« Previous day | Index | Today | Next day » Channels | Search | Join
All times shown according to UTC.
Time | Nick | Message |
---|---|---|
12:13 | tomeu | hmm, should have suggested to the triagers to identify bugs we should really fix for 0.84.1 |
12:14 | erikos | tomeu: i can ask this as homework |
12:14 | FGrose: i will post a link on the blog | |
12:14 | FGrose | erikos: Great |
12:15 | erikos | FGrose: we really need to find out who the Kids testers are ;p |
12:16 | ok - todays subect is 0.86 | |
12:16 | FGrose | erikos: Email them through the wiki talk page, special link in sidebar |
12:16 | erikos | FGrose: ok |
12:16 | we need to define the feedback process | |
12:16 | walterbender | erikos: 0.84 is great, in case you didn't realize |
12:17 | erikos | and rough out the schedule |
12:17 | walterbender: what do you mean? ;p | |
12:17 | walterbender: i look unhappy? | |
12:18 | walterbender | erikos: you should be very proud of your child :) |
12:18 | ishk | can anyone tell me how can i get the information about sugar lab ideas on gsof 2009,the sugar site direct me to this IRC |
12:18 | :( | |
12:19 | erikos | ishk: :) |
12:19 | ishk | :( :( :( |
12:19 | tomeu | one minute of proud silence for 0.84! |
12:19 | erikos | homunq: can you guide our friend? |
12:19 | cjb | maybe there should be a #sugar-gsoc or something? |
12:20 | ishk | :D i think i came to a wrong place,,,am i? |
12:20 | erikos | there is always one stepping into the silence ;D |
12:20 | tomeu | cjb: yeah, I'm thinking that as well |
12:20 | erikos | ishk: #sugar is better |
12:20 | tomeu | ishk: here is for scheduled meetings |
12:21 | erikos | walterbender: ok - i will celebrate a bit - actually we should all do |
12:21 | tomeu: what was about the party? | |
12:21 | sdziallas | a party? when? where? who? :) |
12:22 | walterbender | sounds like linuxtag is the party date? |
12:22 | sdziallas | :):):) |
12:22 | tomeu | ok, deal |
12:23 | erikos | walterbender: ok, so you are coming as well - i read between the lines ;p |
12:23 | walterbender | I'll try. |
12:23 | tomeu | there's also wien, bolzano and paris |
12:23 | and of course, the center of europe and thus the optimal place, prague | |
12:24 | sdziallas | heh. |
12:24 | finds a small hint from tomeu here | |
12:24 | erikos | well it is good to celebrate several times |
12:24 | tomeu | german guys should just hop on the train and come spend a weekend here |
12:25 | I already celebrated it myself a couple of weeks ago | |
12:25 | erikos | tomeu: yeah, i should come visit soon |
12:26 | tomeu | if mel was here, she wouldn't be happy with our meeting discipline |
12:26 | sdziallas | we're having a meeting? heh. ;) |
12:27 | erikos | oups |
12:27 | walterbender | sdziallas: everyibe followed ishk to #sugar |
12:27 | erikos | ok - everyone makes a funny face now |
12:27 | *blink* | |
12:27 | walterbender | #TOPIC 0.86 |
12:27 | erikos | now concentration |
12:29 | walterbender | I am in part responsible, because I asked erikos for a rough idea of the schedule |
12:29 | erikos | walterbender: ok - schedule is a good start |
12:29 | we already said that we should try to follow the Fedora one | |
12:30 | to get our release into it | |
12:30 | as Soas is an important product | |
12:30 | walterbender | which means we can back out from those dates? when is F12? |
12:30 | erikos | yup - let me look it up |
12:31 | tomeu | maybe we should reflect a bit about how well has worked the 0.84 schedule |
12:31 | walterbender | tomeu: good idea |
12:32 | tomeu | IMO, the worst problem was that people didn't had a good way to test sugar |
12:32 | we are still trying to get soas-* to work | |
12:32 | but hopefully that will have improved for 0.86 | |
12:32 | walterbender | tomeu: made worse by the fact that it is difficult to test 0.84 on the XO |
12:32 | sdziallas | walterbender: this will become easier, soonish |
12:32 | silbe | there were some last-minute breakages that frightened me - the keyboard problem e.g. (sorry erikos) |
12:32 | walterbender | tomeu: SoaS will greatly broaden our ability to test |
12:32 | tomeu | maybe we should make a goal to run 0.85 on top of F11-based soas |
12:33 | erikos | so F12 - looks like it is meant to be October 31 |
12:33 | walterbender | tomeu: but also greatly broaden the environment in which we need to test |
12:33 | erikos: what lead time do they need? | |
12:33 | tomeu | that's true |
12:34 | sdziallas | tomeu, walterbender: as I threw the soas & cjb's fedora-xo kickstart files together, they're now based on the same stuff, so everybody get's the cool latest changes. and that way, sugar (0.84) will run better under F11 on the XO. |
12:34 | walterbender | tomeu: but I think we can agree that we do our baseline testing on a Fedora-based SoaS??? That would be a sigiifcant decision. |
12:34 | erikos | walterbender: that is from a draft: http://poelstra.fedorapeople.org/schedules/f-12/ |
12:34 | silbe | more automated testing (both unit tests and sugarbot) would help in the long run, though it cannot replace manual testing fully |
12:34 | erikos | silbe: you mean for the mute key etc? |
12:35 | silbe | erikos: yup, the one where my keyboard stopped working at all |
12:35 | walterbender | erikos: feature submission is end of May |
12:35 | how do people on other distros feel about using Fedora as the baseline? | |
12:35 | erikos | walterbender: it starts yup |
12:36 | silbe: you mean - it was not good to make that change at that date | |
12:36 | walterbender | erikos: oh. we have until end of July... |
12:36 | erikos | silbe: and in the future we should be more cautious? |
12:36 | tomeu | walterbender: for 0.86, I would say so |
12:36 | walterbender: we can revisit for 0.88, of course | |
12:36 | sdziallas | well, feature freeze is not final freeze, right? |
12:36 | silbe | erikos: i don't know what the "right" thing would be. it seemed like a nice, little change but turned out to have rather grave effects. |
12:37 | tomeu | walterbender: saw the email I sent about gnome 2.26, there's a link about how they do their live media images, I guess for testing purposes |
12:37 | ? | |
12:37 | erikos | silbe: we could have reverted it i guess |
12:37 | walterbender | sdziallas: I am just thinking that we need to know what features before we can freeze them |
12:37 | erikos | silbe: on the other hand - we fixed another bug which was only revealed that way |
12:38 | silbe: in general - i think that we will get better in the future and do not need to make so many last minute changes | |
12:38 | silbe | erikos: yes, but i don't like the idea it could have made it into the release |
12:38 | erikos | silbe: sure |
12:38 | sdziallas | walterbender: yeah, definitely agreed. but Fedora's feature freeze just means (afaik) that you need to have a current feature page, which has been approved by FESCo... |
12:38 | erikos | silbe: point taken - will try to be more strict with myself and us in the future |
12:39 | silbe: to avoid such things | |
12:39 | walterbender | sdziallas: OK. But regardless, we should plan to reach consensus on 0.86 features well before then |
12:39 | silbe | erikos: didn't mean to imply you were imcautious (sp?) |
12:39 | sdziallas | walterbender: agreed... |
12:40 | erikos | silbe: nah :) |
12:40 | walterbender | so we at least can agree we need to know <<July what are plans are for 0.86 |
12:41 | that already narrows things down. | |
12:41 | erikos | sdziallas: and the process is maybe the right thing for us as well |
12:42 | sdziallas: that the features need to be aproved | |
12:42 | sdziallas | erikos: yeah, I mean Fedora wouldn't be using the process if it would be bad one... |
12:43 | erikos | or do people think - this would slow down development - and people feel not couraged? |
12:43 | silbe | sdziallas: they how way more people, though |
12:43 | erikos | silbe: maybe we need to work towards this |
12:43 | silbe | i've thought about more strict rules as well, but additional load on the developers would hurt a lot |
12:44 | erikos | silbe: we do not need to make it super strict |
12:44 | sdziallas | silbe: that's also true... ;) |
12:44 | walterbender | erikos: I don't know that we need more strict rules, but we can tell people, here is the schedule, dirven by external events. |
12:44 | if you can meet it, great. If not, your work will probably end up in 0.88 | |
12:45 | and having a set of agreed upon features doesn't preclude people from working on others as well | |
12:45 | but it will enable people to rally around some concrete goals? | |
12:46 | tomeu | yeah, it's important that people understand that there are reason why we have rules |
12:46 | and that we have processes in place to make exceptions | |
12:47 | silbe | IMO the goals for 0.86 should be a) get Sugar to a point where it's interesting for a lot of people to contribute, esp. hackers (=> spread developer load) and b) start with automated testing (e.g. regression tests) to notice new issues faster (increase quality, reduce tester load) |
12:47 | purely subjective of course :) | |
12:47 | walterbender | silbe: good goals. we need to get everyone's goals on the table soon |
12:47 | but it is hard to argue with those two goals :) | |
12:47 | tomeu | silbe: actually, walter started a discussion about high level goals last week, but I failed to follow up |
12:48 | maybe we should do that | |
12:48 | erikos | walterbender: tomeu +1 |
12:48 | walterbender | I would like to have a framework in place for the diascussion |
12:48 | silbe | walterbender: that's a good idea. maybe even do releases more often so not getting a feature into a release doesn't hurt as much. would increase load on the release team, though. |
12:49 | tomeu | walterbender: maybe for the high level goals we don't need more than a plain old email thread? |
12:49 | walterbender | silbe: I am not proposing we churn out more than 1 per 6 months |
12:49 | sdziallas_ | runs, will be back later; sorry folks :( |
12:49 | tomeu | sdziallas_: cheers! |
12:50 | at the end, we'll see people pushing for some goals but having no one who will actually work on them | |
12:50 | so we don't actually need to reach consensus on the mailing list | |
12:50 | silbe | walterbender: personally, i'd go for 3 months (and perhaps even fork new development branches before the release), but that's probably not realistic for now |
12:50 | erikos | phone - sorry :( |
12:50 | walterbender | silbe: not until the secret cloning project is finished |
12:50 | tomeu | apart from cloning, there's the issue of distributors |
12:51 | walterbender | quick, while Simon's back is turned, grab a few skin cells |
12:51 | tomeu | right now, only distros that package sugar are distributing it |
12:51 | walterbender | tomeu: I agree. The bits need to get out into the world to be useful |
12:51 | and to get feedback | |
12:51 | tomeu | and they seem to be sync'ing themselves in two releases per yeat |
12:51 | walterbender | the intermediate (odd numbered) releases help |
12:51 | silbe | walterbender: secret cloning? |
12:52 | oh, think i got it :) | |
12:52 | tomeu | if in the future other people are getting sugar to kids at other times of the year, we can revisit making more frequent releases |
12:52 | walterbender | silbe: did you ever see the Woodie Allen movie, Sleeper :) |
12:52 | silbe | walterbender: not yet. will put it on the list :) |
12:54 | walterbender | so maybe a more aggressive call for feature suggestions and then a workshop or some event where we can drive to some consensus? |
12:54 | silbe | ok. Does SoaS pull the packages from Fedora? |
12:54 | tomeu | after having a bit of high level goals discussion in iaep? ;) |
12:54 | silbe: yup | |
12:55 | silbe | ok, so 6 months for now, fine. :) |
12:56 | tomeu | would be nice to have someone to take care of the feature pages in the wiki, but perhaps our wiki team is a bit overloaded? |
12:57 | erikos | walterbender: haha back! |
12:58 | silbe | it would be great to get rainbow in, BTW </shameless-plug> |
12:58 | walterbender | #ACTION walter will reconcile new features with trac |
12:58 | tomeu | silbe: I think we only need someone to work on integration |
12:58 | silbe | "reconcile new features"? |
12:59 | erikos | so, as we do development releases in the 6 months - we can get feedback earlier |
12:59 | silbe | ok, will try once i get the buildmaster stuff working as expected :) |
12:59 | walterbender | runs to the phone |
13:00 | erikos | and we have Soas where we can get feedback |
13:00 | tomeu | silbe: we talked earlier today about having all features as enhancements in trac, and the ones that are a bit complex, also a wiki page |
13:00 | silbe | ah, ok |
13:01 | so what exactly is walter going to do? add new feature requests in trac or trying to find them? | |
13:01 | tomeu | silbe: making sure they are properly linked and all features in the wiki have a ticket |
13:01 | walterbender | silbe both |
13:02 | but that list is only scratching the surface | |
13:02 | silbe | ok |
13:02 | walterbender | if we put out a call, we need to include a schedule and a process, which is what I asked erokos this morning |
13:03 | needs to spend more time with Typing Turtle | |
13:03 | tomeu | what would a process look like? |
13:04 | erikos | walterbender: i would go for a schedule like the one iposted earlier |
13:04 | walterbender | erikos: +1 |
13:04 | nothing complicated | |
13:04 | but we just need to agree upon it | |
13:05 | erikos | walterbender: yup - nearly two months for feature submission it says there |
13:05 | walterbender: maybe i should just daft one up and we can discuss it | |
13:06 | walterbender | erikos: where is "there"? |
13:06 | FGrose | There are new 'Roadmap' links on the TeamHeaders that can be used |
13:06 | erikos | http://poelstra.fedorapeople.org/schedules/f-12/ |
13:06 | silbe | erikos: erikos: got a pointer? cannot find any mail from you with a schedule in it |
13:07 | erikos | silbe: nah - there is not one yet ;p |
13:07 | silbe: http://poelstra.fedorapeople.org/schedules/f-12/ | |
13:07 | silbe | oh |
13:07 | erikos | silbe: i meant similar to that one |
13:07 | silbe | ok :) |
13:07 | erikos | would until next week work? |
13:07 | if i present a schedule we can talk about next week? | |
13:08 | walterbender | erikos: that is fine. I guess I was looking for the Sugar schedule, not the Fedora schedule :) |
13:10 | erikos | walterbender: yup - next week i will have the draft |
13:11 | walterbender: or we can meet on monday or tuesday for a special meeting | |
13:11 | tomeu | an extraordinary meeting? |
13:11 | walterbender | erikos: I cannot do Monday or Tuesday morning :( |
13:12 | erikos: why not just wait for the usual Thursday time? | |
13:12 | tomeu | yeah, and discuss in the mailing list if you send it earlier |
13:12 | silbe | don't think it's a good idea to do high level decisions on an extraordinary meeting |
13:13 | tomeu | I would actually take all decisions async on the mailing list, in fact |
13:13 | erikos | walterbender: works for me - i just thought people were probably waiting for it |
13:13 | silbe | tomeu: even better |
13:13 | erikos | ok - ml is it then |
13:13 | walterbender | erikos: I think as long as people know it is coming, that is OK |
13:14 | erikos | walterbender: great |
13:14 | walterbender: ok - anything else here? | |
13:14 | walterbender | erikos: wanna post an ACTION? |
13:15 | erikos | walterbender: don't like to post actions for myself ;) |
13:15 | ACTION: Schedule of Sucrose 0.86 - send draft to ml for further discussion (erikos) | |
13:15 | #ACTION: Schedule of Sucrose 0.86 - send draft to ml for further discussion (erikos) | |
13:17 | FGrose | erikos: Let's talk about wiki page to collect goals and ideas for roadmap. There are new 'Roadmap' links for each team/project on wiki |
13:17 | walterbender | erikos: I cannot manage to attend a meeting without getting stuck with an Action... bad habit |
13:17 | erikos | FGrose: ok |
13:18 | silbe | walterbender: you need to delegate more :-P |
13:18 | erikos | FGrose: you mean i should post the roadmap to the wiki - and wait for feedback? |
13:19 | FGrose | erikos: No ML is best, but then some semi static collection of ideas for easy referal |
13:20 | tomeu | if we cloned walterbender, he could delegate on himself |
13:21 | erikos | FGrose: i guess we can directly do that on the roadmap page |
13:21 | walterbender | silbe: says silbe, who has nothing on his plate :) |
13:21 | erikos | FGrose: http://wiki.sugarlabs.org/go/D[…]ease/Roadmap/0.86 |
13:22 | FGrose: oh - i see the links - nice | |
13:22 | silbe | walterbender: did well enough :-P - but actually i promised to work on rainbow once the buildmaster stuff is done |
13:22 | FGrose | erikos: right, but we can structure it better for feedback ideas maybe |
13:23 | erikos | FGrose: suggestions welcome - a concrete idea already? |
13:23 | FGrose | Let me think for a couple of hours |
13:24 | silbe | is the modules page still up to date - e.g. marco as sugar maintainer? |
13:24 | walterbender | silbe: we need to protect you and Aleksey and Sebastian from getting saddled with minutia. |
13:25 | erikos | FGrose: ok - awesome |
13:25 | FGrose | erikos: Mailing list thread is best for wide participation, but point people to wiki for landed ideas |
13:26 | silbe | walterbender: agreed in general - but what do you mean exactly? |
13:26 | tomeu | silbe: marco passed on maintainance, we should update that |
13:26 | walterbender | silbe: nothing in particular... just a general observation |
13:27 | silbe | walterbender: ok, full ack then :) |
13:27 | walterbender | silbe: not that it wouldn't be great to get Rainbow into 0.86!! |
13:28 | silbe | walterbender: it's my personal release goal :) |
13:28 | tomeu | it's quite doable, I think |
13:28 | silbe | AFAICT the datastore part will be the biggest obstacle |
13:29 | but i really need to dive into it to be sure | |
13:29 | tomeu | silbe: yeah, I had some ideas about doing it with dbus, but I think the new rainbow dumped dbus completely? |
13:30 | silbe | tomeu: i think so, yes. the olpcfs2 could be a nice solution if it were already implemented :) |
13:30 | tomeu | basically, the DS needs to ask rainbow if the activity is limited to reading entries with a particular activity_id or to a particular mime_type list |
13:31 | silbe: but the fuse daemon would have the same problem, right? | |
13:32 | silbe | ah, right, rainbow will need to interact with it, not just make it work... |
13:32 | tomeu | silbe: you can also implement olpcfs2 for 0.86 :p |
13:33 | silbe | even if i had the time to do it, i don't think it would be release ready for 0.86. |
13:34 | erikos | tomeu: i read the thing now |
13:34 | tomeu | sure, just joking |
13:35 | any other topics for the meeting? | |
13:35 | can we close it? | |
13:36 | I think I'm quite clean now of non-coding duties, so maybe tomorrow we can make it a bug-fixing day in preparation of 0.84.1? | |
13:37 | silbe | maybe i'll do it as a dissertation if it hasn't been done by then ;) |
13:39 | erikos | #endmeeting |
« Previous day | Index | Today | Next day » Channels | Search | Join