eed to invite me to another email, is that fine [14:57] need* [14:57] Can whoever is in charge of the video feed let us know when it sactually tarts for this session, so I don't have to refresh every 30 seconds? [14:57] ElderDryas, sure thing [14:57] danke :) [14:58] balloons: can I have a spot in the hangout? [14:58] balloons: me too! [14:59] k I sent the url to all I believe [14:59] anyone else who wants to join the hangout? [15:00] starting the broadcast [15:00] anyone else who wants to join the hangout? [15:00] invite please [15:01] anyone else who wants to join the hangout? [15:01] I'd like to [15:01] Names and nicks? [15:02] anyone else who wants to join the hangout? [15:02] balloons, if theres room, i can === kentb-out is now known as kentb [15:04] restored [15:05] balloons: would that include things like "we have a PPA with a proposed new X.org stack"? [15:06] ah, sorry, I thought you were looking for it still. Feel free to delete [15:06] I can join if you have room [15:07] plars: your lower third [15:07] PPAs are going to screw those of us who use things like the BeagleBoard-xM in a desktop role [15:07] gema_: yes, one sec [15:07] skellat: I didn't mean PPA literally as we have them today; obviously we need some kind of "staging areas" which include ARM [15:08] I think we need a second britney instance to give us human testing before packages hit most end users in a rolling release [15:08] pitti_uds: Definitely. I look forward to seeing that coalesce. [15:09] ramping up backports would be great as well, if be able to automate some of the backports testing, that could certainly make approving backports easier [15:11] what is SRU? [15:11] stable release update [15:11] stable release updates [15:11] stable release update [15:11] thank u [15:11] https://wiki.ubuntu.com/StableReleaseUpdates [15:11] https://wiki.ubuntu.com/StableReleaseUpdates for the current ruleset [15:11] heh [15:11] jasnow: hehe :) it's the updates you get offered by e.g. update manager to an stable, released version of ubuntu [15:11] micahg: right, there's a work item on another spec for that [15:12] micahg: displaying backports that need testing in some public place would be good, I don'T think people see the bugtracker [15:12] (also drop the rdepend build deps requirement) [15:12] cjwatson: WI for second britney? [15:12] yes [15:12] any idea which spec? [15:12] cool [15:12] if we are going to take the resources we use for putting out a 6 month release to doing more SRUs and backporting more packages we should limit those to the last LTS and not to every previous release [15:12] the first rolling release session yesterday IIRC [15:12] jtaylor: Do you mean a dedicated page for backported packages? [15:12] bjf: yes [15:12] oh, didn't realise it hady any work items :) [15:13] cjohnston: the problem with that is that some people don't want their apps changing out from under them in an LTS [15:13] vibhav: I was maybe thinking software-center [15:13] right, new versions that introduce interface changes are exactly what some people run LTSes to specifically avoid [15:13] hangout plugin just crashed [15:13] plars, works here [15:13] jtaylor: ah [15:13] so I don't think we can just dump -backports into -updates [15:13] (say) [15:13] backports may be more difficult to support because they can sometimes cascade into updating a bigger dependency stack [15:14] isn't the point of moving to a rolling release to free up resources, devs are not working to support 3-4 versions of Ubuntu, and doing backports will defeat this purpose? [15:14] plars, if you mean the live stream [15:14] janof: it's a trade-off [15:14] the trade of is not having to deal with this when moving to a rolling release, its the main point IIRC [15:14] janof: we can't just drop stuff and leave users without a replacement that's acceptable - trying to figure that kind of thing out is exactly why we're having these discussions [15:14] janof: not really, its only going back one vs multiple [15:15] and I would hope much more targeted [15:15] cjohnston, so the plan is to support the existing, and the next future rolling release, no past versions, is that correct? [15:15] why we are mixing the rolling release with the LTS in terms of stability? [15:15] Am I understanding correctly: the LTS is separate from the rolling release? [15:15] ara, exactly [15:15] at the moment developers generally SRU to every affected stable release regardless of whether there's any pressing reason to believe that there's any users [15:16] janof: continue supporting old releases until the EOL. [15:16] I don't think the main archive LTS criteria should change since it needs to be stable, it'll be much easier to backport newer stuff from a rolling release to the LTS though as we currently have to backport through the regular releases [15:16] janof: there is no fixed plan as yet; there are proposals [15:16] Upgrade from CD was deprecated once the alternate discs went away, no? [15:16] security fixes too [15:16] cjohnston, this would mean that the benefits that we desperately need for the next version, 13.04, would not be feasable until EOL of older versions. [15:16] ara: the proposal to withdraw non-LTS stable releases means that some users will need an acceptable replacement [15:17] janof: how so? not everything would get backported [15:17] cjwatson, yes, I agree with that, but the LTS will stay stable [15:17] janof: let me be clear: we are not going to put the newest version of everything into the LTS [15:17] if anyone else who wants to join the hangout, let me know [15:17] but that is not to say that the current constraints are 100% fixed either [15:17] jtaylor: Instead of software-center, I think it should be more sensible for a universal resource which could be accessed from everywhere [15:18] upgrade from 12.04 to 13.04? [15:19] why would you upgrade from 12.04 to 13.04? [15:19] policy should still be 12.04->12.10->13.04 [15:19] because in order to support upgrades from 12.04 to 14.04, upgrades from 12.04 to 13.04 ought to work [15:19] it needs to work anyway [15:19] vibhav: that already applies to the bugtracker [15:19] it's a lot easier if you don't have to put all the upgrade support back in at the end [15:19] as the 12.04 -> 14.04 upgrade needs to work, we need all the postinst quirks etc. [15:19] ok, good point [15:19] brand new user - had a cd of 12.04, but thought 13.04 was the target [15:19] hi [15:20] cm-t_phone: If there is a LTS version for every seeded package, the idea sounds great [15:20] but they might skip putting a new HWE in .3 [15:22] cjwatson: I think we should test/fix 12.04 -> daily and and 12.10 -> daily every day [15:22] I was envisioning daily ISOs + automatic testing of both the "stable" rolling release and the "unstable" rolling release [15:22] pitti_uds: agreed [15:23] well, both yesterday->today and last monthly->today [15:23] vibhav: that's massive overkill - take for example man-db, really doesn't need a separate LTS version [15:23] the least amount of release versions surely is desirable. utilization of resources. [15:23] cjwatson: Well yes, I was wrong here [15:24] And I would say that's the common case [15:24] right, because one day, today's snapshot will *be* the monthly :) [15:25] so two versions of the RR, daily and monthly ? [15:26] greetings from Miami [15:26] cjwatson: Excluding packages which dont need to have LTS versions [15:26] selena2013: greetings from England [15:26] vibhav: yes, but to be honnest i try to see that from the point of a big company, because they can not spend on migrating all their custom tools and apps on monthly(but we are more in the LTS talk than SRU) [15:27] vibhav: Which is most packages [15:27] Indeed [15:27] cjwatson: you've destroyed my world :( [15:27] I think that with adding the second britney migration would allow people to receive updates in a more delayed fashion which might alleviate the need for monthlies since there will be increased stability and quality in the rolling release [15:27] From this it really sounds like monthly roll-ups would be a nightmare [15:27] is LTS still a term we will use with a RR? [15:27] https://lists.ubuntu.com/archives/ubuntu-devel/2013-March/036798.html .... [15:27] janof: it's two different things [15:28] janof: future LTSes will still be what they are used to be, nobody was debating those [15:28] right, but im saying if we move to RR, the LTS term will be sunsetted and only reflect past releases, is that correct? [15:29] janof: ah, naming hasn't been discussedyet [15:29] That's not true [15:29] None of the rolling-release proposals alter the support terms of LTS releases, past or (AFAIK) future [15:29] cjwatson: I think he means calling them "LTS" doesn't make too much sense any more, as we woudn't have other kinds of releases [15:30] right, it defeats the entire purpose [15:30] *shrug* Might also be too confusing to change [15:30] Least of our problems, IMO :) [15:31] fair enough [15:31] daily builds = new beta / alpha versions [15:32] gema_: we have several scenarios right now, like "default install", "all of main", and the huge "everything that has a .desktop file" [15:32] so a regular user will choose between LTS and rolling release in the future [15:32] gema_: these give quite some good coverage [15:32] stgraber: IMO, SRU's for new versions of packages would be helpful if some interface in the old version was depreceated [15:32] janof: for rolling it would have to be at *least* beta quality [15:32] selena2013, thats sorta my point, thats the same problem we have now [15:32] that choice hurts ubuntu [15:32] selena2013: yep I think that's the idea, rolling release and LTS supported for years [15:32] stgraber: i.e, only update those packages which have been depreceated [15:33] stgraber: we also check for orphan or changed conffiles === kermit66_ is now known as kermit666_ [15:33] the point of moving to an RR, im not not mistaken, is to provide updates faster, and to save resources by not having to support LTS and other versions for years [15:33] plars: do you know if we run the post-install ISO tests on the auto-upgrade tests as well? it seems we should [15:33] LTS, the term and the physical being, wouldnt make sense with a RR [15:33] autopilot is already run before packages are relased... [15:34] thomi: but there would be value to run it e.g. when X is released (and there is no change to unity) [15:34] pitti_uds: iirc it's a slightly different set of tests [15:34] and the tests specify the unity config to use, so I'm not sure AP is going to help you much here [15:34] thomi: but upgrades might change some of its assumptions, so running again after e. g. an LTS->LTS upgrade makes sense [15:34] but I'm not sure if LXC /X is going to be an answer [15:34] i dont have a problem with that , i would recommend newcomers LTS . But if you want the latest and greatest you can choose RR [15:34] pitti_uds: most of the upgrade ones are around making sure that we got the kernel we expected after the upgrade, etc [15:34] we had troubles to run the tests in a VM (but maybe things changed) [15:34] selena2013: yep that's the idea, RR instead of the six month releases [15:34] plars: right, I see that we need some addiitonal ones; but we could also run the post-install iso tests? [15:35] mmrazik: sure, but you get that if you run AP as part of the X release tests. [15:35] pitti_uds: I don't think there's anything that would keep us from expanding those [15:35] question: how are mesa upgrades currently tested? I remember too many cases where a new mesa version broke some rendering in a driver and caused issues in KDE. Can that be somehow catched at all? [15:35] thomi: right [15:35] mmrazik: I just meant that using autopilot to test the *upgrade* of unity (as it stands today) won't get you much extra [15:36] balloons: wouldn't the automated daily images still exist? [15:36] thomi: oh. agree with that. [15:36] to all extends [15:36] the monthly would be the current daily [15:36] dobey, yes, I believe so [15:37] More or less [15:37] the dialsy would become the monthly [15:37] The last proposal I saw did require a configuration tweak to dailies just before each monthly [15:37] Which is cumbersome but hard to avoid [15:37] cjwatson: sorry I muted your quick typing! [15:37] hehe [15:37] Yeah, sorry, I could have sworn I'd muted [15:37] sounded like drum fire [15:38] right. to avoid updating until next month [15:38] why am i hearing german ads now ? [15:38] This is one of the many difficult issues with monthlies [15:38] I definitely wouldn't assume monthlies will happen, there's still lots of debate [15:39] right, i don't want to derail the discussion; but i'd just not do monthlies. maybe have a monthly milestone but not a special image/config for it [15:39] Remember to help the devs to reproduce an issue when they get a bug by saving images/etc [15:39] Saving the image URL is generally enough [15:39] We can extract package versions from that [15:40] being able to match package versions to RC bugs seems important [15:40] dobey: There are many good things about having some kind of monthly rhythm, e.g. dev milestones, marketing, etc.; and I can see the point of images; I just don't buy the upgrade thing [15:40] But yeah, another discussion [15:40] cjwatson: right [15:40] cjwatson: which is why i say milestones without images is something i could get behind :) [15:41] RRs should be rock solid, IMO [15:41] well it depends on what the promise is [15:41] I would expect quality for the end users to be about equal to the regular releases we produce now [15:41] thanks to all developers for the hard work . [15:41] if you promise RRs will be stable and usable, that's what I expect [15:41] s/about/at least/ [15:41] if you tell me they're to be treated as the current, daily development images where things *may* break, then I'd expect that [15:41] draw a picture [15:43] roadmr, ++ [15:43] do-release-upgrade every 2 years instead of every 6 months, basically [15:43] janof: ^^ [15:44] We haven't drawn a picture yet because there are still about a dozen different opinions :-) [15:44] how are we going to provide hardware enablement kernels back to LTS in this new scheme? Similar to the quantal-lts-backports kernel.. [15:44] chiluk_: There was a session end of yesterday about that [15:44] nuts... I was on a plane. [15:44] most server users are probably on the LTS now [15:44] most ... yeah [15:44] chiluk_, there should be a video of that session [15:44] server is a different world than desktop [15:44] yes a lot of them are.. [15:44] servers not running LTS already are either personal home server users, or crazy :) [15:45] I can't imagine many production servers who'd want to be rolling [15:45] but a lot of them are running quantal backports kernels as well. [15:45] chiluk_: http://summit.ubuntu.com/uds-1303/meeting/21598/foundations-1303-hwe-stack/ === udsbotu changed the topic of #ubuntu-uds-community-1 to: Currently no events are active in this room - http://summit.ubuntu.com/uds-1303/community-1/ - http://ubottu.com/uds-logs/%23ubuntu-uds-community-1.log [15:45] those not on the LTS should be able to handle any bumps that regular end users would be able to handle [15:45] dobey: even home servers use lts :) [15:45] for business desktops will be the same [15:45] lts to lts [15:45] barry: i do yes. but some people apparently don't [15:45] thanks [15:45] thanks! [15:45] The last 4 minutes sound like: As a nobody end user, daily (1 desktop, 1 laptop) I have two choices. Scary - developer build take my chances (I'm willing but nervous) or Sad - I only get new stuff every two years. [15:45] o/ [15:45] o/ [15:45] whats the name of the next lts going to be? raring? [15:45] dshimer: that's how it is now [15:45] dshimer, my expectation is rolling = non-lts stable [15:46] jtaylor: We'll see [15:46] not developer edition scary [15:46] thanks everyone [15:46] cjwatson: it must be raring or "larger" for daily ppas which use ~raring suffix [15:46] jtaylor: We clearly can't decrement [15:46] balloons: scary salmon? [15:46] jtaylor: it probably won't be 'rickroll' [15:46] lucky raring is smaller than most other "r" words :) [15:46] jtaylor: But that suffix is horribly misguided for obvious reasons [15:46] gema_, indeed! [15:47] I am glad I don't have to name them x) [15:47] dshimer: a bit like contract phones to an extent, 24 months contract is quite standard, then you may buy a new phone after [15:47] jtaylor: It needs to be changed to ~ubuntuYY.MM before we reach U [15:47] cjwatson: yes but thats a launchpad issue so far I know [15:47] jtaylor: Yes [15:47] Anyway, LP series have a fixed name; renaming would be madness [15:48] i don't think it really matters if it's ~rickroll or ~ubuntuYY.MM [15:48] Hopefully nobody will require us to do that :) [15:48] well, until after z [15:48] dobey: It matters once we ... yes [15:48] Which is why I've been trying to fix this for a while before it becomes a crisis :) [15:48] :) [15:48] We fixed backports a year or two ago === udsbotu changed the topic of #ubuntu-uds-community-1 to: Track: Client | Ubuntu Friendly usability and feedback improvements | Url: http://summit.ubuntu.com/uds-1303/meeting/21628/client-13033-ubuntu-friendly-improvements/ [15:50] i imagine it's not really that hard to fix the recipe builder code though. [15:50] anyway, i guess we need to stop discussing that here now :) [15:50] Shouldn't be, just a matter of finding people to maintain and deploy it given lack of LP manpower at the moment [15:50] right [15:50] There's been an RT open for some kind of upgrade there for months [15:51] But as you say [15:51] ciao :) [15:51] jtaylor: I did file a bug about launchpad not using ~ubuntu..., it's low though [15:52] cprofitt: hello [15:53] cprofitt, zyga-uds,roadmr hangout is up [15:54] balloons: thanks === hikiko_ is now known as hikiko [15:56] I'll start it in a few [15:58] who is the track lead in charge for this one? [15:58] can I get a hangout link, please? [15:58] cprofitt: are you around? [16:00] ara, sent [16:00] anyone else who wants in the hangout, speak up :-) [16:01] starting in a moment [16:03] we should be live now :-) [16:04] ara: hi! could I get an invite to the hangout? [16:06] Is this the public web site for Friendly? https://friendly.ubuntu.com/ [16:06] jasnow, that's correct [16:07] thanks [16:10] cprofitt: \o/ [16:10] like "UPS Tracking" link ;o) [16:11] ara, try turning off your video to lower your bandwidth needs [16:14] Can you report the submit status without the user's identify? [16:15] invite please [16:15] jasnow: status as in, it was accepted/processed, or rejected due to failing tests? is this what you mean? [16:16] anyone else who wants in the hangout, speak up :-) [16:16] bobthebobber, inviting [16:16] yes [16:23] one thing to consider is the difference between "works without any tweaks" and "works if you apply some patches, additional drivers, confi changes" [16:23] *config [16:24] Ubuntu Certified handles this by saying "Certified (with Notes)" [16:25] rrnwexec: yep [16:25] we can do that as well on Friendly [16:26] QUESTION: what about sending incomplete results? if I only run 3 tests (out of 10), shouldn't i be able to submit what i have done? [16:27] rrnwexec: I ping them on your question in a second [16:29] deliberately crashing checkbox! [16:35] there is still a theme out there that says "ubuntu works on everything." Improving SEO of Ubuntu Friendly would help, but we also need a strong(er) awareness campaign. [16:37] rrnwexec: you can help us with the SEO [16:37] :) [16:38] cprofitt: yes, i can blog about it, post to lists, etc. that may help. [16:38] roadmr, doesn't checkbox already support it? [16:38] ara, roadmr I believe I've resumed tests before [16:38] "black hole problem" [16:38] QUESTION: Is Ubuntu Friendly submission an Ubuntu Accomplishment "trophy"? [16:39] ara: resuming tests? yes, but the feature is geared towards resuming after a crash [16:39] ara: cprofitt was talking about "oh crap, I need to go now, I want to pause the test and resume later" - we don't accomodate that use case [16:40] ara: the only option as he mentioned is deliberately crashing checkbox %) [16:41] jedimike_, searching in Google for ubuntu Acer Aspire 3610 brings the staging site first :-\ [16:41] \o/ [16:42] ara: oh dear... well at least we know the pages are SEO ready... [16:42] we need to both get more results on the UF site ... and increase SEO [16:42] jedimike_, maybe it was only deployed to staging? (seo capabilities) [16:42] greetings to zyga ;) [16:43] work with LoCo Advocates [16:44] rrnwexec: for your question, so it doesn't scroll up and disappear (re: accomplishments). Answer is "No" :( [16:45] check before buying on Craigslist [16:46] first things first should be to fix the submission process I'd think [16:48] when the site is ready, and the submission process is better, we could spin up a Global "Test Your Hardware" Day. [16:48] similar to a Jam, but with less jam ;) [16:48] why dont we incorporate ubuntu friendly into the installer? [16:48] rrnwexec, +1 [16:48] +1 bobthebobber [16:49] ubuntu friendly, now friendly +++ [16:49] bobthebobber: poking the hardware in ways that can make it fails or crash at install time... fun [16:50] roadmr: ah yeah, good point, but maybe after the install, and before first boot? [16:51]