Calendar:QA Chat:2007-10-25:Log

[18:29] <Andreas> What is going on with the chat agenda?
[18:36] <mschroeder> so... let's start the qa chat.
[18:37] <mschroeder> I was out of town, so I haven't created a wiki page.
[18:37] <Andreas> ok
[18:38] <mschroeder> 0.7 is finally released.
[18:38] <mschroeder> Hi Clint.
[18:38] <ctalbert> Hi all
[18:38] <ctalbert> mschroeder: 0.7 is *released*?
[18:38] <ctalbert> I thought it was RC3
[18:39] <ulf> Hi Clint!
[18:39] <ulf> ctalbert: surprise, surprise :-)
[18:39] <daniel_> ctalbert: we checked the last fixes of RC3, so ause pushed 0.7 to stage
[18:40] <ulf> it's out; try updating your 0.5 Sunbird; it's all there
[18:40] <daniel_> except for the sources
[18:43] <mschroeder> Now we can start our post-0.7 QA activities. :)
[18:44] <ctalbert> Sorry about that
[18:44] <ctalbert> Not sure what happened.
[18:44] <ctalbert> Ulf: you want me to try an upgrade from 0.5 to 0.7?
[18:47] <mschroeder> Anyone interested in a QA chat? ;)
[18:48] <ctalbert> Yes, I thought it was ongoing.
[18:49] <mschroeder> Good. :) Post-0.7 QA activities.
[18:50] <ulf> ctalbert: you don't have to (I've been doing that way too often fort the last 7 days) But try it out, if you like
[18:50] <ulf> mschroeder: have a party?
[18:50] * mschroeder has started to dive into Automated Testing with XPCShell.
[18:51] <ctalbert> Andy Lawrence asked me an XPCOM question the other day
[18:52] <ulf> mschroeder: means we're no longer supposed to do any manual testing any more? :-) sounds great!
[18:52] <mschroeder> that would be cool but...
[18:52] <ctalbert> He wanted to know if  do_check_true was supported or not.  He wasn't able to get it to work
[18:53] <ctalbert> It evidently works on trunk, but folks I asked weren't sure if it would work on branch.
[18:53] <mschroeder> ctalbert: I reviewed a patch by daniel which should fix that on MOZILLA_1_8_BRANCH
[18:53] <ctalbert> When was that?
[18:54] <mschroeder> https://bugzilla.mozilla.org/attachment.cgi?id=284939
[18:54] <mschroeder> bug 398724
[18:54] <firebot> mschroeder: Bug https://bugzilla.mozilla.org/show_bug.cgi?id=398724 nor, --, ---, daniel.boelzle@sun.com, ASSI, Problems with floating all-day items
[18:56] <ctalbert> Ah, cool, so it isn't checked in yet
[18:56] <ctalbert> Ok, I'll send him an email about that.
[18:56] <ctalbert> thanks mschroeder 
[18:58] <mschroeder> What should QA do better in the next development/release cycle?
[18:59] <ulf> mschroeder: that's a good question
[19:00] <ctalbert> I think trying to focus on broadening our automated tests is a good idea - i.e. the automation summit stuff.
[19:01] <ulf> I'd suggest to start a discussion on the ng about that and later on discuss/sort out things
[19:01] <ulf> what do you think about the aproach?
[19:02] <ctalbert> I think that's a good idea.  Give a bunch more people the ability to drop in and say their two cents.
[19:02] <mschroeder> +1
[19:02] <ctalbert> Overall though, I have to say I think that QA did a pretty fine job with this release.
[19:02] <ctalbert> INMHO
[19:03] <ctalbert> All you Germans should get together and have a party.  :D
[19:03] <sipaq> ctalbert: absolutely
[19:03] <mschroeder> I think Litmus wasn't so effective this time.
[19:04] <ctalbert> Do you think that is because Litmus is very out of date with the product or because we focussed more on ad-hoc testing this time?
[19:04] <ctalbert> or both?
[19:04] <ulf> ctalbert: I like the idea of a release party
[19:05] <mschroeder> ctalbert: It's the out-of-date problem.
[19:05] <ulf> ctalbert: good analysis - indeed I think it's a mixture of both
[19:05] <mschroeder> ctalbert: Most of the new testers like to use it.
[19:06] <ulf> we really need to update some test cases and make them easier to run
[19:06] <ctalbert> I think that it is definitely out of date, which is an issue.  I also think that we can put more interesting tests into it.  But there is still no substitute for someone who really knows the code and the product doing targetted testing around a certain feature.
[19:07] <ulf> absolutely!
[19:07] <mschroeder> that's true.
[19:08] <ulf> and perhaps we should try and get more folks at test days to run litmus test cases 
[19:09] <mschroeder> The number of folks attending testdays depends on the new features.
[19:09] <ctalbert> I think it also depends on how well we "sell" the test day.
[19:09] <mschroeder> This time the best testday was the one after we announced the prototype event dialog.
[19:09] <ctalbert> Here is what I've kind of learned (and this should probably also go to the newsgroup)
[19:10] <ctalbert> New features are imortant
[19:10] <ctalbert> Also, doing the post on sunday night (so people see it on monday) seems to work best.
[19:10] <ctalbert> Doing a very in depth post with more detail always seems to bring more people out than a shorter one
[19:11] <ctalbert> those are the things I've learned so far.
[19:11] <ctalbert> Of course it is never very predictable.
[19:12] <mschroeder> Should we change the testday time or day?
[19:13] <ulf> mschroeder: which would support what ctalbert said about detailed posts (remember sipaqs 'screenshot-post' promoting the the prototype UI bits)
[19:13] <ulf> for me Tuesday afternoon has become difficult
[19:13] <ctalbert> I don't think we should change the day.  I think people are getting accustomed to the Tuesday Test day,but I think that wuld be a good topic for the newsgroup
[19:14] <ulf> but that's more of a personal issue for me (since I have two meetings colliding w. the testday)
[19:14] <ctalbert> One other problem is that every tuesday, the Mozilla QA team has a "bug triage day" and that might force people to choose between the projects
[19:15] <ulf> ah interesting to hear; I didn't knew that
[19:15] <ctalbert> I think that it might be an issue worth exploring.  I've seen several people in the broader community always say that Tuesday is a busy day for them.
[19:16] <ulf> then, shouldn't we try and find a different day?
[19:16] <ctalbert> I think it might be good.  
[19:17] <ctalbert> We should probably take all of this stuff to the newsgroup.
[19:17] <mschroeder> We should at least try it.
[19:17] <ctalbert> And see if we can get a day that a bunch of people can agree on .
[19:18] <ulf> +1
[19:20] <ctalbert> mschroeder: will you write the initial post to the NG?  And I'll comment about this test day reorg stuff?  Or should the test day reorg be a separate post on its own?
[19:21] <ulf> I'd vote for a seperate post
[19:22] <mschroeder> ctalbert: I'll send you a draft for the initial post (as usual ;)).
[19:22] <ctalbert> Sounds good, I was kind of leaning toward that approach myself
[19:22] <ctalbert> ok.  
[19:22] <ctalbert> I'll write the one on the test day reorganization
[19:22] <mschroeder> perfect
[19:23] <mschroeder> sorry guys, I have to leave now.
[19:23] * mschroeder is now known as mschroeder|away
[19:24] <mschroeder|away> will be back in an hour ;)
[19:24] <ctalbert> And I think we're out of time anyway.
[19:24] <ctalbert> Happy 0.7 everyone!