18:00:44 #startmeeting 18:00:44 Let the Jenkins meeting commence! 18:00:51 #charis rtyler hare_brain danielbeck 18:01:00 #chairs rtyler hare_brain danielbeck 18:01:08 #chair rtyler hare_brain danielbeck 18:01:08 Current chairs: danielbeck hare_brain kohsuke rtyler 18:01:10 There 18:01:20 #topic Recap last meeting's actions 18:01:56 I wrote an overdue wiki page and made up the rules for it: https://wiki.jenkins-ci.org/display/JENKINS/Rewards+for+reporting+security+issues 18:02:28 It was a meeting action from ~February or so 18:02:45 LGTM 18:02:52 That's it from me 18:03:13 (Oh, and fork requests are still via mailing list, Jira project is still WIP) 18:03:50 Now that we have the JIRA agile set up, maybe this is a good opportunity to revive the MEETING project 18:04:07 #topic LTS status check 18:04:29 kohsuke: once https://jenkins.ci.cloudbees.com/job/core/job/jenkins_lts_branch/155/ is happy, we are ready for RC 18:04:33 danielbeck: actually, can you run the meeting? I'm in a meeting and multi-tasking here 18:04:38 kohsuke can do 18:04:50 ogondza: great! 18:04:53 danielbeck: thanks 18:05:03 kohsuke Don't close the laptop, there's two items from you 18:05:18 yes 18:06:03 ogondza So you expect that build to be successful, and once it is, KK can post the RC, and you'll post the testing wiki pages? 18:06:24 danielbeck: I will this time, I promise 18:06:56 there was a ;at minute backport, but I expect it to succeed 18:07:01 Got it. I'll check back in an hour or so and start RC 18:07:15 ogondza Which one was that? 18:07:19 *last 18:07:33 https://issues.jenkins-ci.org/browse/JENKINS-10629 18:08:09 \o/ 18:08:17 awesome thanks! 18:08:30 That guy also contained lib updates, but it should be safe enough 18:08:35 anything that was refused? 18:08:42 nope 18:08:56 I will update the ML 18:09:02 great 18:09:20 #topic Proposal - Revisiting JUC in 2016 18:09:27 https://wiki.jenkins-ci.org/display/JENKINS/Proposal+-+Revisiting+JUC+in+2016 18:09:44 kohsuke ^ your topic 18:10:02 Right 18:10:21 So there was a round of conversatiosn about this in the ML 18:10:45 I've seen some sentiment for losing a big European event, 18:10:53 but otherwise people generally seem to be OK with the concept 18:11:00 #info https://groups.google.com/forum/#!msg/jenkinsci-dev/YGEGjYvRzfs/fPRPUg41CwAJ 18:11:04 ... presumably including the name as well 18:11:20 (a collection of butlers is a sneer) 18:11:59 kohsuke: Does CB owns the Jenkins User Conference brand? Technically, could it be granted to another conference like JCPH? 18:13:16 I don't see much reasons to kill such format till there re relatively big conferences in Europe 18:13:40 alyssat This may be interesting for you as well 18:14:33 oleg-nenashev If you look at the proposal, there's CloudBees-Jenkins Summit 18:14:39 oleg-nenashev: nobody owns any name aside from SPI owning Jenkins mark 18:14:53 They're not a lot smaller than the European JUCes we had AFAICT 18:15:16 but no one is stepping up to offer doing JUC in Europe right now, so the conversation about whether that's being killed or not seems little hypothetical to me 18:16:01 Right, and Praqma is using the 'JUES' name anyway, at least so far. 18:16:11 Any further thoughts on the events? 18:16:33 Ideally I'd like to record that this plan is acknowledged & blessed. 18:16:52 No. Jenkins World should be good enough. +1 18:17:17 It does not mean I 100% support it from a community member perspective 18:17:17 sounds like a theme park :-P 18:17:29 It should be! 18:17:50 * rtyler jumps in 18:18:26 kohsuke: fwiw I like the proposal that is linked from the agenda 18:18:40 I'm happy that CB is willing to invest in a full community track 18:19:12 Any more +1/-1 ? 18:19:21 PLUS INFINITY 18:19:38 I'm excited to be able to bring the whole community around the globe to one place 18:19:39 as long as it's in california, or some place fun like big sur :P 18:19:44 montana maybe :P 18:19:54 We should be able to do some interesting contributor event, etc 18:19:54 rtyler sponsors his Infinity to Jenkins community. Cool, we will send him a T-short :D 18:19:55 * rtyler wants to vacation in jenkinsworld 18:20:52 Yep, while there's only a single very large conference left, there's a pretty big travel grant 18:21:05 * rtyler nods 18:21:30 Should be enough to pay for the hotel... 18:21:40 should we move on? 18:21:43 * rtyler shrugs 18:21:44 sure 18:22:15 rtyler's plus infinity beats everything, so no +1/-1 are coming 18:22:21 heh 18:22:42 danielbeck: https://twitter.com/agentdero/status/650709702238597120 18:22:43 xD 18:23:09 kohsuke: sticker time? 18:23:30 #topic Sticker purchase approval 18:23:33 #agreed not much active discussion about the events, but I think people are OK 18:23:34 oops 18:23:39 stickers 18:23:51 so I run out of all the stickers a while ago mostly in JUC 18:24:04 +1 for stickers 18:24:09 In preparation for FOSDEM, SCaLE, and 2016 I need more 18:24:23 Nobody here can even imagine how popular the 'angry Jenkins' stickers are, it's insane 18:24:29 I'm planning to make 500 of all 6 kinds, and maybe another 1000 of the standard sticker 18:24:31 Clearly, people don't see that one often enough 18:24:52 Should cost $1.5K 18:25:02 kohsuke: does it get any cheaper to get more stickers? 18:25:03 kohsuke Is 500 each going to be enough? 18:25:09 like why 1000 instead of 2000? 18:25:14 it's not like the logo changes :P 18:25:32 That's true 18:25:36 Should I just get more? 18:25:47 yes, it's terrible to run out 18:25:53 I guarantee we will go thorugh nearly almost all of the 500 sets 18:26:00 Why not? How much $$$ do we have in reserve? 18:26:00 maybe before the end of the fosdem 18:26:00 OK, then I can like double them 18:26:08 good question oleg-nenashev 18:26:12 500 sounds a lot like 0 after two days. 18:26:13 We have enough money in the bank to buy a cheap car right now 18:26:16 oleg-nenashev 20k+ 18:26:23 do the JAMs get stickets sent to them? 18:26:32 yeah 18:26:34 We can order a truck of stickers.... 18:26:53 hah 18:27:02 JENKINS TESLA 18:27:19 danielbeck: I feel like I have a blessing to move forward 18:27:36 kohsuke How many are you planning to buy now? 18:27:37 kohsuke: double the standard, I'd leave the others the same 18:27:58 with wood wheels 18:28:01 the 6-jenkins' should be more special, we can make it so we give them out in smaller special batches at FOSDEM 18:28:13 ok 18:28:30 we can make a quota for those at FOSDEM and build up excitement around them 18:28:40 Would be cool to have a new Jenkins 2.0 sticker... 18:28:44 like at the top of the hour we do a 2.0 nightly demo and give out those stickers 18:29:02 rtyler You're evil, I like it. Make people work for them! 18:29:09 oleg-nenashev: 2.0 sticker would a be a great idea 18:29:19 danielbeck: I've learned a lot from alyssat and heidi :P 18:29:53 oleg-nenashev: we will have some other special materials around jenkins 2.0 18:30:04 but I'll see if there's anything I can rustle up for a "special editoin" 18:30:07 separate from this meeting 18:30:08 danielbeck: back to you, Mr.Chair 18:30:14 * kohsuke sneaks out 18:31:06 rtyler Do we need a vote? 18:31:07 Ehh, Hudson became too old: http://www.dreamstime.com/stock-photography-butler-image25112082 18:31:10 danielbeck: FOR WHICH? 18:31:18 sticker buying 18:31:25 er, sorry about the caps 18:31:34 danielbeck: I don't think so, no large dissent here 18:31:41 Right, everyone loves stickers 18:31:53 heh 18:31:59 #agreed kohsuke to buy stickers for 2016. A lot of stickers. 18:32:17 #topic Should we require squashing commits in the core? 18:32:21 oleg-nenashev 18:32:34 #info https://groups.google.com/forum/#!searchin/jenkinsci-dev/squash/jenkinsci-dev/kqvuDqXwPZw/oF-KMdNbCwAJ 18:32:39 Шеэы ф ащддщц-гз ещ https://groups.google.com/forum/#!searchin/jenkinsci-dev/squash/jenkinsci-dev/kqvuDqXwPZw/oF-KMdNbCwAJ 18:32:43 sorry 18:32:50 *It's a follow-up 18:33:13 Personally, I hate making those things a rule for us 18:33:32 in a project like this, we all need to tolerate other people's tastes: https://en.wikipedia.org/wiki/Robustness_principle 18:33:33 BTW we need to agree on the common approach 18:33:34 FWIW jglick has good counter arguments 18:33:50 oleg-nenashev, пышь-пышь 18:34:06 english only please :) 18:34:23 rtyler, joke to his wrong keyboard) 18:34:40 * rtyler shrugs 18:34:49 If we don't squash, then OK. If we squash, then it should be a practice for all merges to make a value 18:35:55 I do not like to squash every PR into single commit 18:35:56 oleg disagree with forcing a sqaush 18:36:13 oleg-nenashev: are there existing commit guidelines? 18:36:16 a test and fix could be two separate commits in one PR and having the test separate is golden 18:36:26 this comes to mind for me http://tbaggery.com/2008/04/19/a-note-about-git-commit-messages.html 18:36:40 ogondza, multiple times everybody already said that squashing all doesn't make sense, people just want to see every commit as normal change and not tones of "fix style" 18:36:48 I propose to perform squashes for non-sensitive commits. I don't propose 1 PR = 1 commit rule 18:36:57 ogondza, commit should be logically ended 18:37:20 How about adding something like "maintain meaningful commit history" and request authors to rearrange 18:37:33 Or to do it during the merge 18:37:35 adding it to contribution guidelines ^ 18:37:59 Then you are forcing contributors to deal with `git rebase -i` madness. 18:37:59 ogondza, that's what we initially proposed, but stephenc wants keeping all commits even 30 styling changes commits 18:38:24 jglick, you enforce contributors dealing with 10y old code, rearrange PR is 1 min of work 18:38:59 Persons who merge PRs may perform re-arrange on their own. Or autommate the work in simple cases 18:39:01 jglick, not you directly, but project 18:39:38 Well this is just going to be a flamewar and using IRC is not going to make it any better. I have stated my position and arguments, as have others, so someone in authority should just make a decision. 18:39:38 One of the motivation for Jenkins for me is to be against code policing 18:40:03 kohsuke, you don't process any PRs last 2 years 18:40:05 jglick: So the agenda topic was "voting" 18:40:15 I contributed in the past to some project that enforces checkstyle rules that limited me to a method length of 30 lines, etc., and I hate those 18:40:24 Well who has a vote, and what are the candidates? 18:40:37 jglick, interesting question 18:40:49 Every meeting attendee has a vote 18:41:13 1) Not squash commits at all 18:41:13 2) Agree on a policy to prevent the history pollution 18:41:13 For a project to successfully accomodate lots of contributors each working little time, we need to be lax about these things, even if it has some associated downsides 18:41:17 jglick, let's start with statement that stephenc wants: "Don't chnage commits in PR" 18:41:57 kohsuke What about the argument that overly lax rules turn some people away as it looks like uncontrolled chaos? 18:42:18 kohsuke, frankly speaking i have tones of feedbacks how people hates your code and style :) 18:42:28 I don't think our lack of rule about squashing the commit is one o fthem 18:42:36 kohsuke, and all this people run away from jenkins 18:42:55 * KostyaSha just a fact 18:43:18 Any project should have a feasible amount of rules and common practices 18:43:21 I don't know, I consider it pretty successful that we have this many people contributed so far 18:43:42 I hate redundant commits in PRs, but I can't imagine that people are "running away" due to unsquashed commits. weird code style or bad documentation, perhaps, but unsquashed commits..? 18:44:01 * rtyler agrees 18:44:03 How about we encourage squashing once the number of commits hits two digits? 18:44:13 orrc, today there is a mix of normal commits and unsquashed = trashed 18:44:16 I mean if a PR has 4 commits, how harmful can that possibly be? 18:44:33 orrc, so to dig history you forced jumping between -m1 parent flags and showing all commits 18:44:42 As a Jenkins instance maintainer, I used GitHub a lot to navigate changes and select useful ones for backporting into my internal fork 18:45:01 orrc, it was not a problem because before last years all PRs was good and logically separated 18:45:02 * rtyler runs off 18:45:14 orrc, as soon ad CB started fast development, they started merging PRs as is 18:45:36 orrc, this mades commit history hardly researchable 18:45:50 I have to abstain regarding the comment above 18:46:08 arguments about github ui - wait for future features, develop git browsing tools... 18:46:37 so i can suggest try discuss it by steps 18:47:00 1) Stephenc wants keep PR as is that doesn't allow picking PR for somebodies fixuping 18:47:04 + -? 18:47:30 1) Not squash commits at all 18:47:30 2) Agree on a policy to prevent the history pollution 18:47:45 i see a problem that many useful PRs authors keeping without answers and we need end them manually 18:47:55 ^^^ I propose to vote for the bullets above 18:47:57 so i'm -1 for this point 18:48:08 ================ 18:48:33 -1 on both option 1 and option 2 18:48:51 teilo: And what do you propose? 18:48:54 It doesn't look to me like this is even remotely ready for voting on it. 18:49:00 recomendations to PR authors 18:49:18 This is missing an actual proposal that addresses concerns etc. 18:49:23 danielbeck, i think this discussion mostly depends on kohsuke as he always do veto on any proposals 18:49:27 I would propose to develop recommendations for PR reviewers 18:49:49 oleg-nenashev: should exist for PR authors 18:50:02 aswel 18:50:21 agreed 18:50:23 danielbeck, as kohsuke has only 1 hour a two weeks and very busy, we should provide our opinion to him here 18:50:37 and hope to some resolution :) 18:51:11 It should be agreement within jenkinsci/core 18:51:33 There is no need to have a dictator in such topics 18:51:49 KostyaSha: So your point is invalid IMO 18:52:20 oleg-nenashev, what point? what stephenc wants? it not my point :) just trying to extract logical statements from email thread 18:53:11 If we don't vote now, what are the action items? 18:53:41 I can propose recommendations, but I'll likely get a push-back from kohsuke, stephenc and jglick 18:53:43 if we can't vote and email thread has no final solution - what actions expected from this meeting? 18:53:55 oleg-nenashev Write a coherent proposal and solicit feedback on it. Seems to be right now that opinions are too far apart and concerns not addressed. 18:54:26 danielbeck, he wrote proposal in email thread 18:54:37 danielbeck, and everybody provided feedback 18:54:37 he, Oleg? 18:55:02 danielbeck, yes, he even showed a tool for squashing :) 18:55:25 Nt a tool 18:55:39 Just an aux utility for primitive cases 18:56:13 ogondza, what is your opinion? 18:56:30 I'm having a hardstop within 4 minutes 18:56:31 ogondza, as i see LTS backporting tools can't deal with -m1 merges 18:58:39 seems kohsuke is busy with twitter and posting about stickers 18:58:57 oleg-nenashev, no answers, postpone topic and shift to other? 18:59:06 Probably 19:00:47 KostyaSha oleg-nenashev So we're going to continue this discussion next time? 19:01:14 Yep. Wpuld be great if somebody joins the discussion in the ML 19:01:15 danielbeck, obviously yes 19:01:37 2) for me => as danielbeck said I have already been in the past in the position where I actually find those rule more reassuring than having none stated and you discover when you push/propose the patch. 19:01:58 I'll try to poke KK about it 19:02:03 #topic next meeting 19:02:17 Next meeting on Nov 11 19:02:25 works for me 19:02:44 10 am PST by then I think, so be aware of off by one errrors 19:02:45 can someone link to the mailing list discussion? I can take a look. 19:03:07 *11 am 19:03:07 autojack, yes, channel will have +m soon, will post in jenkins 19:03:08 I didn't arrive for the meeting until a few minutes ago. 19:03:11 OK 19:03:18 #endmeeting