18:55:00 #startmeeting 18:55:00 Meeting started Wed Mar 25 18:55:00 2015 UTC. The chair is colonelqubit. Plugin info at http://wiki.debian.org/MeetBot. 18:55:00 Useful Commands: #action #agreed #help #info #idea #link #topic. 18:55:36 First off, it's Document Freedom Day! 18:55:56 Happy Document Freedom Day all of you 18:55:57 Also, UNCONFIRMED is down to 287, which is great 18:57:15 If anyone out there is in the US and wants some printed materials, I'm happy to drop them in the mail for you -- feel free to run an event next week or next month. The materials aren't dated, and the message is always relevant :-) 18:57:35 (That is, DFD materials, not UNCONFIRMED materials... :P) 18:57:40 and it's great that I'm the no. 1 bug closer atm https://bugs.documentfoundation.org/page.cgi?id=weekly-bug-summary.html 18:57:52 Awesome! 18:58:25 that comes from going through Joel's pings 18:58:49 Yeah, we gained 50 open bugs during the last week, but hopefully we'll be able to shift that trend in the future 18:58:54 btw. it's a fast-paced activity and suited for newcomers 18:59:19 steps clarified before etc. 18:59:33 jphilipz: congrats on reporting the most bugs :P 18:59:40 News from asklibo: Text to table....how???  || Libre office online without downloading-please urgent!!! 18:59:47 colonelqubit: thanks :D 19:00:07 mostly enhancements :D 19:00:08 Sort of a QA-loves-you-and-hates-you all at the same time ;-) 19:00:15 but i put some bugs in there as well :D 19:00:24 Indeed, indeed. A good mix is always healthy 19:01:06 Well hopefully i'll start a compatibility run like after 4.3 was released and those will all be bugs :D 19:01:33 So big news today, of course: LibreOffice Online 19:01:40 News from tdfnew: [Bug 90233] SLIDESHOW Cropped PNG picture appears wrongly truncated in second monitor if picture's transparency is greater than zero  || [Bug 90234] SLIDESHOW Cropped PNG picture appears wrongly truncated in second monitor if picture's transparency is greater than zero  || [Bug 90235] It's a "sum" function bug? 19:01:42 yep beluga mentioned it 19:02:04 collabora is doing some nice work 19:02:34 LibreOffice Online - how this affect QA team? will be standard component? 19:02:34 Not much for QA to do at this point, but I'm sure that once some tools and builds are made available, we'll have some conversations about that work 19:03:15 raal: Based on my understanding of the project so far, that sounds about right 19:03:49 expect it to be several months I believe 19:03:54 before QA is needed at all 19:04:37 raal: Comparing that to the Android work, right now we have the 'LibreOffice Viewer' as a component, but once we have some editing support and further compatibility with additional formats, my goal is to change that to be 'Android UI' (or similar) 19:04:57 colonelqubit: not to a different product completely? 19:05:06 * jmadero supposes this conversation can wait a bit before acted upon 19:05:09 jmadero: well the core code is all the same 19:05:20 so really the difference is the UI on top 19:05:22 true - just wonder about user expectation 19:05:28 Ah yes, good point 19:05:32 it *seems* like a different product 19:05:35 from user perspective 19:05:43 you can use LibreOffice online without downloading LibreOffice 19:05:51 I suppose with Android version...that's less true 19:06:10 I think that it may help to maintain the strong brand if we talk about keeping things as similar as possible 19:06:11 anwho - I'm happy with w/e, as product/component are really just two peas in a pod 19:06:35 So we might start out with a 'LibreOffice Online' component, and then could rename later 'LibreOffice Web/Online UI' 19:07:05 Anyhow, it's great that we need to have these bikeshed-esque conversations :-) 19:07:24 Rather than the "I wonder if one day we might consider making an Android port?" 19:08:35 #topic Stats 19:08:40 Let's look at some stats 19:08:50 needAdvice healthy at 12 19:09:09 bibisectRequests down to 56 19:09:34 preBibisect at 50 19:09:41 (little change) 19:09:53 colonelqubit: on that note, the new ping specifically asks people to test for regressions if they have the time 19:09:59 so we might see a few more regressions marked 19:10:03 jmadero: that's great! 19:10:29 and on that note - gardening page needs updated - action item for..... 19:10:31 someone :-b 19:10:33 perhaps me 19:10:37 Whenever a bug seems like it could be a regression, it's helpful to try to ferret-out that information from the reporter, so we can have them marked 19:10:45 yup 19:11:06 jmadero: updated text for the ping, or? 19:11:11 * colonelqubit isn't sure what the Action item should be 19:11:17 updated text needs added to gardening page 19:11:24 I'm also going to add the table that I have that has dates of every ping I've done 19:11:38 colonelqubit: for my completed action item I present the literary achievement of two added sentences at the end: https://wiki.documentfoundation.org/User:Beluga 19:11:48 #action joel will update gardening page with new text for ping + table with date info 19:12:07 +1 19:12:31 * jmadero wishes there was an extension where action pushed a calendar event to my google calendar :) 19:12:35 Speaking of regressions: 4.4 has 72 (up) 19:12:57 4.3 has 89 (up) 19:13:02 jmadero: Hi 19:13:07 * jmadero waves 19:13:13 and 4.2 has 84 (up a little) 19:13:27 jmadero: Don't forget that Soufiane is a professional for french government :-) https://bugs.documentfoundation.org/show_bug.cgi?id=67712 19:13:29 bug 67712: LibreOffice-Calc major/highest ASSIGNED form controls and draw objects anchored to cell but changes position after reopening 19:13:31 arnaud_versini: just going through some stats from the week :-) 19:14:48 And finally for our stats, a huge jump up to 290 bibisected bugs 19:15:57 Big thanks to mjayfrancis and everyone else who has been helping our devs figure out exactly what/when introduced a regression! 19:17:04 #topic Pending Action Items 19:17:27 beluga_: You mentioned that you worked on your item, right? ;-) 19:17:52 colonelqubit: yep 19:18:03 could not think of anything more 19:18:24 Sounds great! 19:18:42 [DONE] ACTION: beluga_ will expand wiki page notes on recruiting to include mentoring (and 'how to talk to devs') 19:19:06 If anyone else has suggestions on how to improve beluga_'s notes... y'all know where to find him! 19:19:28 colonelqubit: I added the NEW ping - to be honest I think it needs cleaned/rearranged 19:19:31 but it's what I'm using right now 19:19:33 * colonelqubit nods 19:19:36 it's a bit wordy - not perfectly organized 19:19:46 steve-_-: ping 19:19:58 pong colonelqubit 19:20:15 steve-_-: Did you have a chance to work on media tests during the past couple of weeks? 19:21:13 sorry now. I started. got stuck due to wiki issues. then got stuck thinking how to make the wiki usable. beluga_ investigated mediawiki live editing or whatever that plugin was called. and then I was back to work with other things 19:21:18 now = no 19:21:42 up upcoming easter holidays I hopefully have more capacity 19:21:47 no worries, just checking in :-) 19:22:32 (Off-topic: My friend down the road gave me some fresh eggs that are pink, brown, light brown, and green... it's like they're pre-dyed for easter!) 19:23:14 On the topic of media testing, having the baseline is important as a reference tool. I just saw this thread on reddit: https://www.reddit.com/r/libreoffice/comments/3016i0/need_advice_on_an_unaccepted_format_of_video_in/ 19:23:56 pardon my ignorance what is "baseline" tool on osxß 19:23:58 ? 19:24:09 Unfortunately on Windows there's not much a choice for video.. 19:24:43 steve-_-: Not a baseline tool, but having a baseline established, as in having a set of reference test results against which we can compare changes to the code in the future 19:24:44 just wanted to say, that is an especially bad occasion to play with LO and it's broken video support 19:24:52 absolutely 19:25:00 we need to know what's going on to get a grip on things 19:25:31 * colonelqubit checks the list 19:25:54 Walrus flag is top-secret. I'll have more info soon ;-) 19:26:58 Re: Educational opportunities with RedHat, I heard back from my contacts, and I'll be working with them over the next couple of weeks 19:27:27 colonelqubit: Need one in france :-) 19:29:27 Remy de Causemaker, Professor at RIT teaching classes in FOSS, just joined RedHat as a Fedora Community Manager/Outreach person, which is great news for us and for FOSS collaboration with universities in general 19:29:45 News from asklibo: Conditioal Text not working in Office Writer 19:30:52 I spoke with him at LibrePlanet this past weekend, and he's just getting started there; I'll continue to stay in touch 19:31:04 arnaud_versini: need one what? a flag? 19:31:06 colonelqubit: cool news. also there's tons of work to be done in educational work / university outreach. they are all MS office invested... 19:31:42 Yep 19:31:44 so long as the message isn't "use it for free and we'll work for you" 19:31:49 which is a hard message not to spread 19:31:50 steve-_-: RH has some materials and I think is hoping to expand in the area of teaching info re: (Free and) Open Source software 19:32:08 That aligns very well with the work being done at OpenHatch 19:32:59 As some of these projects are being developed, we have an opportunity to be a guinea pig in these relationships being formed -- having students work on QA or documentation or etc... as a part of class participation/graded work 19:33:04 well universities are a great place to spread the idea of foss software free as in freedom not as in beer and the people there have the intelectual potential of grasping that concept... 19:33:25 steve-_-: yeah, at lower levels the message is harder 19:33:32 if we're trying to convince faculty/administration to change 19:33:46 they want "cheap" and we give them "free" - but that doesn't give them access to free bug fixes 19:33:54 so the message is hard to tailor correctly 19:34:10 steve-_-: Indeed. I think that's one of the key goals of these programs, and a smart investment from the RH side, as if you can get this message to students in CS at uni's across the US (and beyond), then you're going to have a much bigger pool of candidates coming out the other side interested in working on projects, using the software, etc.. 19:34:43 exactly. all students that hear MS Office in their introduction / first semester are basically lost 19:34:59 and those are the once going into industrie etc making important decisions in the future 19:35:00 jmadero: That's why an integrated cirriculum is a stronger position 19:35:15 here's one attempt at "tailoring the message" http://commonstransition.org/ 19:35:32 colonelqubit: yeah - and perhaps setting up a collective group that contributes jointly to bug fixes 19:35:39 steve-_-: I think that students are coming into college, high school, and even middle school with existing experience and biases 19:35:45 like a group pool of funds with one message - that they then can either tender or go directly to companies 19:35:51 I suppose many educational facilities would see the same bugs 19:36:01 If you have an IBM PC at home running WP 5.1, that's what's going to be comfortable and natural to you as a kid 19:36:05 pooling resources (so long as there is one voice at the end) is a good idea 19:36:33 yes but if you as a lecturer say, I want your work in a free file format 19:36:38 that is a very strong stance 19:36:47 from dave_largo's input, it seems like when they moved to LibreOffice, they just cut their budget to $0, and now are unwilling to give a dime 19:36:48 and should get people started to think about things like that 19:36:50 that's a problem 19:36:54 steve-_-: +1 19:37:23 most user s never ask themselves why they are stuck on doc or so 19:37:33 raal identified these as potential non-bugs https://bugs.documentfoundation.org/show_bug.cgi?id=89961 https://bugs.documentfoundation.org/show_bug.cgi?id=89997 19:37:35 bug 89997: LibreOffice-Calc normal/medium UNCONFIRMED Named ranges shift when distant cells are deleted 19:37:42 jmadero: Sure, and that's why I don't ignore the $0 price tag, but push that less as the ultimate selling point 19:37:43 steve-_-: I agree - my interview with DFD people basically said the same thing ;) 19:38:10 yes there's lots of work to do. but that#s the place free software has to be 19:38:17 #topic Anything Else? 19:38:22 and research results should be free as well 19:38:35 colonelqubit: any update about locking priority/severity? 19:38:37 Okay, I think we've covered all of the topics pending from last time 19:38:41 colonelqubit: can we look at a handful of bugs? like those I mentioned just now 19:38:54 sorry didn't mean to interrupt any ongoing meeting (: 19:38:58 jmadero: that is something that we can do 19:39:07 I'll look into that more after the meeting is over 19:39:10 so (1) is it something we want to do; (2) how hard to implement 19:39:14 steve-_-: you don't have to suggest topics 1 week beforehand :) 19:39:16 colonelqubit: I gave you those links right? 19:39:17 * colonelqubit thinks there might have been some small hurdle 19:39:22 jmadero: which links? 19:39:36 where bugzilla people on their ML told me how to implement it 19:39:37 if they're on a redmine bug report, that's probably easiest for me to keep track of 19:39:39 with the documentation 19:39:45 ah, excellent 19:40:03 jmadero: I think we can just disallow changing that field except to a subset of bugzilla users 19:40:19 So far, we haven't gone through and made subgroups aside from administrators 19:40:35 So the first conversation is: What groups do we want/need to have? 19:40:49 is cloph back from holidays? 19:40:57 colonelqubit: http://www.bugzilla.org/docs/4.4/en/html/cust-change-permissions.html 19:41:09 yes, and also restarted the mac tinderbox a couple of times :-) 19:41:17 colonelqubit: I think at least to begin with we just need like "Admins" "contributors" and "users" 19:41:24 contributors have rights to alter priority/severity 19:41:33 ah ok, just saw something happened today and the newest folder is empty. ignore me and move on then... 19:41:56 and getting contributor perms can be ultra low bar - basically request it, and you shall receive 19:42:10 Yep, I like it 19:42:18 colonelqubit: another thing is changing default priority/severity 19:42:23 jmadero: if we had a Dev role, could we filter them in queries? 19:42:30 IMHO that should move to something like - - - - 19:42:39 beluga_: not that I know of 19:42:47 queries are limited in Bugzilla :( 19:43:13 jmadero: well I do think there may be some benefit of having the reporter be able to participate in that process 19:43:18 jmadero: would be useful when you do a ping.. to not ping reports created by core devs 19:43:29 but you're just suggesting that they have to make a conscious effort to do so? 19:43:35 beluga_: hmm not sure - a lot of times devs close their own bugs after a ping 19:43:37 steve-_-: still uploading righ now - 15 min ETA for the main installer 19:43:39 because they forgot about them 19:43:54 colonelqubit: well I don't think we can have it both ways to be honest 19:43:54 ok :) I just noticed having to badger them on IRC 19:44:04 awesome cloph thx! good to have you back. hopefully you had a nice and relaxing time. 19:44:09 colonelqubit: if we lock it out, and we want uniform priority/severity 19:44:14 users who know nothing, can't be involved with the process 19:44:29 if we keep the contributor threshold very low - then a reporter who reports often....should ask for contributor perms 19:44:54 colonelqubit: and I think that the BSA2.0 will allow users to be involved - if you report through BSA2.0 - and it asks you a series of objective questions, it prioritizes it automagically 19:44:59 got away from work-related stuff at least - relaxing - not so much (moving and stuff :-)) but nice nevertheless - thx 19:45:37 jmadero: ah, okay 19:45:39 colonelqubit: from what I see - users either (1) don't prioritize at all because they have no idea what's going on; or (2) they falsely push their bugs too high 19:45:43 neither of which is useful 19:45:54 I thought the plan was that we (skilled labor) prioritizes 19:46:02 colonelqubit: yeah that's what I'm saying 19:46:05 everyone/reporter (etc) can makr severity 19:46:10 but currently the normal - medium is hard because that is a priority 19:46:17 so it's hard to see what has been skillfully prioritized 19:46:24 vs. what is just sitting in default status 19:46:38 colonelqubit: oh - no sorry, I think priority and severity should be left to QA 19:46:44 as both have important ramifications 19:46:47 Okay, that's fine 19:46:58 personally, I think having two fields is kind of overkill for how we use them 19:46:58 unless others disagree :) 19:47:14 colonelqubit: that's another bird that we might want to kill at some point :) 19:47:16 (enhancement could just be a flag, I feel) 19:47:19 but right now - I think it's rough 19:47:30 I rarely touch importance, severity quite often 19:47:36 I mean I think that a normal bug means a bug that prevents high quality work - if it's set to high - it means it prevents it for a lot of people 19:48:15 ultimately I think as a stepping stone, defaulting to some unprioritized state and locking them out 19:48:32 It's just that QA and devs aren't strictly integrated: Basically unless it's a crasher or MAB, my impression is that the devs work on whatever region of the code to which they've been assigned 19:48:36 means QA (1) knows what is not prioritized; (2) hopefully takes more time to prioritize things; (3) we can abandon MAB at some point 19:48:46 colonelqubit: devs agree that MAB sucks 19:48:55 if we can provide them with objective priority/severity consistently 19:48:57 they are happy to abandon it 19:48:58 indeed 19:49:16 Okay, so I'll take a look at that process. jmadero: Send me those notes, or a reminder ping 19:49:22 colonelqubit: +1 19:49:41 colonelqubit: so action item for you is to look into that and another is to walk me through Android install :) 19:49:46 so I can start breaking things 19:49:50 #action Robinson will create 'contributor' group for bugzilla to hold editing perms for priority AND severity 19:50:05 colonelqubit: along with the group probably a new wiki 19:50:12 with instructions on how to get perms (email QA list?) 19:50:19 "request for bugzilla permissions" 19:50:24 #action Robinson will restrict access to priority/severity fields to 'contributor' group 19:50:51 colonelqubit: or maybe a link on the bugs themselves right next to the locked fields "request permissions" or some such thing 19:50:54 #action Robinson/Joel will document on the wiki how all this priority/severity/contributor group stuff is supposed to work 19:50:57 that is just a mailto link 19:51:08 I dunno - just brainstorming 19:51:23 basically very low bar - but not no bar at all ;) 19:51:37 #action Robinson will something something...Android install for Joel 19:51:51 +1 19:51:53 good verbage 19:51:54 I think that's it. 19:51:56 beluga_: you're up! 19:52:17 yep like I mentioned, raal identified these as potential non-bugs https://bugs.documentfoundation.org/show_bug.cgi?id=89961 https://bugs.documentfoundation.org/show_bug.cgi?id=89997 19:52:19 bug 89997: LibreOffice-Calc normal/medium UNCONFIRMED Named ranges shift when distant cells are deleted 19:54:19 beluga_: Should we just check these, or? 19:54:21 * colonelqubit reads 19:55:12 colonelqubit: yeah I pinged erAck about those some time ago, but I guess he was busy (or bad connection) 19:55:52 is moggi about? :-) 19:56:03 (hmm... 11 hours idle doesn't bode well...) 19:56:38 beluga_: Leave those two to me. I'll chat with the devs 19:56:44 ok thanks 19:56:46 colonelqubit: you're the man :) 19:57:02 then some report hijacking attempts 19:57:05 #89941 19:57:07 LibreOffice-Writer normal/medium UNCONFIRMED No hyphenation in user fields - words also break at end of line https://bugs.documentfoundation.org/show_bug.cgi?id=89941 19:57:10 #action Robinson will wrangle https://bugs.documentfoundation.org/show_bug.cgi?id=89961 and https://bugs.documentfoundation.org/show_bug.cgi?id=89997 19:57:11 bug 89997: LibreOffice-Calc normal/medium UNCONFIRMED Named ranges shift when distant cells are deleted 19:57:24 #88302 19:57:25 LibreOffice-Writer normal/medium UNCONFIRMED Embedding graphics slows down scrolling massively https://bugs.documentfoundation.org/show_bug.cgi?id=88302 19:57:49 beluga_: hijack? 19:58:07 colonelqubit: yep, users appearing out of nowhere with a different issue 19:58:12 hijacking the report 19:58:25 split function would be needed or delete 19:58:41 88302 might not be different, but we have no way of knowing, as the original reporter did not attach a file per your request! 19:58:47 I dunno 88302 seems maybe good? 19:58:49 * colonelqubit nods 19:59:00 * colonelqubit sighs 19:59:07 Roger in 89941 doesn't mention or show hyphens at all 19:59:15 (or lack of that is) 19:59:22 This is why I'd like bug trackers to keep track of individual test runs (user + software + hardware + time) as tuples/nodes 19:59:36 so that if someone gloms-on to bug #1, we can just peel that off and make it bug #2 19:59:52 News from asklibo: numeración de documento completo 19:59:55 or if bug #4 and bug #5 are dupes, then we can just put the repro steps/results side by side in the same report 20:00:19 buuuuut, I don't have time to re-code bugzilla that much right now :P 20:00:40 beluga_: with 88302, I'd say that if the example given is slow, then mark it new and proceeed 20:00:49 colonelqubit: ok i'll do it 20:00:50 (whiteboard: 'perf' and all that) 20:01:21 If it's fast, I'd say something like "Per the provided document, I don't see a problem" and mark it WFM. 20:01:31 it was slow 20:01:52 We have so many bugs and so many bug reports, it's important not to get mired-down by one or two of them. 20:02:20 (bringing them to the QA meeting is a good strategy :-) 20:04:18 then there's this one where I commented https://bugs.documentfoundation.org/show_bug.cgi?id=90086 the document was apparently originally a .docx as confessed here: https://bugs.documentfoundation.org/show_bug.cgi?id=90087 20:04:20 bug 90087: LibreOffice-Writer normal/medium RESOLVED NOTABUG Impossible to remove a Frame 20:09:13 beluga_: re 89941, I'd just ask the 2nd commenter to clarify why he thinks he's seeing the same problem (and politely mention that if it's different, he should please file a new bug) 20:09:35 colonelqubit: ok I'll do it 20:10:05 I'll take a look at the next pair of bugs in just a second, but I wanted to see if anyone else had stuff to discuss... 20:10:12 Anyone? Topics? 20:11:27 colonelqubit: letter kerning on osx in writer? 20:11:31 you got this on your radar? 20:11:36 bug #? 20:12:51 http://imgur.com/YrpcWjb let me look up the bug # 20:14:29 https://bugs.documentfoundation.org/show_bug.cgi?id=87373 20:14:31 bug 87373: LibreOffice-graphics stack major/high NEW Kerning broken in 4.4b2 (OS X only) 20:16:03 beluga_: If the document was originally a DOCX, it would be helpful to know if it was converted, and what converted it. As q and NEEDINFO? (recent update, so likely to hear back) 20:16:45 colonelqubit: it was simply saved in LibO 20:16:46 Also, please ask to upload file directly. That way avoids further possible modification by google drive, etc. Also puts the document someplace it won't expire, as well as adding it to our set of automated-test documents. 20:17:14 colonelqubit: comment 2 in 90087 "Yes, from LO Writer 4.4.1.2" The document was too big 20:17:34 25 megs 20:17:40 Ah, okay. 20:17:45 * colonelqubit ponders file size limits 20:18:42 Okay, any other topics for the meeting? 20:19:06 Then we'll end here. Thanks for attending! 20:19:31 Our next meeting will be April 8th, in 2 weeks. Go out and enjoy the rest of Document Freedom Day! 20:19:38 #endmeeting