Below is the transcript of a committer's meeting held on #dspace at irc.freenode.net at 16:00 GMT on April 22.
We intend to hold the next meeting, on #dspace April 29th at 20:00 GMT You can see the upcoming (weekly) schedule as well as other DSpace developer events at: http://www.google.com/calendar/embed?src=3mfp5qsv0kejvsbh558lmshujk%40group.calendar.google.com&ctz=America/New_York Anyone is welcome to attend. Meeting Summary: We agreed to meet every week for 60 minutes, alternating between 16:00 and 20:00 GMT. We discussed moving the majority of votes and discussions off dspace-commit and onto dspace-devel. We're working on the next steps to move SVN from SF to OSL. ---- bradmc: Hello all. We'll wait another minute or two. Start brainstorming up topics. [12:01pm] Stuart-Lewis: Hi all [12:01pm] bollini: hi all [12:03pm] mdiggory_: Good morning/afternoon/evening [12:03pm] bradmc: Let's start tossing out topics. [12:03pm] bradmc: Meeting Frequency / Length / Next Meeting. [12:03pm] Stuart-Lewis: Can we discuss the need for more committers, how many, and what we should do about it? [12:04pm] Stuart-Lewis: I can give brief feedback with initial survey results [12:05pm] bradmc: Followups on 1.5.2, if any. [12:05pm] bradmc: Status of SVN to OSUOSL. [12:06pm] bradmc: Shall we start with the meeting administrivia, then pick up the committer need questions, and proceed from there? [12:06pm] Stuart-Lewis: ok [12:06pm] bollini: go ahead [12:06pm] mhwood: Yes [12:07pm] bradmc: We've been going at pretty much one week intervals, as we had a lot of catching up to do. Is it time to settle into a two week interval schedule? [12:07pm] bradmc: Are folks happy with the two alternating times? [12:07pm] scottatm: Yes [12:07pm] bradmc: Is 60 (nominal) minutes the right length? [12:07pm] mdiggory_: Yes [12:08pm] mdiggory_: I have to leave in exactly 52 minutes... [12:08pm] bollini: could be insufficient with a frequence of two weeks [12:09pm] Stuart-Lewis: Yes (Alternating) Yes (60 mins) Yes (biweekly) [12:09pm] mdiggory_: I think the weekly meeting is proactive and keeps us engaged [12:09pm] mhwood: I'm happy with the times, but I live in GMT-5 so I would be. [12:09pm] mdiggory_: something the community needs. [12:09pm] Stuart-Lewis: Weekly is fine by me too [12:09pm] bradmc: Okay, I'm going to set a 60 minute weekly wednesday meeting, alternating times. We can adjust when if situations change. Thanks. [12:10pm] mdiggory_: k [12:10pm] bradmc: Do we need more committers? Do we need more developers? [12:11pm] Stuart-Lewis: Yes + Yes [12:11pm] bradmc: (Yes + Yes) [12:11pm] bollini: more committers: yes - more developers: yessssssssss [12:11pm] mdiggory_: Of course, yes + yes [12:12pm] bradmc: How many, and what to do? (Aside from completing Ben's status) [12:12pm] mdiggory_: bradmc: thats not a question I think we should answer [12:13pm] mdiggory_: I mean "How many" [12:13pm] scottatm: Yes, it's not about numbers it's about community. [12:13pm] mdiggory_: "What to do", thats one to answer [12:13pm] mdiggory_: scottatm: agreed [12:13pm] Stuart-Lewis: Do we maybe want to introduce a committer emeritus staus that we can choose to move to (and back from if wanted) that individuals can choose to show they no longer have the time? This will help us judge effort available, while still recognising the valuable input they have given? [12:14pm] Stuart-Lewis: E.g. http://www.eclipse.org/projects/committers-emeritus.php [12:15pm] mdiggory_: possibly, but with the new adoption of Apache policies as a basis for our community, thats not so important now [12:15pm] Stuart-Lewis: How so? [12:15pm] mhwood: Some good things have been done to make it easier to be a developer, and more good things are in progress. I think that the more support there is for soft-coupled addons and for making community assets known/accessible, the better. [12:15pm] mdiggory_: majority vote vs unanimous [12:15pm] bradmc: [Good point Mark] What does emeritus mean? That we don't chase them for votes or help, but they can choose to decloak whenever they want? [12:16pm] bradmc: That was good point mdiggory, followed by good point mhwood. ? [12:16pm] Stuart-Lewis: Yes - it would mean something like that. [12:17pm] bradmc: I tend to like that idea, if it's done with minimal formality. It just codifies what is already occurring. [12:17pm] mdiggory_: Stuart-Lewis: we should not get gridlocked when voting on items now [12:18pm] Stuart-Lewis: Oh yeah - wasn't thinking of formality, just mainly to help us judge effort available. [12:18pm] mdiggory_: I propose another idea... Voting and Proposals should happen on the developers list in public instead of in private commiters list [12:18pm] Stuart-Lewis: Those of us here are possibly the wrong people to be asking about that status. [12:18pm] bradmc: +1 public votes. [12:19pm] scottatm: +1 public votes [12:19pm] bollini: -1 public votes for committer, +1 for all developer questions [12:19pm] bradmc pokes jrutherford on emeritus [12:20pm] Stuart-Lewis: For committer votes I think this might be dangerous? What if the most recent one had happened in public? It may have caused upset in public. [12:20pm] mdiggory_: +1 on public votes for commiter and development [12:20pm] Stuart-Lewis: But +1 othwrwise [12:20pm] mhwood: Publish results on -devel? [12:20pm] scottatm: Only security issues, non development issues, committer votes should be on the committer email list. [12:20pm] mdiggory_: Stuart-Lewis: that might have been a good thing [12:21pm] Stuart-Lewis: Maybe - would have given the individual in question a chance to represent themselves. [12:21pm] bradmc: I think we should eventually get to the point that committer votes are public, but I'm willing to take one step at a time. Note: I intend to post this IRC log. ? [12:22pm] mdiggory_: The point of exposing proposals/voting for new commiters is that it breaks down the "exclusivity barrier" of the commiter group and forces them to voice their oppinions in manner appropriate for the community [12:23pm] Stuart-Lewis: Do we not do that already? (voice approperiate...) [12:23pm] bradmc: I think that when we were using defacto unanimity, privacy was required due to the high probability of vote failures. Under current rules, that isn't as much of an issue, and I agree with Mark that the sunshine helps. [12:24pm] Stuart-Lewis: But might people be scared on making a nomination in case it gets shot down, and someone gets hurt? [12:24pm] bradmc: Yes, I think we voice them appropriately, but when absent a mechanism to resolve them, it would be rather disruptive. But we've solved that problem (or at least pushed it elsewhere). [12:25pm] Stuart-Lewis: I want to make a nomination now, but if you all -1 that person, he'll not thank me for having nominated him. [12:25pm] mdiggory_: Stuart-Lewis: With majority vote, this becomes less of a problem. [12:26pm] bollini: we have already a different case that is not appropriate to make publicly available [12:26pm] mhwood: If you stand up in a large group and propose a vote among a subset, you need to make clear who is eligible and why it's taking place in the broader context. [12:26pm] Stuart-Lewis: But there is still the potential to shoot people down in public even with majority voting [12:26pm] mdiggory_: Nominations should be based on the individuals reputation and contribution. [12:26pm] bradmc: At the risk of seeming inconsistent, I think that you should garner support via private channels until you are confident, and then nominate with some confidence. [12:26pm] bollini: I'm thinking about a "temporary" (hope) refused committership [12:26pm] bradmc: At that stage, you are likely to have some sense of if issues are likely to arise. [12:26pm] mdiggory_: Its best not to nominate someone unless you've got a stong sense that theres a majority who will agree with you [12:27pm] mdiggory_: I also wouldn't nominate someone your unsure would accept the position [12:28pm] mdiggory_: This sounds like a topic it may take a couple meetings to vet out [12:29pm] Stuart-Lewis: Can we continue to -commit for a few weeks so we can vote for a few more a.s.a.p. and ramp up the effort a bit right now while we are sorting this? [12:30pm] mdiggory_: Given my limited time this morning, can we run through the agenda quickly and then return to topics if folks want to? [12:30pm] bollini: I agree with Stuart [12:30pm] bradmc: Sure. What topics did you want to get to, Mark? [12:31pm] mdiggory_: OSL... we should probibly work out some of the things that need doing and who/how we tackle them? [12:31pm] mdiggory_: then 1.6/trunk adjustments we will need to make as well [12:31pm] bradmc: Okay, to close the last topic, I propose we go with scottatm's suggestion for now: Only security issues, non development issues, committer votes should be on the committer email list. [12:32pm] Stuart-Lewis: Sounds good to me. [12:32pm] mhwood: OK [12:32pm] bollini: ok [12:33pm] bollini: OSL: I have only make a single checkout test and it has worked fastly [12:33pm] mdiggory_: likewise, we should address some of the issues arising with people trying 1.5.2. Are there any problems with the release [12:33pm] mdiggory_: bollini: ok, I need to go and verify accounts. I havn't been to the admin interface this week/ [12:34pm] bradmc: I've gotten a couple of documentation notes on 1.5.2 that I'll roll in and keep up to date at dspace.org. [12:34pm] Stuart-Lewis: (Just an aside - to be fair to SF, their SVN service has been more stable since Christmas) [12:34pm] bradmc: But SF is still not fine-grained access. [12:34pm] Stuart-Lewis: Are backups etc all checked and proven at OSL before we move? [12:35pm] bollini: we need also an owned maven repository [12:35pm] Stuart-Lewis: bradmc: Oh yeah, wasn't saying we shouldn't move, just commenting on its improved stability [12:35pm] mdiggory_: bollini: yes [12:36pm] mdiggory_: ok I have mwood, bollini and the dspace 2.0 commiters as new users in the OSL svn [12:36pm] bradmc: Ugh. That doesn't seem like enough testers. [12:36pm] mdiggory_: Others should go and get accounts so I can give you commiter rights [12:37pm] mdiggory_: https://dspace.osuosl.org/trac/dspace/register [12:37pm] Stuart-Lewis: Are backups etc all checked and proven at OSL before we move? [12:37pm] gabriela: I can have a look at the OSL, I'll get an account shortly [12:38pm] bollini: what kind of test we need to perform? should we check the grain fine permission? [12:39pm] bradmc: Next step: get all committers on there. After that: validate infrastructure (backups, permissions, etc). Then sort out all the links / references to be changed (dspace.org, wiki, flagging SF, ...). Then set the cutover date. [12:40pm] mdiggory_: gabriela: since theres no notifications I receive on new users, just email me after you do it [12:40pm] Stuart-Lewis: mdiggory_: I have just created a stuartlewis account [12:41pm] mdiggory_: I still need to add you into the groups I also want to reorg the repo a little bit. [12:41pm] mdiggory_: ok, will focus on setting this up later today, after the eeting [12:41pm] scottatm: mdiggory_: I have created a ScottPhillips account. [12:41pm] gabriela: mdiggory_: I have just created an account, username gabriela [12:42pm] mdiggory_: thanks everyone, will get this tuned up for testing on your end. [12:42pm] Stuart-Lewis: mdiggory_: Can all this be documented in the wiki so the rest of us can administrate it too? (just the basics like adding a new committer http://wiki.dspace.org/index.php/NewCommitter) [12:43pm] mdiggory_: bollini: for instance, I will setup a "dspace" project sibling next to the dspace2 project, we will control access so that you cannot commit to dspace2, but can to dspace [12:43pm] mdiggory_: Stuart-Lewis: Yes... I can use all the help I can get [12:43pm] mhwood: Haven't used Trac before and not yet familiar with the organization of the doco. I suppose that when I find in the browser what I want to check out, I just feed the same URL to svn? [12:43pm] Stuart-Lewis: Great ? [12:44pm] mdiggory_: mhwood: we will be clarifying that in the docs, but the svn repo is not the same as the browser [12:44pm] mdiggory_: the browser is like ViewVC [12:45pm] mhwood: OK, will watch that space. [12:45pm] mdiggory_: not, I'm not recommending using Trac for anything but svn account management [12:45pm] mdiggory_: not = note [12:45pm] mdiggory_: JIRA for issue tracking, Recommend starting to use crucible for source code browsing [12:46pm] mdiggory_: For new, I think we keep the listservs at SF [12:46pm] mdiggory_: (typos) [12:46pm] Stuart-Lewis: They seem to work ok there, if a little slow. Less admin for us if they stay there. [12:46pm] mhwood: SF archive interface is terrible but at least we're used to it. [12:47pm] Stuart-Lewis: Shall we do the 1.6/trunk svn rejigging before or after the move to OSL? [12:47pm] bradmc: +1 Jira, Crucible, listservs at SF. I can pick up the Crucible piece. [12:47pm] bollini: list can wait for me... instead we need to solve the maven repo issue [12:47pm] mdiggory_: We just need to "repoint" that at Atlassian [12:48pm] bradmc: Ah, you mean the public one. Fine. Foundation also has a license to run it locally. [12:48pm] mdiggory_: bollini: we can work on the maven repo under the same vhost name and just move it wherever we deem appropriate [12:49pm] mdiggory_: the OSL host may be ok for it intially, but we will want to consult with OSL about any possibly better solutions [12:49pm] mdiggory_: "under the hood" I mean [12:50pm] mdiggory_: they run some file/https services that may be more appropriate to locate it on [12:50pm] mdiggory_: but biggest issues are the svn [12:50pm] bollini: should we use the maven.dspace.org as central repo also for stable release? [12:51pm] mdiggory_: bollini: no, I feel it is best to use the maven central repo for that [12:51pm] Stuart-Lewis: Shall we do the 1.6/trunk svn rejigging before or after the move to OSL? [12:52pm] mdiggory_: Stuart-Lewis: We can do either IMO, but it will be easier on folks working on 1.6 if they don't have to setup twise [12:53pm] Stuart-Lewis: So after then? [12:53pm] Stuart-Lewis: Or as part of the move? [12:54pm] mdiggory_: pushes the issue of migration to OSL higher in priority if we say so [12:54pm] scottatm: What is the difference between maven.dspace.org and the maven central repo..... I thought that the central repo just collected material from maven.dspace.org? [12:54pm] mdiggory_: scottatm: yes, central repo rsyncs maven.dspace.org/releases [12:54pm] mdiggory_: but not snapshots [12:55pm] scottatm: ah, okay. Thanks. [12:55pm] bollini: ...but rsync has a big delay [12:55pm] mhwood: Thank you, I was unclear on why there is a maven.dspace.org at all. [12:55pm] mdiggory_: but releases do not happen often [12:55pm] bradmc: I tend to think before, although announced after. Go ahead and get SF looking how we want it to look. Then the move will be clear to anyone looking later. But make the announcement of how to work on 1.6 with a pointer to OSL after we've moved. [12:55pm] mdiggory_: the delay is acceptable IMO [12:56pm] bradmc: Last few minutes of this meeting. Any reason not to go straight to -devel with this transcript? [12:56pm] Stuart-Lewis: Fine with me [12:56pm] bollini: ok to send [12:56pm] mdiggory_: but also bollini you can customize your local maven settings.xml to use maven.dspace.org if you like [12:56pm] Stuart-Lewis: Do we want a quick straw poll on how many extra committers we think we need / can find with the time to devote? [12:57pm] mdiggory_: so before I have to leave, sounds like we need to formalize/document the process for SVN migration so we all know what that means [12:57pm] scottatm: +1 to send transcript to dev list. [12:57pm] mhwood: Transcript OK [12:57pm] mdiggory_: Stuart-Lewis: As many as it takes to maintain the projects we will all be working on. ? [12:58pm] Stuart-Lewis: Initial 1.6 survey results http://www.wordle.net/gallery/wrdl/770765/DS1.6_initial_survey_results (1=stats, 2=batch metadata editing, 3=embargoes, 4=better import/export) [12:58pm] mdiggory_: Also want to see those submitting new modules taking larger ownership/commitership stake [12:59pm] bradmc: I think the question is how many active committers we need, and clearly we're under that number by at least two at the moment. But I don't know that we'll know the right number until after we've passed it. [12:59pm] bradmc: Let's close up (keep going unofficially if you want). Thanks everyone. [12:59pm] Stuart-Lewis: So shall we all put forward a new names on -commit, and then if some names get mentioned several times we can vote on them? [1:00pm] bradmc: I think I'd start some proposals (not votes) on -commit, yes. [1:00pm] mdiggory_: Yes, thanks, will leave laptop "lurking" ------------------------------------------------------------------------------ Stay on top of everything new and different, both inside and around Java (TM) technology - register by April 22, and save $200 on the JavaOne (SM) conference, June 2-5, 2009, San Francisco. 300 plus technical and hands-on sessions. Register today. Use priority code J9JMT32. http://p.sf.net/sfu/p _______________________________________________ Dspace-devel mailing list Dspace-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/dspace-devel
