/srv/irclogs.ubuntu.com/2010/04/29/#ubuntu-reviews.txt

bdmurraynigelbabu: it makes sense to me if we only want notifications of bugs with patches00:00
nigelbabubdmurray:  only trouble now is we we dont know how many people are working on reviews00:47
nigelbabuok, you can know by mooking at moderation queue00:47
bdmurraynigelbabu: actually any comment on a bug that the team is subscribed to will end up in the moderation queue - which also means metoo's will go there - which I think is a good thing00:52
nigelbabubdmurray: metoo?00:55
bdmurraynigelbabu: if some random person comments on a bug that happens to have the team subscribed it won't go to the mailing list00:56
nigelbabuah00:56
nigelbabuonly not everyone is subscribed00:57
nigelbabubdmurray: is there anyway to be midway?00:59
nigelbabui.e. have only members of the team and people who've applied be able to post00:59
bdmurraymaybe one could update the list of posters with the list of team members?01:00
nigelbabuor we can say subscribe to list but keep web only if you dont want mails01:00
nigelbabuthat way we can know how many people actually review bugs, so easier for dholbach to deal with membership requests01:01
nigelbabubdmurray: btw, the graphs are supposed to show u-m-s and u-u-s too? aren't they redundant?01:03
bdmurraynigelbabu: not for the long term historical perspective01:13
nigelbabuoh, ok :)01:14
nigelbabubdmurray: you're admin on the list right?01:14
bdmurraynigelbabu: of the patch review mailing list? yes01:15
nigelbabuok :)01:15
nigelbabuI'll think of something to add to the wiki01:15
nigelbabualso, is it possible t get a list of mails in moderation queue? (I dont know how mailman works)01:16
nigelbabuso on patch day if I want to know who helped and in what way01:16
bdmurraynigelbabu: probably but I'm not certain how to best get it in a usable format01:24
nigelbabubdmurray: any thing would be fine.  even a tar perhaps01:27
=== yofel_ is now known as yofel
dholbachgood morning06:45
* ajmitch waves06:49
ajmitchso I suppose we've got a whole pile of fixes to review that should have gone into lucid06:49
nigelbabuajmitch: just fyi, we have a whole pile of stuff from before  dapper ;)07:09
ajmitchI'm not at all surprised07:10
nigelbabuthough for now we're giong from latest patches to older07:10
nigelbabuCiemon: the point where we are in the release cycle also influences the "getting into debian vs getting into ubuntu"09:50
Ciemonok.. but the importance of the bug is also a major influence?09:53
nigelbabuof course09:55
nigelbabudholbach: I'm working on a new Getting Started page, and we'll move the Review Guide elsewhere10:00
dholbachnice10:00
dholbachgive me a shout if you want me to doublecheck or review or anything10:01
nigelbabuyes, once I finish :)10:01
nigelbabuI need your help later in writing html with python10:01
dholbachok :)10:01
* ajmitch sees some life in the channel10:15
nigelbabuajmitch: ;)10:16
* ajmitch is just bored waiting for a slow computer to install lucid packages10:17
nigelbabuI will only do a clean install10:17
ajmitchI only do a clean install on getting new hardware10:17
ajmitchmy original ubuntu install used to be a debian install carefully upgraded to ubuntu :)10:18
nigelbabuwow10:18
ajmitchthere wasn't as much divergence from debian when hoary was in development10:19
* ajmitch feels so old :(10:20
nigelbabuhttp://pad.ubuntu-uk.org/Z4QAoFdt7v10:21
nigelbabuajmitch, dholbach, persia (and anyone else intersted) thats a potential new getting started page ^10:21
* ajmitch reads 10:22
nigelbabuI'm making the joining M/L a requirement.10:22
nigelbabuIt helps keep track of contributors10:23
ajmitchok10:23
* ajmitch joins it10:23
ajmitchyou don't prefer them to be in a LP team?10:23
persiaI like it, except that I disagree with the mailing list bit.10:23
ajmitchgood evening persia10:23
nigelbabupersia: did you see my reasoning?10:24
persiaBut that's mostly from a cui bono point of view, rather than any specific set of objections.10:24
persiaYeah.  I still don't see any benefit *to me* to being on the mailing list.10:24
nigelbabuno, there isn't much10:24
nigelbabubut for new contributors, yes10:24
ajmitchnigelbabu: it refers to ~ubuntu-reviews, should that be ~ubuntu-reviewers?10:24
persianigelbabu: How is there benefit *to the new contributor* to join the mailing list?10:25
nigelbabupersia: I know thta they have contributed beforehand10:25
nigelbabuI see the mails everyday10:25
persiaHow is this a benefit to them?10:25
nigelbabuor we can checkup on the mailing list to see it10:25
ajmitchthat doesn't help them, just doing the work helps them10:26
nigelbabueasier for membership?10:26
nigelbabuoh yeah, it helps us10:26
ajmitchyou could see activity from a person whether they're on the list or not10:26
persiaNot really, since anyone who does anythig reasonable will have a couple good examples anyway.10:26
nigelbabuAgreed. Scratching.10:26
nigelbabuRemove the whole thing off? Or just give an option?10:27
ajmitchGive it as an option if they want to keep up with general review activity that's going on10:27
persiaMInd you, it makes it *lots* easier to track stuff.  I just don't think there's any value to making something mandatory or requiring it unless it can be made to offer a benefit to the person who must perform the action.10:27
ajmitchBut keep the high-traffic warning :)10:27
nigelbabu10 mails a day is high traffic?10:28
ajmitchThis would end up with every comment & status change to every bug the team is subscribed to, right?10:28
nigelbabuI said high traffic becuase it would be when everyone subscribed10:28
nigelbabuno, only when subscribers comment or change anything10:28
nigelbabulike if emet makes a change, I wont see it because he isnt subscribed10:29
persiaI'm not sure I understand the value of the mailing list at all then.10:29
nigelbabuonly benefit is getting a crack at the newest patches10:29
persiaIf it tracks *everything* that's useful.  If it is a discussion forum, that's useful.  If it tracks an arbitrary subset of stuff, seems pointless.10:29
ajmitchnigelbabu: I thought this would be for tracking bugs that ~ubuntu-reviewers is subscribed to, and that you'd keep the team subscribed to bugs10:29
ajmitchOr do you prefer to have the reviewers unsubscribe the team, like with -sponsors?10:29
nigelbabuwhen the review is over, they are to be unsubscribed10:30
nigelbabupersia: You've got a point there.  Should we move our disscusions to the M/l from now?10:30
ajmitchso it'll track all the comments & status changes on the bug until that point10:30
ajmitch(assuming that the mailing list is the contact address for the team)10:31
* persia isn't a big ML user :)10:31
* ajmitch has the firehose on for a select set of packages10:31
nigelbabuajmitch: no.  Oddly, it does not.  It only does so for the subscribers of the M/l (yeah, even I'm confused)10:31
ajmitchnigelbabu: Something is fishy there10:32
nigelbabuajmitch: I dunno mailmain backend, so not much clue10:32
ajmitchcontact address is set for the team, so it *ought* to be getting info for any bug the team is subscribed to10:32
nigelbabupersia: changed the M/L stuff10:32
nigelbabuajmitch: stuck in moderation I guess10:33
ajmitchsince LP is meant to be whitelisted for lists.u.c, I think10:33
ajmitchI may be wrong there10:33
* ajmitch isn't a team member anyway :)10:33
nigelbabupersia: +1 for changes now?10:34
persianigelbabu: Sure, but I still think the concept of the mailing list needs more thought :)10:39
nigelbabupersia: you dont want it at all?10:39
persiaI think there needs to be a point.10:39
ajmitchabout the only thing I'd need to be in the team for is unsubscribing the team, right?10:40
persiaI don't understand the value *to anyone* of a mailing list that tracks an arbitrary subset of changes.10:40
* nigelbabu is not the person to talk bout it10:40
nigelbabuajmitch: yup10:40
nigelbabus/to talk bout/to talk to 'bout/10:40
ajmitchI'll put it off until I annoy dholbach about it then :)10:41
nigelbabuhah10:41
nigelbabuany corrections, please make it https://wiki.ubuntu.com/NigelBabu/SandBox10:51
nigelbabumeanwhile, I'll get the review guide sorted10:51
* ajmitch is *still* waiting for synaptic to install packages :)10:53
nigelbabuokay ^ that link is redundant now :)10:59
ajmitchwell this upgrade is going well11:01
ajmitchX just went byebye11:01
nigelbabuthe wiki has now changed a bit11:13
ajmitchwhich page?11:14
nigelbabuhttps://wiki.ubuntu.com/ReviewersTeam11:15
ajmitchReviewersTeam/GettingInvolved ?11:15
nigelbabusee the header11:15
nigelbabuthe new getting involved page is in + review guide created11:15
nigelbabunow to edit review guide11:15
* ajmitch sees some bugs on the patch-accepted-debian query that he should really be touching asap11:16
nigelbabuajmitch: yaay! (that moves fast most of the time ;)11:19
ajmitchthough it's being handled by someone on the server team as expected11:19
ajmitch(the php5/mysql issue)11:19
ajmitchthat was a case where the patch originated from upstream & debian11:20
nigelbabuah11:21
nigelbabuthe getting involved page looks nice now?11:21
ajmitchyeah, looks good11:21
ajmitchmaybe some minor grammatical fixes, but apart from that it's good :)11:21
nigelbabuplease go ahead and fix as needed :)11:23
nigelbabuwhat is the easiest way to test patches?11:27
nigelbabu(the only way I know is to create a deb)11:28
dholbachapply the patch (in most cases something like "patch -p1 < ~/patch") or in the case of a merge proposal: "bzr merge lp:.....", then rebuild the source package (something like "debuild -S -uc -us"), then pbuilder it11:29
nigelbabuwell, thats the way I thought too :)11:29
nigelbabuthen dpkg -i11:29
dholbachright11:30
dholbachit's the best way because you afterwards know that the package still builds and if you messed something up, you can easily replace it with the "old package version" again11:30
nigelbabuhm, makses sense :)11:30
nigelbabusaw the new wiki pages?11:31
dholbachnot yet, I'll review in a sec11:34
ajmitchthe hard part is testing if the patch did the right thing11:40
ajmitchespecially if it was to fix some obscure race condition that's hard to reproduce in testing11:40
dholbachwould it make sense to add a step 0 to the list that is something like "establish step-by-step reproducing instructions"?11:42
ajmitchit would, that should be an important part of knowing that the bug can be closed as being fixed11:45
dholbachnigelbabu: ^ do you think that'd make sense?11:46
nigelbabudholbach: yes it does11:47
dholbachgreat11:47
nigelbabuthis is the review guide right?11:47
nigelbabusaw the getting started page?11:47
ajmitchyeah it'd be for the ReviewGuide page, in the workflow section I think11:48
dholbachnice work11:50
dholbachit looks much better11:50
ajmitchaw, silenced in -release, how unfair :)12:17
nigelbabuajmitch: I think mailing that attorney should help12:42
nigelbabuProkurist means attorney I think12:43
ajmitchI'll let ogra deal with it, he's in .de12:43
nigelbabuha12:43
* ajmitch doesn't think there'd be much of a legal case, it's more a matter of not doing stupid things12:44
nigelbabuthat too12:45
nigelbabuok, I'm going to start coding12:49
nigelbabuI need to do that review overviewtoday12:49
ajmitchok, cool12:50
* ajmitch has tomorrow off, so might try & catch up on ubuntu stuff 12:50
nigelbabuI haz working statistics!13:42
nigelbabunow we can know exactly how many bugs would be in review queue if there was no date criteria13:42
ajmitchsounds good13:43
nigelbabubdmurray: poke14:36
=== yofel_ is now known as yofel

Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!