# QA/Meetings/2014/July 16

< QA‎ | Meetings‎ | 2014
 Meeting of the QA Team previous - next Date and Time (UTC) 2014-07-16 17:30 UTC G+ Hangout http://bit.ly/LibO_QA_call_hangout Chair Robinson Tryon Secretary Robinson Tryon Started 17:44 Adjourned 19:17 Participants Joel (jmadero), Robinson (colonelqubit), Jay (jphilipz), Joren (jorendc1), Florian R. (Reisi007) Pad http://pad.documentfoundation.org/p/qa Talkyoo Room Number 537138

# Agenda + Minutes

Minutes:

Content from the meeting itself ("the minutes") will be displayed in blue boxes like this.

## Before we Start

A few minutes before we officially start the call, we'll try to have a few people on hand to help assist newbies set up their teleconferencing gear, get things muted, and log on to all of our communication channels:

## Start

Put start time and attendees list in the table

### IRC Minutes

EXPERIENCING PROBLEMS WITH GOOGLE HANGOUT TODAY ---> Falling back to IRC chat (No Talkyoo meeting today)

jorendc: then probably wiki text hates you
jmadero: really really really hate it
jorendc:
jorendc: how come?

• colonelqubit laughs

jmadero: jphilips: if you can't see the problem with 4.3, I quit on bibisecting it
jmadero: because I always screw it up
jmadero: I butchered it this time
jmadero: and now don't know how to fix it
jorendc: lol

• colonelqubit had trouble finding german beer here

jorendc: eating now, will join after that

• jorendc lasagna time o/

colonelqubit: just budweiser for me
jorendc: Is wine okay too

• jorendc takes a sip of great German White wine

jmadero: blah - [beep]ing world cup
jphilips: think i need to restart the net as nothing in google in working
jorendc: jmadero: you mind inviting me to the hangout ?
Reisi007: jmadero: ouldn't start the video call because of an error.
Reisi007: Please try again in a few moments.
jphilipz: jmadero: about that bug you asked about, just open the file in 4.1 and then 4.3 and you should easily see the problem
jmadero: jphilipz: I can't see the problem in 4.3

colonelqubit: Like groundhogs
colonelqubit: Who's in the hangout?
Reisi007: jmadero: Couldn't start the video call because of an error.
Reisi007: Please try again in a few moments.
jmadero: setting up a new one

colonelqubit: Please Stand By -- Technical Difficulties
Reisi007: colonelqubit: ^^

• colonelqubit tells jorendc1 to hang out!

jmadero: we're 10 minutes behind schedule, let's get this started
colonelqubit: we'll always have paris....I mean IRC as a fallback
Reisi007: Liongold: Updated CSS right now
Reisi007: jmadero: Couldn't start the video call because of an error.
Reisi007: Please try again in a few moments.
jphilipz: crashed again
Reisi007: oh
Reisi007: same error
Reisi007: other service?
Reisi007: or do we type?
Reisi007: //we need our own service
Reisi007: ^^
jmadero: let's try again in 5 and then if that fails let's do IRC
Reisi007: jmadero: lets do 5 min IRC
jmadero: sure, Robinson can start in a moment

• Reisi007 is trying G+ in the background

colonelqubit:

------ We'll use IRC for now -------
'(start: 17:44 UTC )

jorendc1: +1
colonelqubit: wait, nevermind. we don't need etherpad for this
colonelqubit: Okay, who's in the meeting?

• jphilipz raises his hand
• jorendc1 raises his hand

colonelqubit: cool
colonelqubit: Okay

----> Project Management
colonelqubit: We now have a QA project in redmine
colonelqubit: does everyone have an account?

• colonelqubit looks arund

colonelqubit: around
jphilipz: i dont
jmadero: need to get comfortable with it
colonelqubit: http://redmine.documentfoundation.org/

• Reisi007 is in

jphilipz: whats the purpose of it
colonelqubit: jphilipz: we can use it to keep track of TODO items, etc..
jphilipz: okey dokey

• jphilipz wonders if he has anything TODO

jorendc1: colonelqubit: also action items of our calls?
jorendc1: I mean, all action items, or only the big ones?
colonelqubit: jorendc1: Yes, some ACTION items are good
jmadero: I'd say any that are lingering and big enough not to get done fast
colonelqubit: We can keep track of our open ACTION items by just querying redmine -- that could be handy
jmadero: like "Joren will email Joel to say Germany should have lost the World Cup"
jorendc1: colonelqubit: so we should use ACTION-RM or so?
Reisi007: @all: Nothing new about the dev guys who raised their hands
colonelqubit: what's the RM?
colonelqubit: oh, redmine
jorendc1: yes
colonelqubit: naw, I think we'd just put stuff in Redmine
jorendc1: so we know we have to create a RM issue for that
colonelqubit: if the # of todos in redmine filed against QA gets to big, we can start to think about a specific prefix to use
jorendc1: okay, and who's going to create the issues in redmine, to one who is assigned to the task during the call or is that you ?
colonelqubit: jorendc1: I guess I'd probably do it at the end of the call
jorendc1: great +1
jmadero: but I guess once the initial pain in the ass backlog is done

• colonelqubit nods

jmadero: I think new action items per call is <5 total
colonelqubit: Any other q's re: redmine?
jmadero: so how are we grouping them?
jmadero: like is every item a new ticket
jmadero: even if they are very similar
colonelqubit: I'm happy to use them and then throw them away
jorendc1: indeed, with a clear scope so we can close them 'as fast as possible'
jorendc1: we all know the endless bug reports
colonelqubit: jorendc1: feel free to open sub-items, if that helps to keep track of progress
jmadero: and then I don't think we have to discuss every action item
jmadero: like "whiteboard wiki is done" no need to talk about it unless we need additional feedback
colonelqubit: yep
jmadero: or if it's a bigger task and we want to celebrate
jmadero: should cut some time off call
jorendc1: +1

----> UNCONFIRMED bugs
colonelqubit: Down to 962 from 985
colonelqubit: Numbers are looking good
jmadero: right after this call a 60-120 minute blitz
jmadero: we'll ping dev list to see if any bored developers want to help
jorendc1: I'm in
colonelqubit: I was vaguely following conversation about our triaging methods/closing bugs
colonelqubit: what was the result of that?
jmadero: just complained that we close bugs when users report them shitty
jorendc1: ah yes, was a gentle rant as far I can see
jmadero: and he never replied back
colonelqubit: heh
jmadero: and I see no reason to change it
jorendc1: agreed
jmadero: and if that means we miss a nasty bug one of these days - you know that sucks
colonelqubit: Sounds good -- I just wanted to check in with y'all and make sure we're on the same page
jmadero: I believe strongly that users have some responsibility in this chaos as well
jorendc1: well, our working method did survived all his rant
jorendc1: so I guess our base is pretty solid
jmadero: even though we didn't plan that
jphilipz: got disconnect, bad connection today for some reason
colonelqubit: jphilipz: no worries
colonelqubit: what was his last quip about melting ice?
Reisi007: ice? like ice cream colonelqubit
Reisi007: hangout still not working
jmadero: his statement boiled down to if the instructions are a little manageable, QA should take the time and energy to reproduce every document from scratch
jmadero: I was irritated as well
colonelqubit: The comment was "Let some moderate heat melt your ice field."
jmadero: makes me thinks he was a bit loony
colonelqubit: I thought it was some pun, but over my head it went
colonelqubit: okay, moving on
jorendc1: lets not waste our time with his comment anymore. I think we proved with his 'gentle rant' our work method is solid

----> Regression watch
Reisi007: colonelqubit: one nasty bug fixed today

Regressions in 4.3 not in 4.2:
colonelqubit: up from 37 to 41
Reisi007: nice to have it in .0
IZBot: LibreOffice-Chart normal/medium RESOLVED FIXED Threaded 3D chart rendering ... tdf#81110
Reisi007: fdo#idk

• colonelqubit shrugs

Reisi007: fdo#81333
colonelqubit: neither do I, but good to hear that we've got a fix
IZBot: LibreOffice-Writer major/high RESOLVED FIXED Selecting table when comment present on page crashes Writer tdf#81333

Regressions in 4.2 not in 4.1:
colonelqubit: Regression count for 4.2 is up to 99 (from 85)
colonelqubit: 99 regressions, and we ain't fixed one
jmadero: but maybe mostly just good QA work
jmadero: I'm not convinced it's actually climbing
colonelqubit: hopefully yes
jmadero: how many marked as Major+
colonelqubit: numbers are 65, 85, 99
colonelqubit: so slowing down at least
jmadero: lol half full kind of gu
jphilipz: must be all my doing

Total open bibisected
colonelqubit: Total open bibisected bugs is up to 108 (from 95)
colonelqubit: I'll nudge the devs again
jorendc1: colonelqubit: great
jmadero: yeah - I think that we should prioritize those
jorendc1: I think michael S. is doing a great job here
jmadero: then at least devs can try to tackle the more serious ones
jorendc1: I think Michael S. is one of the biggest regression hunter
jmadero: yeah I've seen him very active
jorendc1: my kudo's to him
colonelqubit: Perhaps we should chat w/him about what else we can provide?

 ACTION: Joel will dig through bibisected and regressions confirmed bugs and prioritize according to flowchart


colonelqubit: good, good

colonelqubit: NeedAdvice: down to 29 (from 32)
colonelqubit: I'll continue to chip away at that stack
jmadero: for now let's just skip hard ones
jmadero: let that number get manageable
jmadero: maybe once it's under 15 or something
jmadero: at least let them pretend like they are catching up
colonelqubit: I try to take a handful to them when things aren't too busy in the ESC call
colonelqubit: but most of the bugs that make it that far just make them groan or sigh
jmadero: if a dev can't understand it
colonelqubit: Many of them are know/expected problems
jmadero: tell them to be more aggressive with NEEDINFO
colonelqubit: *known
jmadero: and if that's the case, mark them as NEW
jmadero: just get them out of Unconfirmed
colonelqubit: Sure
colonelqubit: I think the problem is that they often exercise some weird aspect of the code, and while what they do might not be very common, there is debate about whether the bug is kosher or not
colonelqubit: But I agree: Good to either get them in NEEDINFO and get feedback from user, in NEW if we know it's a problem we want to tackle (eventually), or just have them ping another dev if it's unclear if it's actually a bug or if it's our bug
jmadero: yeah, hopefully if we stop adding, we can at least get like -3 a week or something
colonelqubit: jmadero: I'm okay with the current #s -- I'll try to get it down closer to ~15, as you mentioned

----> LibreOffice Updater
jorendc1: colonelqubit: just want to mention: interesting comment of SteveBell on the update topic. tdf#68274#c5
IZBot: bug 68274: LibreOffice-Installation enhancement/high NEW provide better update mechanism (was: Use Sparkle update with Stable / Beta / Nightly channels to keep users on the latest releases on macOS)

• colonelqubit goes to read bugzilla

colonelqubit: jorendc1: "Kendy suggests using Mozilla's software updater"
jphilipz: sorry guys but i gotta run to the airport, was hoping the meeting would have been finished by this time
jorendc1: colonelqubit: not sure, but you were investigating options for a kind of branches, right?
jorendc1: (stable, master, ...)
colonelqubit: jorendc1: yes, that would be great
jorendc1: colonelqubit: so just want to mention that here
jmadero: see if they have thoughts
colonelqubit: I believe we brought up the idea of using Mozilla tooling for that, but I can't recall why it didn't go forward
colonelqubit: I think it's a great idea
Liongold: I'll do install it. About the update, maybe Opera got some info out, if you're interested.
jmadero: so I'm sure their stuff is not MPL
colonelqubit:

 ACTION: Bring up proposal to use Mozilla updater tooling for LibreOffice (Robinson)


Liongold: I know. But they wrote about some utility which helps about patching or something.
Liongold: If you want I'll find the post.
colonelqubit: Liongold: Got a link handy?
jmadero: sure, we've been searching for a better updating mechanism for some time
jmadero: I think everyone in QA thinks updating is a nightmare
colonelqubit: "a better" -> "an"
jmadero: sure but that requires futher steps
jmadero: which is fine, it's a workaround
jmadero: ideally it works within the software
jmadero: and you can select a branch
colonelqubit: +1
Liongold: jmadero: colnelqubit: They mention something called bsdiff
Liongold: http://blogs.opera.com/desktop/2014/04/opera-developer-22-now-easier-update-ever/
Liongold: Tell me if it's good when you have time to read it....
Liongold: They also mention UAc, no code but explanation.
colonelqubit: bsdiff is vaguely familiar
colonelqubit: I'll bring the topic to ESC and then we can go from there
Liongold: Ok, glad I can help.
colonelqubit: IF anyone would like to send me some notes/ideas before tomorrow, I'll read them over
colonelqubit: then we'll see what the devs say

----> Bugzilla Migration

colonelqubit: no news from my side. I've been wrapped up with OSCON and the Boston Hackfest lately
colonelqubit: I'll have some time in August to dedicate to the process
jmadero: it's going to get harder and harder for me to find time
colonelqubit: but we do need to coordinate between Infra, Norbert, Joel, me, etc..
colonelqubit: yeah, I know
colonelqubit: We'll make it happen
jmadero: sure, worst case we do it in Bern
Reisi007: jmadero: extracted it a bit more, so integration is easier
jmadero: colonelqubit: but in August, weekends are rough now, after August 15 my schedule becomes really rough for at least a month
colonelqubit: (at least the planning)
jmadero: yeah, I'm fine with that
colonelqubit: jmadero: okay, perhaps 2nd half of September might be sanest idea

• colonelqubit nods

jmadero: and ping those who need pinged a 100 times
jmadero: until we get a commitment

• jmadero prepares armor, shield and sword

colonelqubit: +1
jmadero: I bet FDO guys are irked
jmadero: we bugged them a bunch

----> Bibisect and Bibisect GUI
jmadero: no news as far as I know
Reisi007: colonelqubit: no news
colonelqubit: Reisi007: any news from you?

 ACTION: Joel will poke the new guy again on email that showed up with dev experience

• colonelqubit nods

jmadero: see if I can get him happy again
jorendc1: lol
colonelqubit: jmadero: sentence trying are write you?
Liongold: lol
jmadero: Captainmajid: any time for a bug squashing session after our meeting, tends to be a bit more entertaining when we go as a group
jmadero: plus more satisfying seeing the #'s drop fast

----> Hackfests
colonelqubit: The Boston Hackfest will be in a week and a half
colonelqubit: Planning a few activities, including some intro to QA: https://wiki.documentfoundation.org/Events/2014/US_Summer_Hackfest#Activities
colonelqubit: If any of you have suggestions, please include them there
jmadero: and if anyone is around that weekend
jmadero: please join QA channel - we'll have at least some online stuff going on
jorendc1: if I do have time, I'll surely jump in

• colonelqubit warns jorendc1 about breaking LibreOffice during the hackfest

Liongold: I'll try as well.... from Malta of course

• jorendc1 looks away
• jorendc1 knows from nothing

jorendc1: knows nothing* jeez
jmadero: it's okay, no one speaks English well in this chat anyways
colonelqubit: That 'innocent' routine ain't gonna work this time!

• colonelqubit bleeps and bloops

jmadero: jorendc1: soms ben je boos dat Engels de dominante taal in de wereld?

----> ( Frozen Items)
colonelqubit: Anyone want to discuss one of our Frozen/Long-term items?
colonelqubit: I thought it was the Sweedish Chef
jmadero: nothing on those frozen items here
jorendc1: I think the sentences the Sweedish Chef forms are better constructed then English->Dutch in Google Translate
colonelqubit: Ok, then let's move to the New Item

----> NEW ITEM: Demonstrate our Interoperability
jmadero: one is a brainstorming event, maybe we can actually discuss next call
colonelqubit: preach!
jmadero: so I'd love to somehow demonstrate our interoperability more clearly

• colonelqubit gets out his brain slicker for the brain storm

jmadero: so my idea is to somehow track lots of things - and provide test cases
jmadero: so for instance "basic text in one paragraph, nothing special"
jmadero: I see a long list of tests
jmadero: right now I think there is just a lot of FUD out there about where we are
jmadero: and honestly, we don't know where we are
colonelqubit: So you're talking about compatibility tests?
jmadero: yes, but stripped down to the most basic level
Reisi007: jmadero: LibO has unit tests
jmadero: in some very organized fashion, that we can keep adding to
Reisi007: automated
colonelqubit: Some kind of programmatic proof that we can properly handle a variety of documents in ODF, DOCX, etc.. ?
jmadero: I want something that many in QA can contribute to
Reisi007: jmadero: or the test script
Reisi007: which pulls from FDO
colonelqubit: Unit tests aren't all bad
jmadero: but I'd love to have like 5,000 really basic test cases
colonelqubit: with a little training, I think that Joel and others could start to write some unit tests
Reisi007: colonelqubit: agreed
colonelqubit: I think that we do need to think about the overall message, as Joel says
jmadero: so that's me, lol I want something that "many" can contribute to
jorendc1: indeed, I know on writer-side (ooxml etc) for every fix in the docx section, an unit test is added
jmadero: and we need to be transparent
colonelqubit: Indeed
jmadero: to really discuss where we stand as a project
jmadero: and have easily accessed samples for the public to see
colonelqubit: So we have a few different motivations/groups that use LibreOffice
jorendc1: but please don't forget it do consume CPU time during build
jmadero: "oh yes, clearly this works well"
jorendc1: so we might now want to 'randomly' test something
Reisi007: needed 1) a PC 2) libo 3) file + expected result in 2 folders
jmadero: or "shit, that particular thing is broken"
jorendc1: and test it 1000x due overlapping test cases
jmadero: LibreOffice and MSO will be required
Reisi007: another tool we need
jmadero: but that's my brainstorming idea
colonelqubit: +1
jorendc1: +1
jmadero: and I've suggested similar things in private behind the scenes
jmadero: others have agreed that we could use some clarity in this aspect
colonelqubit: I'll be seeing Italo at OSCON next week
jmadero: like we know image handling is "mangled"
colonelqubit: I'll try to find time to chat w/him about some of the aspects
jmadero: but who knows how bad it is without clean sample cases
colonelqubit: I think our biggest win is to promote it as an overall user experience
colonelqubit: that we're doing this so that our users can experience maximum interoperability with their peers, with

----> NEW ITEM: Whiteboard Naming Conventions
jmadero: okay now whiteboard, there are multiple things here
Reisi007: bibisect and bibisectRequest
Reisi007: +1
jmadero: is it just ooxml and older mso files
jmadero: or does it include like rtf, html
Reisi007: jmadero: compatibility with other formats
jmadero: is it "anything with any other format"
jmadero: I always thought it was just the major competitor
colonelqubit: okay, now go to town
jmadero: so thoughts on interopability - this would add to our clarity/transaparency
jmadero: if we start using this status more
jmadero: so is it every format not equal to odf
jmadero: or it is the major competitor filter
colonelqubit: I'm a big fan of KISS
colonelqubit: Keeping it simple
jmadero: and I'd love to have this finalized and done
Reisi007: jmadero: keep it short and simple
colonelqubit: Well I'd like to have consistency, so I'd push Filter:xxx more
colonelqubit: that being said, I believe that devs will grumble
jmadero: colonelqubit: so my thought is we'd have Filterocx, and then "interoperability" also
jmadero: but if it was Filter:HTML, then no interoperability
jmadero: so you could easily query every single interoperability issue
jmadero: and ignore those other not so popular formats
jmadero: basically my argument is we are mostly concerned with odf, then ooxml, then microsoft old format
jmadero: and lastly, those other formats
jmadero: so highlighting all issues with MSO with "interoperability" is nice
colonelqubit: So what do we use 'interoperability' for right now?
colonelqubit: i.e. are we consistent?
Reisi007: jmadero: colonelqubit : As I won't remember/ use them no vote from here
colonelqubit: Reisi007: you won't remember what?
jmadero: colonelqubit: we are not consistent
jmadero: that's why I brought it up
Reisi007: colonelqubit: the filerocx thing
jmadero: Jay is using it for everything
Reisi007: I love typos
Reisi007: ^^ or they love me
jmadero: or do we just want to "retire" interoperability
jmadero: and swing fully to "Filter:XXX"
colonelqubit: Standardize, standardize
colonelqubit: (is my suggestion)
colonelqubit: New QA folks and new users don't know what 'interoperability' means
jmadero: so every other filter status should be retired
colonelqubit: QA/Bugzilla/Fields/Whiteboard#interoperability <-- NOT intuitive
jmadero: and we should move 100% to Filter : XXX
jmadero: can we get a quick vote
colonelqubit: I really feel like the way that we use Bugzilla is bespoke and esoteric
Reisi007: pls lowercase
jmadero: Reisi007: let's deal with that next
jorendc1: +1
jmadero: for now just the filter isssue
Reisi007: filter:xyz
colonelqubit: For long-time QA and devs, there's just comfort with what we know, but that serves as a limitation to new participants
Reisi007: not Filter:XYZ
colonelqubit: and means that what info is tagged is less useful
jmadero: colonelqubit: sure but the reality is no one new is adding these
jmadero: and if experts can quickly go through 100 bugs and add "filter : ..."
jmadero: Jay has been doing some good work here
colonelqubit: oh, i'm not suggesting that new people add them (at least not at first)
jmadero: and I know some devs like it
colonelqubit: I'm saying that new people can't *read* the current descriptors on bugs
jmadero: filter : docx seems fine?
colonelqubit: "Filter:XXX" is much more readable
colonelqubit: Is 'filter' the right prefix?
jmadero: yeah I was told that by devs
slacka: So I think I found a strange bug/behavior in calc
colonelqubit: I mean, it makes sense w.r.t. how our code is formatted
jmadero: slacka: give us a few
jmadero: yeah I think it's obvious
jmadero: they know what it means
colonelqubit: (side note: We did have...might STILL have... a couple of different filters for one format, but I think that's mostly going away)
colonelqubit: And I think that was just export filters
jmadero: I'm happy with the proposal
jmadero: and if we push it hard - I will
jmadero: and sooner or later appreciate it
jmadero: Reisi007 clearly has an opinion here
jmadero: I'm happy where it is
jorendc1: colonelqubit: no, we do have a different filter for SVG files if you do file>open or insert them as an image
jmadero: just wanted to figure out if we need/want to change it
jorendc1: colonelqubit: but it's a minority
colonelqubit: Yep, so the proposal is to use "[Ff]ilter:XXX" to describe bugs that relate to a given filter/format
Reisi007: jmadero: a second, switched device
jmadero: so .svg would be filter : svg
colonelqubit: (no spaces, just to be clear)
jorendc1: filtervg (without spaces I hope)

• colonelqubit same-fives jorendc1

Reisi007: " filtervg"
Reisi007: islandCase
colonelqubit: I'd like to see consistency from one word to two
colonelqubit: So yes, either "Camel" and "CamelCase" or "island" and "islandCase"
jmadero: right now we have been using CamelCase for some time
jmadero: to go back and change those will suck
jmadero: and devs will get email pings
Reisi007: jmadero: ok, it is just error,prone
jmadero: I don't see how it's worse than islandCase
colonelqubit: does that make sense?
jmadero: so right now we're neither of those
colonelqubit: I know, and it's another source of confusion
colonelqubit: take mediawiki -- it uses Capital and CapitalCamelCase
colonelqubit: (okay, so it doesn't enforce the latter part, but it does upper-case all first words in page names)
jmadero: so, what the solution, we accept what it is
colonelqubit: I'm a big fan of consistency
colonelqubit: is it confusing to have things like 'needAdvice' ?
colonelqubit: maybe that's fine
jmadero: lol I think it looks a little goofy, but not confusing
jmadero: if that's what we decide to use

• colonelqubit agrees

jorendc1: +1
Reisi007: Less error prone
Reisi007: If you write quick, you ignore the case
jmadero: and we can monitor errors I suppose easily
Reisi007: So you leave it small
jmadero: but that's no good at all, then we might as well just do islandcase
jmadero: which to me is not bueno
jmadero: people using whiteboard should be experienced enough to slow down a little
jorendc1: lol, lets not spend to much time discussion or we should start or end with a case lets vote for islandCase or CamelCase
jorendc1: as long as we are consistent in it
jorendc1: +1 for islandCase
jmadero: yes, agreed, let's just be done with this
jorendc1: (I hate and forget always to case the first letter)
jmadero: sure, okay fine, Robinson can we agree with this
jmadero: I'm fine with that, consistency is more important
colonelqubit: I looked at Mozilla's bugzilla, and they use their whiteboard for other stuff... no help there
colonelqubit: So jmadero what did you vote for?
jmadero: I changed it just to get consensus
Reisi007: Island case
colonelqubit: island, islandCase ?
jmadero: Joren and Reisi have feelings for islandCase
colonelqubit: That's fine with me
jorendc1: okay, islandCase it is
colonelqubit: we'll need to do some updating
jorendc1: Neeext
jmadero: and we can start squashing
jmadero: I did 2 more bibisects while we did that
jorendc1: I'm ending my build here. /me had to revert some commits that broke debug-build
colonelqubit:

 ACTION: Update tags in the whiteboard to use 'islandCase' (ALL)

 [Ed. note -- Wikipedia/Google don't seem to know about 'islandCase', so I chose 'wimpyCaps' as a descriptive term]


Liongold: I have to go. Bye.

• colonelqubit waves

jmadero: jorendc1: still willing to squash?
jmadero: might be just you and I
colonelqubit: We should email out to QA and Dev list when we start to update the whiteboard tags, just so that everyone's on board
colonelqubit: and knows that we're changing things

 Colonelqubit --> Will do that nowish


jorendc1: jmadero: I'm doing writer. also in reverse chrone
colonelqubit: I think that devs will be most happy with that
jorendc1: jmadero: but I hope my build is done in 5 minutes
colonelqubit: Okay, what's next?
jorendc1: colonelqubit: +1 that
colonelqubit: yep
colonelqubit: that means that things like 'perf' can be left as-is

• colonelqubit sighs

colonelqubit: don't even get me started with keywords
colonelqubit:
colonelqubit: Okay folks, thanks for the meeting
jorendc1: jmadero: lets discuss for 20 minutes it is biBisect or bibisect (abreviation of binary bisect
jorendc1:

• jorendc1 runs

jorendc1:
colonelqubit: I'll get these minutes formatted and out, and write something about islandCase for QA and Dev lists
jmadero: he's not here to speak up
Reisi007: Filter:rtf
jmadero: if it affects more than just docx
Reisi007: BUT could we getbthat from attached test doc
jmadero: oh geeze, no no meeting is done

colonelqubit: So in my mind the quick test for ooxml would be if the problem affects more than one type of format
colonelqubit: So unless it's explicitly known at the time of bug filing, we'd just see 'filterocx'
colonelqubit: and then later, if we determine it's more pervasive, then we can add 'filter : ooxml'
jorendc1: jmadero: that'd be filteroc I guess?
colonelqubit: what names are actually used in the filters?
jmadero: if it affects doc, xls
jorendc1: because mso97 = doc/xls/...
colonelqubit: we have that filter list somewhere..
jorendc1: jmadero: grouping them = WW8
jmadero: if it affects basically all of the formats
jmadero: no a dev replied to Jay saying that
jmadero: he said that's just word
jorendc1: ah yes, WinWord8
colonelqubit: https://cgit.freedesktop.org/libreoffice/core/tree/filter/source/config/fragments/filters
jorendc1: grrr, I think my build is stuck on sd_html_export_tests
jmadero: colonelqubit: I don't see anything relevant
jmadero: for "all filters from MSO 97"
colonelqubit: If the plan is to say 'filter : blah', then that really is talking about our import or export filter
colonelqubit: vs. 'format : blah'
colonelqubit: does that make sense?
jmadero: ...I just listened to devs
jmadero: they said the extension/filter/format whatever should be called "filter:"
colonelqubit: Devil's advocate: What if there's a potential bug in MSO's ODF importer?
colonelqubit: or how they handle docx?

• jorendc1 hits colonelqubit in the eye

jmadero: let devs deal with that
colonelqubit: do we still label it 'Filter : docx' ?
jmadero: we're just identifying the stuff after the period
jmadero: if the bug is demonstrated in .docx
colonelqubit: I guess either way, we're going to have deal with it ourselves -- no guarantee that they're going to patch their stuff for us/compatibility
jmadero: this is meant to just be a simple thing to add so that we can query

• colonelqubit totally agrees

jmadero: ideally if every bug was marked
jmadero: devs could have a hackfest where they say
colonelqubit: Okay, well I guess we'll punt on the question of Word97 vs. Word2003 binary formats
jmadero: let's make docx filters better
jmadero: and just query every bug with that filter
jmadero: it's not perfect, but it's a matter of time spent
colonelqubit: We can just use 'filter : doc' for now
jmadero: the harder we make it
jmadero: the less likely anyone will do it
colonelqubit: heeeee's back
jphilipz: i'm back, i'm assuming its still going on right?

• colonelqubit waves the flag

jphilipz: anything i miss
jmadero: everything is moving to filter:yxyx
jphilipz: roger
jmadero: if it affects all of MSO, either ooxml, or mso97
colonelqubit: DONE with meeting: 19:17 UTC
jmadero: oh yeah no more CamelCase
jphilipz: what is mso97
jmadero: affects all those old formats
jphilipz: okey dokey
colonelqubit: jphilipz: I'm going to email out to the QA and Dev lists about the changes
jmadero: and yes we know there is a difference between MSO 2000 and 1997
jmadero: but, to make it easy
jmadero: this is better than we've been in the past
colonelqubit: okay, food time for me
jmadero: if we mark every bug
jphilipz: there is also mso xp and 2003
jmadero: devs will have a much better idea of bugs
colonelqubit: thanks all -- have fun squashing bugs!
jmadero: if it affects all of them

Meeting ENDS

## The State of QA

Including any Opening Discussion

• xx- Bugzilla migration
• Re: need to reset passwords
• Tweak the change-password page and the front page to indicate that people will need to reset their passwords
• Create an 'About our new Bugzilla' video as a part of our 'Intro to QA' series
• Bugzilla:
• SUGGESTION: Use bugs.documentfoundation.org/ to host Bugzilla
• We can have both LibreOffice and some of the Document Liberation Project libraries all exist as top-level products in the bugtracker
• QUESTION: Should we refocus ourselves as the TDF QA Team (to cover LibreOffice, DLP, etc.. ?)

### weekly status

* https://bugs.documentfoundation.org/page.cgi?id=weekly-bug-summary.html


### UNCONFIRMED Bugs

• July 2nd: 985
• Today: 962
• Difference: -23
• ACTION: Joel will come up with strategies (bloggers? other marketing?) to get us under 500 UNCONFIRMED over the summer (by the end of August)

### Regression watch (Bjoern)

* Regressions in 4.3 not in 4.2

• Previous Counts: 32, 37, 41
* Regressions in 4.2 not in 4.1

• Previous Counts: 65, 85, 99

### Topics for ESC?

Does anyone have any topics/concerns that I should bring to the Engineering Steering Committee (ESC)? (Robinson)

• Previous counts: 80+, 32, 32, 29

## Pending Items

### PENDING ITEM: Auto-updates for regular users and QA Testing

• ACTION: Investigate auto-updates for regular users and/or for QA/testing purposes (Robinson)
• ACTION: Open discussion again re:auto-update for QA on the mailing list (Robinson)

### PENDING ITEM: Bugzilla Migration etc.

• ACTION: Organize time to stress-test new Bugzilla VM (Joel)
• ACTION: Make sure to suppress outgoing email during our testing of the test Bugzilla (Robinson)
• ACTION: Resolve url for hosting our Bugzilla instance (Joel)
• Deal with acronym if using tdf.org url (tdf# ?)
• ACTION: Set up call and/or thread(s) with the useful people on a mailing list (Robinson)
• May 22 - Started process
• June - Need to follow-up again with key people re: current progress (Robinson)

### PENDING ITEM: Bibisect Repositories

• ACTION: Chat w/Bjoern, etc... re: updates to bibisect repos (Robinson)
• Propose docs on testing w/3.3.0 separately to place bugs in pre-LO category (Robinson)
• ACTION: Provide documentation for mac bibisect on wiki (Norbert)
• Perhaps ask Cloph for help re: Bibisect stuff, and keep Norbert's cycles free for Bugzilla migration

To review:

• IDEA: Add bibisect into regular Developer workflow
• IDEA: Create graph/tool to show which builds contain a given commit

### PENDING ITEM: GUI for Bibisect

• ACTION: Review BibiGUI -- any Windows/C# Python volunteer dev (Bjoern)
• June 19 - Joel may have found a volunteer?
• ACTION: Create an EasyHack a bug for the process of rewriting the existing C# code in Python (or Java) (Florian R.)

Related:

• ACTION: SI-GUI: Contact the people running "interesting" Windows tinderboxes and ask for their assistance (Florian R.)

### PENDING ITEM: Hackfests

• ACTION: Check on status of LibreOffice building in the cloud for hackfests (Robinson)
• Included in the notes for hackfest planning

Update on:

## Long-term/Frozen Pending Items

These are long-term or frozen items (i.e. on ice until something else happens) that we don't want to forget about...

### PENDING ITEM: MozTrap Localization

• ACTION: MozTrap translation (GUI and tests) continues to be high-priority for QA (Needs someone to shepherd)

To review:

• Other project who might want to work on translating GUI w/us?

### PENDING ITEM: Feedback pages

• ACTION: Sophie will shepherd native-lang versions of Feedback page

To review:

• Next steps for NL versions (bug #?)

### PENDING ITEM: US Merch Store

• (Future progress)

### PENDING ITEM: Most Annoying Bug List

• ACTION: Joel will propose stricter guidelines for what qualifies as a MAB

To review:

• How is the test going? (using the 'Priority' field in Bugzilla)

References:

### PENDING ITEM: Talk to CS Department/others at University

• ACTION: Put up a message on the message board at his University and see who is interested (Joel)
• Need to meet w/Charlie R. to talk with his Uni contacts as well

## New Action Items

All items proposed between meetings go here

## Announcements

• ANNOUNCEMENT: Our next meeting will take place... July 30th (Wednesday) at 17:30 UTC, unless otherwise noted on the QA/Mailing List.

## End

Put meeting-adjourned time in the table

# Topics

(Add topics below and reference them as #Example Topic in the Agenda/Minutes above)