18:00:58 #startmeeting tor-browser 18:00:58 Meeting started Mon Apr 25 18:00:58 2016 UTC. The chair is GeKo. Information about MeetBot at http://wiki.debian.org/MeetBot. 18:00:58 Useful Commands: #action #agreed #help #info #idea #link #topic. 18:01:06 hello everyone! 18:01:15 hi! 18:01:18 and welcome to a new tor-browser meeting 18:01:27 who wants to go first today? 18:01:42 Yawning: yeah, that'll work fine. 18:03:10 it seems this falls back to me again, fine. 18:03:40 alright, last week i was busy preparing the releases and addressing things found in the network review 18:03:42 (k) 18:04:07 i am quite happy with the 6.0a5 outcome, we basically got all of the things in we wanted, somehow at least 18:04:14 thanks everybody 18:04:41 Thanks to you for pulling it all together! 18:05:09 this week i'll help getting the releases out, then i plan to look at all the firefox bugs that did not make it into the dev docs 18:05:12 thanks 18:05:24 yeah, thanks for spending your weekend on it, GeKo! 18:05:39 and i reserve some time for things that explode 18:06:04 boklm helped here, too, fwiw 18:06:14 so, i am glad, we made it :) 18:06:39 today i looked at the remaining esr tickets, i guess we can talk later about how to proceed 18:06:47 that's it for me for now 18:07:02 * mcs will go next 18:07:07 Last week, Kathy and I created fixes for #18800 and #18603. 18:07:17 We reviewed many, many ESR45 fixes. 18:07:24 We also spent some time on #18619 but did not find the root cause of the browser console messages. 18:07:32 This week we will work on other tickets that have keyword=ff45-esr. 18:07:42 (I guess we will talk later about how to divide up the tickets) 18:07:48 That’s all for now. 18:08:03 * arthuredelstein can go now 18:08:09 Last week I worked on patches for #16326, #18741, #16998, and #16673. 18:08:22 This week I will work on a better patch for #16998, reviewing our patch for #15640, and trying to fix #18743. 18:08:41 If I have more time I'll try to work on more ff45-esr patches. 18:08:59 That's it for me 18:10:17 * boklm will go next 18:10:34 This past week I fixed #17895, reproduced builds of the new releases, and started looking at #18887 18:10:43 This week I planning to fix #18887 to get the test suite working with ESR45, and help getting the releases out 18:10:53 That's it for me. 18:13:48 hm. releasing without running the tests what could go wrong :) 18:14:09 but i guess we don't have good alternative options 18:15:30 mcs: a second look at the dev doc things in #18545 would be helpful in case i overlooked something 18:16:03 GeKo: I forgot about that. Kathy and I will take a look. 18:16:12 maybe putting that on the agenda for this week because if i messed up we would still have time to fix stuff 18:16:18 thanks 18:16:28 who else is herer for a status update? 18:16:46 *here 18:17:33 * GeKo suspects mikeperry is here but probably being shy 18:18:51 anyway, i heard the network code review is done and there are two remaining things we need to look closer at 18:18:52 thanks mike 18:19:00 i plan to do that this week 18:19:22 okay, discussion time 18:19:36 first item on my list: reminder of deadlines 18:19:50 i forgot to mention that last week which is my bad 18:20:13 but we had this discussion in the past: deadline for patches should be thursday the week before the release 18:20:31 given that we usually need at least one who is reviewing the code 18:20:54 sometimes two reviewers are required in case the code is complicated c++ 18:21:06 there are exceptions to this rule 18:21:41 but generally aiming for monday or the weekend for patches is not a good idea :) 18:21:53 makes sense to me :) 18:22:08 i try to remember announcing things the next time, though :) 18:22:23 second item: when do we want to get the stable out? 18:22:29 I am here, but distracted 18:22:36 aha! welcome! 18:22:39 is there something needed from me? 18:22:54 i think we are fine, i just summarized your status update 18:23:09 (the network review being done) 18:23:40 oh yeah. yep. that's finally over :) 18:24:38 i am trying to remember what we did last time. i guess it was a soft launch one or two weeks before the EOL of esr31 18:24:58 that sounds again like a good idea to me 18:25:32 So “soft launch” means available for use but we do not enable updates? 18:25:44 Seems like a good thing to do again. 18:27:17 based on esr 45.1 ? 18:27:31 yes, based on 45.1 18:28:05 i wonder about the update thing, though. i guess it would not be a soft launch if we enabled updates 18:28:39 even if we would not urge user to update given that this is no security related update 18:30:39 so looking at the timeframe week 21 May 25-29 seems to okay? 18:31:10 i guess we want to have some time for fixing urgent things we did not catch with our tests/the alpha 18:31:12 “soft launch” could mean we update but also tell people that it is OK to continue to use 5.5.x (and they could downgrade if 6.0 is really bad for them) 18:31:43 yeah, i might be more inclined to that model 18:32:30 but the release date should be unaffected by the update/no update question 18:32:41 Also note that May 30th is Memorial Day in the US (aka summer kickoff for a lot of people) 18:33:05 (Kathy and I will be afk that weekend including Monday the 30th) 18:33:09 How do we tell people these things (whether or not to update)? Is it something we can do in the UI? 18:34:42 mcs: ok. does that speak for releasing before it or after it? 18:34:55 There is a way to mark an update as not critical / not a security update but we have never done that (I don’t remember exactly how that changes the behavior of the UI). 18:35:29 When is ESR38 EOL? One week after that? 18:35:48 June 7 18:36:24 If we can get everything done in time, earlier release might be smarter. I don’t know. 18:37:11 let's aim for May 23-27 18:37:42 So patches need to be done by May 19. 18:38:12 well we are flexible here given that we don't have a fixed release date on tuesday :) 18:38:41 True. 18:39:46 but having all patches done by May 22 leaving possible reviews and release preparations for that week 18:39:49 sounds like a good plan to me 18:40:32 me too 18:40:35 yes, sounds like a good plan 18:41:28 Let’s try to get the tricky patches done earlier ;) Kathy says “earlier is better.” 18:41:45 yes, agreed. looking at the tickets tagged with ff45-esr that should be doable 18:42:14 this week we should get all the mozilla bug reviews done 18:43:37 One way to get only some people updating (a soft launch) would be to not show the Update warning (and big black arrow) on the about:tor page. But the "Check for tor update" menu option would still result in an update. But I guess that requires a modification to torbutton? 18:43:51 mcs: #18770 sounds like it might take some time as well 18:44:30 arthuredelstein: no, we only need to add the new version to the RecommendedTBBVErsions file but not remove the old, i.e. still current one 18:44:42 that's was actually my plan 18:45:26 GeKo: Ah, ok - so that results in not showing the update warning on about:tor? 18:45:49 yes. it is only shown if the version you are running is not in that file 18:45:54 which torbutton fetches 18:46:04 great 18:46:25 arthuredelstein: #18811 might be one candidate for getting done earlier as well 18:47:05 good suggestion. I'll aim to work on that soon. 18:47:45 apart from that nothing is actually jumping out for a early as possible treatment 18:48:03 but i guess looking at it again next week after the update might be a good idea 18:49:00 that were my discussion items. do we have something else/additional comments? 18:49:46 I would say anyone who starts to work on a ticket should make themselves the owner so we avoid overlap 18:50:04 (for the ff45-esr ticket effort) 18:50:15 good idea, yes, please 18:50:18 yes 18:52:53 okay, thanks for the meeting then. just a final thing: 18:54:15 i looked at our 38.1.0 branch because i was curious whether we have to deal with less patches now that another ESR cycle is almost over 18:54:45 it turns out, no, we are starting with almost the same amount of patches as back then which is :( 18:55:00 but, hey, what did i expect :) 18:55:03 *baf* 18:55:07 #endmeeting