=== salgado-afk is now known as salgado === mrevell is now known as mrevell-lunch === mrevell-lunch is now known as mrevell === salgado is now known as salgado-lunch [15:01] #startmeeting [15:01] Meeting started at 09:01. The chair is bac. [15:01] Commands Available: [TOPIC], [IDEA], [ACTION], [AGREED], [LINK], [VOTE] [15:01] hello everyone. welcome to the first reviewer's meeting of 2010. [15:01] who's here? [15:01] me [15:01] me [15:01] me [15:01] me [15:01] me [15:02] me [15:02] me [15:02] me [15:03] let me try to round up some folks [15:03] * gmb has summond the Bugs team reviewers [15:03] me [15:04] me [15:04] ... see? [15:04] me [15:04] gmb: and your leader? [15:04] bac: Has been pung. [15:04] BjornT: ping [15:04] bac: jtv has been having connectivity troubles and danilo is off. [15:05] thanks henninge [15:05] well, let's get started. if you notice an absence from one of your team members please follow gmb's good example and harass them. [15:06] [TOPIC] agenda [15:06] New Topic: agenda [15:06] * Roll call [15:06] * Action items [15:06] * invite other teams to do lazr-js code reviews? [mars/barry] [15:06] * Peanut gallery (anything not on the agenda) [15:06] * Reminder: Who can do JS reviews? All reviewers? [henninge,allenap] [15:06] * Proposed coding standard for YUI modules. [Edwin] [15:06] * Cleaning up outstanding approved branches on +activereviews [bac] [15:06] * New developers as mentats? [bac] [15:06] * Meeting frequency [bac] [15:06] bac: I am the sole member of the Code team on this meeting, but I welcome harassment from others. [15:06] abentley: what about rockstar? [15:06] bac: He does the other one. [15:07] [TOPIC] action items [15:07] New Topic: action items [15:07] rockstar joins ameu [15:07] first, an unlisted item -- we all owe many thanks to barry for his long service getting the group together and chairing. thanks barry and have fun in foundations! [15:07] yeah, thanks barry! [15:08] ees a jolly good fella [15:08] barry: Thanks! [15:08] [TOPIC] * invite other teams to do lazr-js code reviews? [mars/barry] [15:08] New Topic: * invite other teams to do lazr-js code reviews? [mars/barry] [15:08] * barry blushes - you're welcome! i have no doubt bac will great improve the governance of this team :) [15:08] me [15:08] i'm not sure if this is leftover from our last meeting so long ago [15:09] mars isn't here, so do you recall barry? [15:09] i vaguely remember an ml discussion about this from way back last year [15:09] i think it would be a good idea to do cross-team reviews of lazr-js, but iirc mars was -0 on it [15:10] i don't remember why (something about the code not being ready yet?) [15:10] ok, i'll take it on to review the ML to see if i can find a discussion and talk to mars to see if we want to pursue it. [15:10] [TOPIC] * Reminder: Who can do JS reviews? All reviewers? [henninge,allenap] [15:10] New Topic: * Reminder: Who can do JS reviews? All reviewers? [henninge,allenap] [15:11] henninge: is this a current issue? if so, please proceed. [15:11] bac: it came up in a review [15:12] I think allenap wasn't sure if he could review my JS code because he was never officially knighted as "JS reviewer" [15:12] all reviewers can and should do js reviewes [15:12] *ui* reviews are a different matter [15:12] barry: that's what I remember, thanks. [15:12] well [15:12] yup [15:12] i agree, by now there should be no reason why anyone can't do js reviews [15:12] that's my feeling. though i know i've seen some people who don't consider them experts defer. i've done it myself. [15:13] we did have a similar JS review approval process [15:13] graduated reviewers were the UI/AJAX team members [15:13] that attended the Berlin sprint [15:13] and if someone doesn't feel comfortable enough then they can work with someone else [15:13] bac: right. that's not to say a reviewer can't ask for help, with js or even python [15:13] where JS coding guidlines were established [15:13] but we never graduated anybody after that [15:13] and didn't make the process very formal either [15:14] fjlacoste: i'm nearly certain we decided to throw everyone in the deep end :) [15:14] fjlacoste: perhaps we consider those people as resources but everyone should attemp to do JS reviews to their comfort level [15:14] right [15:14] everyone was considered a mentee [15:14] well [15:14] we didn't setup a formal mentoring process around this [15:14] we shuld clarify that situation [15:15] and update the reviewer pages [15:15] accordingly [15:15] +1 on updating the reviewer page [15:15] perhaps we need a volunteer to herd the JS reviewers. anyone? [15:16] I thought we didn't have such a group? === fjlacoste is now known as flacoste [15:17] henninge: i don't recall [15:17] bac: i think this should be someone from the UI/AJAX team [15:18] flacoste: agreed. [15:18] EdwinGrubbs: would you be interested? [15:18] bac: What I meant is, if all reviewers are JS reviewers, there is no such special group, is there? [15:19] bac: to herd js reviewers? don't we have a list of them already in the wiki. [15:19] And it already says: https://dev.launchpad.net/ReviewerSchedule [15:19] A Note on JavaScript reviews: Any reviewer can handle a JavaScript review, if they feel comfortable doing so. For now, we ask that their review by seconded by one of the JavaScript specialists. [15:20] thanks noodles775 [15:20] it looks like there is no action necessary. [15:20] let's move on. [15:20] [TOPIC] * Proposed coding standard for YUI modules. [Edwin] [15:20] New Topic: * Proposed coding standard for YUI modules. [Edwin] [15:21] noodles775: So this means there is a "JavaScript specialists" group. Do we have an easy way to find its members? [15:21] * henninge waits for the clarification on the wiki ... ;-) [15:21] abentley: the list on that page identifies them I think... [15:21] abentley: that page lists javascript reviewers in the last column [15:21] (you can update yourself as a resource of course) [15:22] noodles775, bac: sounds fine. [15:22] you may also want to look at the inconsistencies we currently have with JS module names and the namespaces they define: https://pastebin.canonical.com/25818/ [15:23] That looks like a good topic in itself :) [15:23] EdwinGrubbs: the floor is yours for your YUI topic. [15:23] I knew milestone_table would bite me [15:23] I'm suggesting that we name our JS modules more like how python modules must be named. More info is available at https://dev.launchpad.net/ReviewerMeetingAgenda but I"ll summarize [15:24] 1. The module name should match the directory structure. E.g. javascript/registry/timeline.js should use YUI().add('registry.timeline', ... [15:25] 2. The namespace should match the module name, so we should put methods in the namespace like this Y.registry.timeline.someFunction() instead of Y.registry.someFunction(). [15:26] does anybody disagree with that plan? [15:27] Not me - it would be good to not have to think about those decisions :) [15:27] +1 [15:27] your next question should be who volunteers to fix these [15:27] +1 [15:27] +1 [15:27] +1 [15:27] +1 [15:28] so we seem to agree it's a good idea. which leads to curtis' question of who and when to do the clean up. [15:28] I can open up bugs for the inconsistent modules and assign them to the respective teams. [15:28] since the code is already divided by app, each team can take care of their own [15:29] EdwinGrubbs: that would be great. [15:29] thanks Edwin [15:30] [ACTION] Edwin to file bugs on JS naming inconsistencies and teams will take care of doing the clean up. [15:30] ACTION received: Edwin to file bugs on JS naming inconsistencies and teams will take care of doing the clean up. [15:30] [TOPIC] * Cleaning up outstanding approved branches on +activereviews [bac] [15:30] New Topic: * Cleaning up outstanding approved branches on +activereviews [bac] [15:31] i noticed yesterday when doing OCR that we've got a large number of approved branches that haven't landed. [15:32] is this work abandoned after review, blocked, other? [15:32] bac: Mine were blocked on test suite issues, but are now moving again. [15:32] if the former perhaps the state of the MP can changed to reflect it and clear out that list. [15:34] bac: maybe each reviewer at the start/end of their shift try and chase those MPs in question [15:34] i'm glad that tim created the list and think we should strive to keep it minimal. any other thoughts? [15:34] it's a bit of a bother, but it will probably help and is not hard to do [15:34] bac: For the first case, they can be marked "rejected" or "work-in-progress", as appropriate. [15:35] abentley: right. [15:35] e.g. jelmer's branch was approved, but it turns out there are some issues that require further investigation. [15:36] intellectronica: perhaps. or we might just monitor it weekly, perhaps keeping it as an item for this meeting until the backlog is handled [15:36] for now, i just ask that each developer look at his branches and take the necessary action. [15:37] [TOPIC] * New developers as mentats? [bac] [15:37] New Topic: * New developers as mentats? [bac] [15:38] we've hired a few new people and i was wondering if anyone was ready to enter the reviewer mentat program. [15:38] i think team leads should be responsible for nominating their developers as appropriate [15:39] moving on to the final topic [15:39] [TOPIC] * Meeting frequency [bac] [15:39] New Topic: * Meeting frequency [bac] [15:40] in the past we have met weekly. starting the new year i think we should look at whether we want to continue weekly meetings or move to biweekly. [15:40] * bigjools agrees with bac [15:41] I would prefer to continue meeting weekly, because it's easier to remember. [15:41] I say stick with weekly, if there's nothing to discuss it's not a problem is it? We just finish quickly. [15:41] i think bi-weekly will be enough. maybe we can alternate the eu and pacific meetings, so that if you really want to join a meeting on a given week you can have the option of joining out of office hours [15:41] bigjools: there is a bit of overhead to a meeting [15:42] bigjools: alternatively, if there are no/few items on the agenda i can pre-emptively cancel the meeting [15:42] not a metric! [15:42] bigjools: but, as a rule it would go on as planned on a weekly basis [15:43] I'm +0 on keeping it weekly for the sake of my godawful memory. [15:43] I don't see the problem personally [15:43] i just don't want to cause interruption to everyone's schedule if the meeting is not serving a purpose [15:43] also, many of the topics we discuss in these meetings can probably be discussed more productively on the mailing list anyway [15:44] bac: Well, if it's going to break up an important piece of work we can always send our apologies... [15:44] ok, it sounds like there is enough sentiment to continue weekly. [15:44] lastly [15:44] [TOPIC] Peanut gallery [15:44] New Topic: Peanut gallery [15:45] anyone have an item they'd like to (briefly) discuss? [15:45] * bigjools raises hand [15:45] go bigjools [15:45] * abentley raises hand [15:45] abentley on deck [15:45] very quickly, the current edge non-updating is because we let an API change land which didn't have security protection [15:46] so this is a reminder to be vigilant when reviewing API changes [15:46] * bigjools out [15:46] bigjools: Thanks for raising my topic [15:46] heh [15:46] thanks bigjools and abentley [15:46] I wanted to ask if we think there is anything else we should do. [15:47] The outcome of the discussion was that this should have been caught in review. [15:47] reviewers' checklists? [15:47] bigjools: Wouldn't hurt. [15:47] maybe a template for review replies. Didn't we used to have one? :) [15:47] bigjools: You mean for review requests? [15:47] no, replies [15:48] I remember using barry's reviewing tool [15:48] bigjools: Must have been before my time. [15:48] Of course, we've exposed a lot of APIs, and if this is the only security issue we've had, we're doing pretty well. [15:48] But what if it's not? [15:49] it's not the first time it's happened [15:49] Would it be a good idea to audit our API? [15:49] I think security should be constantly on reviewers' minds [15:50] we can continue this on the list [15:50] we're OOT [15:50] yep [15:50] bigjools: Okay. [15:50] thanks for coming and contributing [15:50] #endmeeting [15:50] Meeting finished at 09:50. [15:51] Thanks bac! === salgado-lunch is now known as salgado === EdwinGrubbs is now known as Edwin-lunch === Edwin-lunch is now known as EdwinGrubbs === noodles76 is now known as noodles775 [21:00] #startmeeting [21:00] Meeting started at 15:00. The chair is bac. [21:00] Commands Available: [TOPIC], [IDEA], [ACTION], [AGREED], [LINK], [VOTE] [21:01] Hi all, welcome to the AsiaPac version of the Launchpad Reviewers Meeting. Who's here today? [21:01] hello? echo? [21:01] rockstar, mwhudson, thumper: ping [21:02] hi [21:02] hi [21:02] wgrant, ping? [21:02] hey [21:02] [TOPIC] agenda [21:02] New Topic: agenda [21:03] * Roll call [21:03] * Action items [21:03] * invite other teams to do lazr-js code reviews? [mars/barry] [21:03] * Reminder: Who can do JS reviews? All reviewers? [henninge,allenap] [21:03] * Proposed coding standard for YUI modules. [Edwin] [21:03] * Cleaning up outstanding '''approved''' branches on [[https://code.edge.launchpad.net/launchpad/+activereviews| +activereviews]] [bac] [21:03] * New developers as mentats? [bac] [21:03] * Meeting frequency [bac] [21:03] * Peanut gallery (anything not on the agenda) [21:03] [TOPIC] summary of the AMEU meeting [21:03] New Topic: summary of the AMEU meeting [21:04] so we went over the items listed above. the lazr-js one was a non-starter as barry didn't remember what it was about and mars was awol. [21:05] i see :) [21:05] a much run-around regarding JS reviews, noodles reminded us the page at https://dev.launchpad.net/ReviewerSchedule listed all of the JS gurus. everyone should do JS reviews but refer to the experts if needed [21:06] edwin made a very well received suggestion about JS namespaces and is going to file bugs so each team can clean up their code to adhere to the new idea. [21:07] thumper i championed the idea of cleaning up approved branches on +activereviews. a few people had reasonable explanations for why branches lingered but most are still a mystery. hopefully we can drive that list down [21:07] why do the work and get it approved only to let the branch linger if there is not a clear blocker? [21:07] awesome [21:08] i encouraged team leads to nominate new developers to start the mentoring process when ready but that only applies to one or two people, and certainly no code folks [21:09] since we went about six weeks without a meeting and nothing caught fire i floated the idea of doing these reviewer meetings less frequently. to my surprise the concensus was to stay with weekly meetings. [21:09] bac, code tema is already in the process of cleaning up their namespaces. [21:09] i reserved the right to cancel a meeting if there is nothing on the agenda. so if you have something to talk about please put it on the wiki. [21:09] rockstar: great! [21:10] and that was the meeting. took us 50 minutes to cover all of that. [21:10] so do any of you have anything for the "Peanut Gallery" section? [21:11] * mwhudson doesn't think so [21:11] * rockstar doesn't [21:11] oh, yeah, we heaped lots of praise on barry for chairing these meetings for so long [21:11] bac, also, I don't think we're going to have this meeting next week, since we're all sprinting. [21:12] rockstar: ok, great. you going to miss your snow? [21:12] bac, I think I'll risk it. [21:13] ok, then. thanks for coming. if no one has anything else let's call this meeting done. [21:13] #endmeeting [21:13] Meeting finished at 15:13. [21:14] thanks bac [21:15] thanks bac [21:15] np [21:15] hey thumper are you CHR? [21:17] yep [21:17] thumper: have you been answer email to feedback? if so, please CC the feedback list when you do. [21:17] er, 'answering' [21:17] I just got to it yesterday [21:17] and yes, I'll be CCing the list [21:17] ok. [21:17] see my email from yesterday [21:17] thanks [21:18] i'll look