18:55:04 #startmeeting 18:55:04 Meeting started Wed Apr 8 18:55:04 2015 UTC. The chair is colonelqubit. Plugin info at http://wiki.debian.org/MeetBot. 18:55:04 Useful Commands: #action #agreed #help #info #idea #link #topic. 18:55:08 Huzzah 18:55:24 Hello assorted QA folks 18:55:42 let's take a quick look at our stats 18:57:04 310 UNCONFIRMED bugs -- looks like we need to work a bit more at chipping away there 18:57:22 needAdvice: 11 18:57:31 Not too bad 18:59:14 So let's see... several things to bring up: Android testing 18:59:27 mjayfrancis had some notes for me to relay 19:00:08 But first, let's take a look at our previous action items 19:00:37 I haven't had time to work on Bugzilla, as is perhaps evidenced by our glorious DFD banner still waving proudly there 19:00:48 (I'll find time to replace that later today) 19:01:12 jmadero1: you here? 19:01:22 have you had time to "update gardening page with new text for ping + table with date info" 19:02:43 * colonelqubit hmms 19:03:01 Is steve here? 19:03:49 nope 19:03:58 arnaud_versini: are you here still? :-) 19:04:11 * arnaud_versini is busy for now, sorry 19:04:17 no worries 19:05:13 #topic Android Testing 19:05:28 Very quickly, we're currently jumping into Android testing right now 19:06:07 colonelqubit: don't forget to send me the mail :) 19:06:33 A number of you are already involved here, and I'm definitely looking forward to seeing LibreOffice for Android testing happen on different devices, versions of Android, etc.. 19:06:58 sophi: yeps -- that's on the list after the meeting, minutes, sending stats for bugzilla to florian, and ...something else I forget 19:07:24 colonelqubit: QA charts for the AB meeting maybe ;) 19:07:29 maybe. 19:07:41 * colonelqubit wonders if he imagines sleep if it will make him less tired 19:08:10 colonelqubit: week-end is not so far ;) 19:08:35 sophi: yeah, but email flows in all weekend 19:08:52 colonelqubit: yes, I know :) 19:09:31 If others in QA have an Android device and would like to become more involved with our testing program, please talk to me, Sophie, or Joel, and we'll help you get started. 19:10:02 beluga_: any topics from you? 19:10:16 colonelqubit: not really 19:10:57 sophi, raal, jmadero1? 19:11:07 colonelqubit: nothing on my side 19:11:19 Okay, let's move to mjayfrancis' notes 19:11:32 #topic Regressions 19:11:55 In what will come as a shock to all of you, mjayfrancis wanted to talk about regressions! ;-) 19:12:30 He's currently working on assembling a bibisect repository containing all released versions of LibreOffice 19:12:31 News from asklibo: What happened to my toolbars? 19:13:30 This should help us to more quickly drill-down and find particular regressions, especially bugs that were introduced in the "blind spot" that exists between the start of the project and the 3.5-series builds 19:14:03 (where we don't yet have bibisect repo coverage) 19:15:00 mjayfrancis' repo will be built for GNU/Linux, and will be similar to repos that Norbert has built for Mac. Perhaps sometime soon we can try to have a similar repo available for Windows as well 19:17:23 The 2nd topic from mjayfrancis regarding regressions is "Regression Triage" 19:18:20 I know that he's talked about this a bit with several of you, but the short summary of the idea is that we'd like to triage all of our confirmed, open bugs to a point where we can sort them into categories 19:20:42 (1) A problem that predates LibreOffice "Inherited from OOo", (2) A regression from some working version of LibreOffice, and (3) What we're calling an 'implementationError' -- a bug that was introduced in the code at the same time that a new feature was added 19:20:52 For more details, see https://wiki.documentfoundation.org/QA/Bugzilla/Fields/Whiteboard/Advanced#Regression_Triage 19:21:57 Right now, we've got over 5,000 open, confirmed bugs that need to be triaged into one of these 3 categories: https://bugs.documentfoundation.org/buglist.cgi?cmdtype=dorem&remaction=run&namedcmd=Bugs%20which%20need%20regression%20triage&sharer_id=27686 19:23:10 Because we've got such a large number of bugs in this set, we're suggesting that people focus on the most recent bugs and incoming bugs; we'll deal with the backlog as time goes on 19:25:05 I've found that dealing with already confirmed bugs is faster than unconfirmeds :) 19:25:22 beluga_: for regression-triage? 19:26:06 colonelqubit: yeah just wanted to say that ~5000 bugs might seem daunting at first thought, but the pace of going through them is different than uncos 19:26:26 * colonelqubit nods 19:26:28 basing this on my exp. with going through Joel's pinged NEW bugs 19:26:41 I think it's really important for us to keep up with our UNCONFIRMED bugs as well 19:26:55 even though we're down to 300, we've still got a consistent pool of bugs that need attention 19:27:41 I think we should try to find better ways to visualize what's in that pot, so we can identify what deserves our attention the most 19:29:22 Matthew has dealt with some older uncos in the last couple of weeks 19:31:08 excellent 19:31:22 Okay, I think that's about it for mjf's notes 19:31:26 #topic Anything Else? 19:31:52 Anything else to bring up at the meeting? 19:32:44 colonelqubit: nothing from me 19:34:10 Okay, thank you guys for showing up! Big thanks to mjf for sending in some notes, and a big thanks to everyone who is participating in the regression triage or Android testing. 19:34:32 * sophi disconnect, enough for that long day :) bye all 19:34:37 * colonelqubit waves 19:36:03 We have our next meeting scheduled in 2 weeks (April 22nd), however I'm scheduled to hop on a plane a few hours after the meeting, so I might be in transit and unable to join. If we don't have a meeting then, we'll have one two weeks after, on May 6th 19:36:10 bye! 19:36:33 #endmeeting