18:00:59 #startmeeting 18:00:59 Let the Jenkins meeting commence! 18:01:13 #chair hare_brain rtyler danielbeck KostyaSha 18:01:13 Current chairs: KostyaSha danielbeck hare_brain kohsuke rtyler 18:01:15 oops 18:01:22 :D 18:01:27 I meant to type in myself but got KostyaSha instead 18:01:29 oh well 18:01:41 two times yourself?! 18:01:42 #topic Recap last meeting's actions 18:02:05 The only action from last time was that I get some way to annotate the LTS changelog 18:02:11 Aye 18:02:20 And that's mine 18:02:26 That hasn't happened AFAIK, so I'll just edit the file on the site 18:02:49 It continues to be high on my TODO list FWIW 18:03:00 From an earlier meeting, I had the list of infra admins, and it also includes Artifactory data now: https://wiki.jenkins-ci.org/display/JENKINS/Infrastructure+Admins 18:03:00 #topic LTS status check 18:03:06 Oops 18:03:18 should I set the topic back? 18:03:19 * ogondza forgot to push on LTS rc release 18:03:38 That's just FYI. Other than making Tyler an Artifactory admin when he's available, it's finish IMO. 18:03:45 * rtyler nods 18:03:52 thanks for that danielbeck 18:04:13 I also noticed we never followed up on jenkinshosting.com, so we'll need to do that some time 18:04:28 Yes, it's a good work towrad more transparency 18:04:46 Would be interesting to see any past communication between the guy and KK, he wrote to the list once but that was it 18:04:48 Does Alan need permissions? He rarely appears in the community now IIRC 18:04:55 danielbeck: I can forward it to you 18:05:06 okay, then I'll follow up on that 18:05:06 I've not seen mindless in eons 18:05:07 Basically he's very happy to go through the formality 18:05:33 I communicated to him that for us to protect the mark we need it, but we aren't trying to shut you down. 18:05:45 I mean shut him down 18:05:50 Yes. I think we never intended to not grant that use. 18:06:00 #action kohsuke to send danielbeck my correspondence with JenkinsHosting.com guys 18:06:05 FWIW the wiki page on granted uses is outdated as well 18:06:18 #action danielbeck to follow up with them 18:06:22 #action danielbeck to update the granted trademark usage page 18:06:23 their site has hudson screenshots 18:06:25 lollz 18:06:27 :-) 18:06:37 Let's move on 18:06:45 They did start as HudsonHosting.com 18:06:53 OK, so now really LTS status 18:07:09 I swear to god that I did produce & posted RC binaries 18:07:28 So I need to chase down where the lapse is in automation 18:07:37 no announcement to the list 18:07:38 http://mirrors.jenkins-ci.org/war-stable-rc/1.625.1/ 18:07:46 and no wiki page for RC testing 18:08:01 Right, no announcement to the list was my fault 18:08:11 I completely lost track, my apology 18:08:13 oh I see, it didn't hit mirrors 18:08:26 anyway, we are ready 18:08:34 I see that the calendar was adjusted already to mark today as 1.625.1 RC date 18:08:44 yep, I did that yesterday when we noticed 18:08:46 I have backported one more issue and the revert what KostyaSha asked for 18:09:14 #action ogondza prepare the LTS testing wiki 18:09:16 OK, then I should build 1.625.1 RC2 then 18:09:26 Do we merge any security stuff into 1.625.1 ? 18:10:07 I think .2 is better in general for security unless it's urgent? 18:10:14 +1 18:10:25 oleg-nenashev: historically, I believe we've done that right before the release, which means fixes skip RC soak tests 18:10:56 I remember having that discussion with jglick when we started doing it 18:11:03 with the exception of the master/slave system 18:11:43 kohsuke: I just want to make sure, because several PRs have been merged. I also vote for .2, but we need to make sure that the current fixes won't be taken into the release by the release bot 18:12:08 Is it better to have this conversation in jenkinsci-cert list? 18:12:30 I don't want to get too specific here about undisclosed issues or fixes we are planning to release in 1.625.1 if any. 18:12:43 Yes we can take this to jenkinsci-cert. 18:12:46 agreed 18:12:47 ack, this is a process change 18:13:17 ogondza: is what I just described a process change from what we've been doing? I was trying to explain the current process. 18:13:37 So we're done with LTS discussion? KK will post RC 2 and make sure it gets distributed correctly. 18:13:50 Yes 18:14:03 #action kohsuke to post 1.625.1 RC2 18:14:13 ... and announce it 18:14:15 No, though preferring .2 for security fixes was proposed several times though not sure if sainted on meeting so I assume it is not a process 18:14:25 kohsuke: ^ 18:14:37 Yeah, indeed that's new 18:14:47 I have some concerns about that, but I will take it to jenkinsci-cert 18:14:49 Let's continue that discussion on the cert list 18:14:51 #topic Q4 patron program 18:15:18 danielbeck: yours 18:15:21 No changes to Q3. I've never received the final second CloudBees message, but apparently I will soon 18:15:49 I think we can just agree to continue to run the same messages. 18:15:56 https://github.com/jenkinsci/patron/pull/4/files 18:16:18 +1 18:16:57 Shall we move on? 18:17:02 anyone else ±1? 18:17:05 +1 18:17:09 * rtyler shrugs 18:17:10 sure 18:17:36 Looks like nobody's opposed at least 18:18:09 Kif, what makes a man turn neutral 18:18:16 only if it not a huge unclosable banner 18:18:16 #agreed same patron messages in Q4 as in Q3 18:18:32 KostyaSha Same message format as always 18:19:26 FWIW I believe the document only says new blessing needed for new messages 18:19:40 Anyway, shall we move on? 18:19:54 yep 18:20:00 #topic Time to think about Jenkins board elections? 18:20:09 ... from orrc: 18:20:53 #info the 2-year old "latest" proposal is here: https://wiki.jenkins-ci.org/display/JENKINS/Board+Election+Process+Proposal 18:21:20 Is this page still relevant? https://wiki.jenkins-ci.org/display/JENKINS/Governance+Board 18:21:45 *relevant/up-to-date I mean. 18:21:45 uh, I think s/abayer/rtyler and yes 18:21:57 So far I've been failing to implement the account app change, but I agree we need to do it sooner 18:21:58 what will happen if CB hire newly elected non CB person? 18:22:12 I'm still +1 on proposals, btw orrc's concern must be addressed 18:22:30 batmat: right, abayer has transitioned a seat to rtyler 18:22:34 recorded live on 100K podcast 18:22:42 very official like :P 18:22:55 so, 2 CB and hare_brain ? 18:22:56 We should announce that more formally too 18:23:52 Ok, updated the page, is is OK now? 18:24:13 "Any company must have <50% of board members" seems to be a reasonable addition 18:24:34 +1 18:24:38 I'm not sure it does though 18:24:41 oleg-nenashev: orrc: I wanted to actually discuss if it's so inherently bad that CB employees have majority 18:24:50 it also seems to me that now nobody from board really uses/supports jenkins for some projects 18:25:00 kohsuke's clearly involved with how CB policy is made 18:25:12 * i mean practise 18:25:22 but if you take somebody like danielbeck, I'm less convinced that his CB involvement "taints" his ability to act as a board member 18:25:52 I think we should look to how something like Ubuntu is operated here 18:26:06 or openSUSE 18:26:19 KostyaSha, not sure what you mean 18:26:26 if CB hires all the people who are passionate and spend lots of time on jenkins, we can't just *not* have a board 18:26:31 kohsuke: I think that it's pretty important. It will prevent formal complains that a company may dominate in the community and push any decision 18:26:48 I guess my concern is that people who are getting stuff done, like danielbeck, gets necessary authority to act. 18:26:54 hare_brain, i mean that abayer configures and supports apache's jenkins and infra. Do you support any jenkins installation as user? 18:27:11 oleg-nenashev: Yes, I totally understand the motivation behind "no majority" rule 18:27:37 Of course I do. 18:27:52 My team runs Shutterfly's release infrastructure, which is based on Jenkins. 18:27:57 the purpose of this topic is not to elect board members right now 18:28:08 hare_brain, good 18:28:18 can we not compare jenkins credentials and discuss the topic at hand 18:28:56 It's a bummer orrc is not here 18:29:00 it is 18:29:02 How are we to go forward with this one? A thread on the ML about the different models? 18:29:16 maybe orrc will be able to express more things in async? 18:29:25 and others btw 18:29:26 so there's a qualifying question that I'd like to ask about that 18:29:42 Is there a public record if orrc's concerns? 18:29:43 there's a definite bias we're going to get by keeping topics like this, and the Jenkins 2.0 discussion, on the -dev mailing list 18:29:53 hare_brain Only the meeting topic I think 18:30:19 there are plenty of users on the mailing list, are their opinions not important to weigh in here? 18:30:47 @rtyler: I think it's perfectly fine to have Kohsuke on the board, but CB should not have a 2/3 majority on the board. For the sake of "could look a bit shady" for people that don't know the motivation of each board member. 18:31:03 rtyler: sure, I suppose kohsuke just thought it was more something technical first, before visible/user changes. That's my understanding 18:31:10 fredg02: I understand that bit I'd like to quantify it more than "looks shady" 18:31:19 but* 18:31:50 Maybe we can go the UK monarchy route and define a separate body under board that has additional authority delegated to act 18:31:52 if in a year, all the highly active particpants in the project work for CB, does that mean we have no government? :P 18:32:12 * rtyler doesn't want to become belgium 18:32:13 * rtyler ducks 18:32:20 So KK becomes KKK (king KK?) 18:32:20 let's add oleg-nenashev's addition "Any company must have <50% of board members" to the proposal 18:32:41 so 2/3 or <50%? 18:32:57 less than the majority ;) 18:32:57 I would also like us to consider as a community some other projects which have similarly influential corporate participants 18:32:59 Well the proposal was to expand the board to 5 members. 18:33:08 well, if we add that then IIUC there's then an immediate issue? 18:33:13 hare_brain: ok, I see. 18:33:32 fredg02 So skip whoever wold get elected if they'd push it over 50%? Or have only one candidate per company? 18:34:05 rtyler: that was my thought to give us time to discuss that on the ML, with people having a bit more time to explain (and maybe push towards) some other examples. 18:34:09 danielbeck: +KKK as an obvious exception? 18:34:10 what if I have a ceritifed letter from Sacha saying CB does not tell me how to act on the board? :P 18:34:13 I don't mind having 2 CB ppl on the board if the board has 5 members. 18:34:28 Oh boy, I hope this KKK thing won't stick. 18:34:31 haha 18:34:35 oleg-nenashev, diods?! 18:34:39 We already got slave s and executors going around. 18:34:47 hehe 18:34:59 lol 18:35:02 if a board member accepts a job at CB, do they then lose their seat 18:35:16 rtyler: right. Or FIFO ? 18:35:18 :) 18:35:20 hah 18:35:20 rtyler: I would prefer it to be "at the time of an election" 18:35:28 FWIW I'm for ogondza and vjuranek in a three person board so it's for once not CB that's considered a problem :-) 18:35:51 hah, ETOOMANYFEDORAS 18:36:19 teilo: I would prefer "any time", but I'm not sure how democracy should work here 18:36:19 danielbeck: I think that is a good point, does the rule apply to all corporate entities or just CB 18:36:33 same rules for everyone! 18:36:37 what if I work for Hive Holdings, and KK works for CloudBees and we both work for Sacha Enterprises 18:36:40 does that count? 18:36:53 (redhat's not quite like that, but amazon and google are definitely fragmented like that) 18:37:09 alphabet... you mean 18:37:29 I'd think if we embrace a rule like that, then I think it's the spirit that counts 18:37:31 I'm saying this not to be annoying (which I'm sure I am) but that whatever structure if there's going to be a guard against undue corporate influence that it's structured properly 18:37:44 rtyler: I don't think there is a need to be so formal. WEANAL 18:38:12 this acronyme is even more crappy 18:38:15 heh 18:38:31 Board not having enough bandwidth has been a problem and it is still a problem today 18:38:39 rtyler, it easy define! all guys that pushing workflow are CB :) 18:38:47 I think my problem is that if it's hard to quanitfy ('looks shady') and there aren't strict guidelines in place 18:38:55 then "looks shady" can be used as a weapon against the board in the future 18:39:19 depending on where one is in the community, one will perceive shadiness differently 18:39:25 rtyler: then there should be a disclosure policy and recusal in event of conflict of interest 18:39:38 As I said, my concern in having the proposed rule is that it excludes lots of people who can spend time on the project, which is somewhat anti-meritocratic 18:39:54 mmars: aha, now you're getting to the heart of it I think! 18:40:00 So if we are to adopt the proposed rule, I want us to spend some time thinking about it. 18:40:14 kohsuke: IIRC Jenkins contribution does not require the board membership 18:40:17 so let's say danielbeck's job description is "work on Jenkins FOSS full time", does that present a conflict of interest? 18:40:26 @kohsuke: so you think we run out of candidates if we enforce the <50% ruleß 18:40:27 ? 18:40:52 kohsuke as a CB investor and executive obviously *does* have a conflict of interest 18:41:06 oleg-nenashev: for code commits, yes, but when you interact with outside, being able to represent the community does help 18:41:21 @rtyler: yes, but it's ok as long as he does not have majority 18:41:26 e.g., talking to SPI, ffis, or driving code of conduct investigation 18:41:36 It seems there is a question of scope of the governance board. For example, the board has nothing to do with workflow as a feature of Jenkins, other than overlapping people. 18:41:48 It's just a "Board of Directors", which gives advices and acts as arbitrator. That's why it's important to keep it independent from both formal and informal POVs 18:42:10 fredg02: I wouldn't say we run out of candidates, but the pool will be smaller for sure 18:42:39 I actually think that could potentially have the opposite effect 18:42:41 @kohsuke: but that's the lesser evil, I think 18:42:57 since now CB as a company would have an interest in putting a few candidates forward that are in their interests 18:43:03 as opposed to people volunteering themselves 18:43:31 hare_brain: oleg-nenashev: right, that line of thoughts leads to the board delegating some more exectuive abilities to another group 18:43:45 anyways, I agree with your point about there being some overlapping issues here hare_brain 18:44:47 just a time check, we've got 15 minutes left here 18:45:01 so let's nail down some actions to take between here and the next meeting to get this moving forward 18:45:12 fredg02: did you add the 50% thing to that proposal? 18:45:25 we have not voted yet 18:45:37 IMO, the board should not be involved in deciding feature roadmap, technical decisions, things about Jenkins-as-a-product. It's scope should be Jenkins-as-a-community. I can pretty clearly separate the two concepts. Not sure if that's how people generally feel. 18:45:39 There's also a question raised by teilo in such case 18:45:52 the goal is to prevent the board from agreeing to thinks like "let's show advertisment of on every Jenkins installation" while has majority in the board 18:45:52 hare_brain: +1 18:46:10 #action hare_brain to write out a revised proposal for the scope of the jenkins board 18:46:13 :) 18:46:37 s/thinks/things 18:46:54 fredg02 exactly right 18:47:24 * teilo notices I never did move the servlet 3.0 18:47:43 rtyler, the current Governance document describes it pretty succinctly and accurately. https://wiki.jenkins-ci.org/display/JENKINS/Governance+Document#GovernanceDocument-board 18:48:17 We can stick in an explicit "The board's authority does not extend to features or roadmap." 18:48:17 hare_brain: I think what's more important, to what you wrote earlier, is to write what the board doesn't do 18:48:25 saying we're queens doesn't really explain much :P 18:48:45 It does for me, so I must have written that part 18:48:57 some Queens have been dictators as well :-) 18:48:58 heh 18:49:19 hare_brain: should I take that AI? 18:49:29 rtyler: It means that the goverment can reduce your fundings :) 18:49:37 heh 18:49:49 Take work off my plate? Sure. :) 18:49:54 well, right now the project technically owes me money, so I don't think my funding can be removed any more 18:50:35 rtyler: sorry I didn't follow up on that one, let's follow up and fix that 18:50:42 heh 18:50:53 I should declare jenkins as a dependent on my taxes 18:51:03 #action kohsuke to take previous action from hare_brain 18:51:06 other action items from this? 18:51:11 Too late, it's already in my tax return 18:51:40 @rtyler re-election sometime soon? 18:51:43 if oleg-nenashev or fredg02 can update or append their corporate influence proposal to the doc that'd be good 18:51:51 see https://wiki.jenkins-ci.org/display/JENKINS/Board+Election+Process+Proposal 18:52:09 I think we are mostly saying very similar thing, so hopefully a clarifiying the role of the board and adding that rule would let us bless it 18:52:22 We need someone to implement the change to accountapp 18:52:36 fredg02: maybe? I think it makes sense to do it next february on Jenkins' birthday 18:52:46 that way we can keep the number of dates we all have to remember down :P 18:53:11 rtyler: sounds good 18:53:14 That's a great idea 18:53:26 I think that would also help get us into a rhythm 18:53:43 Any volunteer to help the accountapp code change? 18:53:46 wasn't that just the 100k? 18:54:01 #action rtyler to put proposal for board elections on feb 2nd annually 18:54:15 kohsuke: what's the scope of the code change? 18:54:25 kohsuke: I've been pondering updating that app anyways :p 18:54:30 such a PITA for management 18:54:38 hopefully not that much 18:54:40 heh 18:54:43 got JIRA? 18:54:43 I talked about that briefly with danielbeck 18:54:51 *where* is the code? 18:54:52 rtyler Adding a page for voting, select which of the candidates record the choice, and export into a format that can be fed into an STV tool 18:54:52 #action oleg-nenashev to update proposals regarding the per-company limit of board members 18:55:13 #action kohsuke to file a JIRA on what needs to change in the accountapp to make board elections viable 18:55:24 #info code is https://github.com/jenkins-infra/account-app 18:55:40 mabye I can get gus to help pretty it up a bit ;) 18:56:00 Clearly the highest priority 18:56:11 heh 18:56:18 Line up in the queue behind me, rtyler. I got a long list of things for him 18:56:41 kohsuke: drinking contest? :P 18:56:41 OK, I think it's good enough for this topic for today 18:57:08 kohsuke Maybe start a thread on the list about this? This is a pretty small group after all 18:57:10 shall we cover the servlet 3.1 topic? 18:57:15 agreed! 18:57:17 Yes, and yes 18:57:24 very US timezone biased 18:57:26 #topic time to move to Java 8 and servlet 3.1 18:57:43 teilo: does that you mean you didn't read the Jenkins 2.0 post or you cannot wait for that? 18:58:30 Are there good solutions for the Maven Project Plugin to build with older JDKs? People complain about the recent Java 7 requirement already, and I expect it to not get better once LTS has it as well. 18:58:37 kohsuke: 1 do it now not later. 2 the propsal said servlet 3.0 not 3.1 18:58:48 oh 18:58:55 what's new in servlet 3.1? 18:59:15 non blocking io 18:59:20 danielbeck: I think it's the fundamental limitation of that plugin 18:59:33 but some of non-blocking I/O was a part of 3.0 19:00:03 Servlet 3.0 allowed asynchronous request processing but only traditional I/O was permitted. 19:00:05 #info https://blogs.oracle.com/arungupta/entry/what_s_new_in_servlet 19:00:08 kohsuke Yes, but with almost 1M jobs out there we should consider the impact of moving to Java 8. 19:00:23 Because "It's just master, you can build what you want" is a load of shit 19:00:45 i'm out of time, catch you folks after lunch 19:00:47 s/build/build with 19:00:49 o/ 19:01:12 Yes 19:01:40 Servlet 3.1 part doesn't feel like a problem 19:02:09 I think this conversation is probably better suited for the ML 19:02:22 Especially given that we are running out of time 19:02:42 Right. There's not going to be a decision today. 19:03:12 teilo: if you want to propose this before 2.0 you should send a proposal to the ML 19:03:22 #topic next meeting 19:03:48 #info the next meeting will be Oct 14th the same time 19:03:55 When does DST end? 19:04:07 end of oct 19:04:14 OK, so we are still safe 19:04:18 #endmeeting