Below is the transcript of a committer's meeting held on #dspace at irc.freenode.net at 16:00 GMT on May 7, 2009.
We intend to hold the next meeting on #dspace May 13 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 spent a few minutes on OR2009 logistics; endured an IRC failure, and just started on how to extend the existing event system, and considering an external workflow framework. Transcript: bradmc: Hello folks, we'll start in a couple of minutes. What topics should we consider today? [12:01pm] mhwood: I see these left undiscussed from last week: GSOC, OR09, 1.5.2. Sorry, no more context on my notepad. [12:04pm] bollini: If we have enought time I like to hear some opinions about best practise using the event system [12:05pm] bollini: see Larry comment on DS-202 http://jira.dspace.org/jira/browse/DS-202?focusedCommentId=10306&page=com.atlassian.jira.plugin.system.issuetabpanels%3Acomment-tabpanel#action_10306 [12:07pm] bradmc: mhwood: yes, those were my seed topics last week. On OR09, I wanted to get some feedback on who will be there, and solicit commitments to be in the room on Thursday afternoon during the DSpace workshop session. [12:09pm] StuartLewis_: Is there a meeting today? Or have I messed up my dates/times? [12:09pm] bradmc: StuartLewis_: Yes, it's just quite quiet thus far. [12:10pm] StuartLewis_: OK ? [12:10pm] mhwood: There was a request for suggested topics. [12:10pm] StuartLewis_: A brief update on the status of 2.0 work? [12:10pm] bradmc: bollini is interested in discussing practises for using the event system, see comments on DS-202 (Jira) [12:11pm] bollini: bradmc: btw we need to add more "modules" on JIRA: at the moment we need at least SWORD (DS-199 & DS-200) [12:12pm] bradmc: bollini: Okay. [12:12pm] tdonohue: well, should we touch on OR09 first, as that's coming up soon? [12:13pm] bradmc: Sure. I know you'll be there ? [12:13pm] mhwood: I'm going. [12:13pm] tdonohue: yes, i will ? I'm also willing to be on the panel during the DSpace workshop session (as I had mentioned over email) [12:14pm] bollini: Sorry, I will be not present [12:14pm] StuartLewis_: Me neither this time ? [12:15pm] bradmc: The workshop session is open to all; first half will focus on 1.x issues, second on 2.x issues. We're hoping that any committers that are at OR will come to a least part of the session and participate. There are some conflicts with other tracks, so we're aware that people will slide in and out. [12:15pm] grahamtriggs: I'll be there [12:15pm] bradmc: Tim, anything you want to discuss vis-a-vis your presentation? [12:15pm] ben_atmire: I will be present, and so will Mark Diggory [12:16pm] bollini: are there possibilities to get a video recording? [12:16pm] bradmc: We're also expecting that there will be a BOF session after the DSpace presentations earlier in the week. [12:16pm] bradmc: There is slated to be video, yes. [12:16pm] bradmc: Assuming all works well, I think you'll be able to see it. [12:17pm] bollini: good news [12:18pm] tdonohue: bradmc: Um...yea, I guess I wasn't fully aware I was running the first half of the workshop ? At least until I saw the final schedule...is there anything in particular folks want me to cover for the 1.x issues? [12:19pm] tdonohue: I was planning on concentrating more on customization/configuration of 1.5 + XMLUI + Maven (as that's what I know the best) [12:20pm] bradmc: That's fine, Tim, we don't expect you to fill the first half yourself; we expect interaction from the floor. [12:20pm] tdonohue: oh, sure...i was planning on that...I'm wanting to make it interactive and lots of time for Q&A (similar to what happened with Graham & my talk at OR08 last year) [12:21pm] mhwood: The area where documentation is thinnest is the organization of the stock Theme transforms, I think. [12:21pm] bradmc: That sounds like the right approach. [12:23pm] tdonohue: well, then i think i can take things from there....I'll give a few good examples of customizations...and then try to open things up.... One final question, are we attempting a relatively even split (45 mins on 1.x, and 45 mins on 2.x)? [12:23pm] bradmc: Actually, the workshop is 90 min 1.x, 90 min 2.x [12:23pm] bradmc: (That's thursday) [12:23pm] tdonohue: oh...i missed that ? ok...even longer than I expected ? [12:24pm] bradmc: bollini: there is now a category of "SWORD" under dspace 1.x on the Jira. [12:24pm] tdonohue: so, are we pulling up a panel of committers for the 1.x talk as well? (i.e. should I work that end at the end of my talk) [12:24pm] tdonohue: bradmc: We can also just chat offline about this...so I don't hijack this meeting [12:26pm] bradmc: Okay, thanks, Tim. [12:26pm] bradmc: Other topics for general discussion today? Shall we pick up the event system question? [12:28pm] mhwood: Yes? [12:29pm] bollini: I'm assuming that we can go ahead with event system... [12:29pm] bollini: I'm thinking to it as the main opportunities to make dspace "pluggable" without core modification [12:30pm] scottatm: The event system is not a silver bullet to solve the problem of pluggability, but it is one component that can be used to solve the problem. [12:30pm] mhwood: It seems to be intended specifically to expose changes to the content. That leaves a lot uncovered. [12:33pm] bollini: I agree with Larry, if we change in a consumer the model at the moment it is really easy to create infinite loop. but... what do you think, should this use case covered? [12:34pm] mhwood: I found that I had to add another, unrelated event stream for sensing *accesses* of the content. There was resistance to extending the scope of the existing stream. Looking back, I think that was correct. [12:35pm] scottatm: Seems like the consumer can be coded to prevent the case of recursion? [12:36pm] bollini: scottatm: should be the consumer or the event system to take care of this? [12:43pm] bradmc: Looks like an IRC failure. [12:44pm] bradmc: Appears we had an IRC failure. Last message was from bollini: scottatm: should be the consumer or the event system to take care of this? [12:46pm] scottatm: bollini: The consumer [12:46pm] scottatm: mhwood: the event system could keep track of events generated from other events, some kind of linking system between events. But i don't necessarily think its needed. [12:46pm] mhwood: It sounds like solving a problem we should instead just avoid creating. [12:47pm] bollini: sorry, I have had an issues with IRC... I have missed the last 4-5minutes messages [12:47pm] tdonohue: and now everyone's suddenly back...what happened? [12:48pm] mhwood: leguin.freenode.net bounced? [12:48pm] bradmc: Looks that way. [12:48pm] tdonohue: bollini: I don't think much has happened in last 4-5 mins...most everyone was kicked out (with the exception of about 3 of us) of IRC [12:49pm] mhwood: If you received "bollini: The consumer" then you missed nothing that I know of. scottatm and I repeated our remarks. Anybody else? [12:51pm] mhwood: So: is it better to expand the mission of the event system, or to add more plugin points? [12:51pm] mhwood: Or add hair to the event system to (try to) prevent looping? [12:53pm] bollini: I think that we need to follow both strategies... add more plugin points and add more event [12:54pm] bollini: what is the plan for the 2.0? should we have a Workflow Engine? [12:55pm] bradmc: We don't have a firm plan around a Workflow Engine. Does anyone have a firm recommendation on an appropriate one? It's a pretty fluid space right now. [12:55pm] mhwood: We have one now, don't we? Custom-built for DSpace. Do you mean, should it be replaced with a more general one that we can just adopt? [12:56pm] bollini: mhwood: of course, we need to move on an external framework... something like Jbpm [12:58pm] bollini: my concern is about some overlap that we could have between workflow engine and our event system [12:59pm] mhwood: My understanding is that, at the moment, the event system fires when a submission is exiting the workflow, but it's been some time since I looked at it. [12:59pm] bollini: the event that I suggest to add in DS-202 is pretty related to a workflow transaction [1:00pm] bradmc: bollini: I think that any integration of an external framework will inevitably result in having to rework the current event system. That's not a reason to stop extending the current system for the time being. [1:01pm] bollini: mhwood: no we don't have an event for this, at the moment we have only content related event [1:04pm] bollini: what do you think if us open an issue on JIRA about adding more type of events and ask to vote (developer list) for it? [1:05pm] bradmc: I think that's a good proposal. [1:05pm] bradmc: We're past the 60 minute mark. Should we wind down? [1:05pm] bradmc: Note: Next week I'll be in transit during the chat time. Will someone volunteer to record the discussion, please? [1:06pm] StuartLewis_: I can do if you want? [1:06pm] mhwood: I can do that. [1:08pm] bradmc: Thanks, guys. Between the two, we should end up with a copy ? [1:08pm] StuartLewis_: ok [1:10pm] mhwood: It's on my calendar. [1:12pm] bollini: I will open an issue on JIRA about the event system... [1:12pm] bollini: are we closing? [1:13pm] mhwood: It seems so. It's very quiet here. Are there other issues? [1:14pm] bollini: nothing that can't wait until next week... [1:15pm] bollini: bye... [1:15pm] StuartLewis_: bye [1:16pm] mhwood: 'bye all. [resumes lurking] [1:16pm] bradmc: Thanks all, bye. ------------------------------------------------------------------------------ The NEW KODAK i700 Series Scanners deliver under ANY circumstances! Your production scanning environment may not be a perfect world - but thanks to Kodak, there's a perfect scanner to get the job done! With the NEW KODAK i700 Series Scanner you'll get full speed at 300 dpi even with all image processing features enabled. http://p.sf.net/sfu/kodak-com _______________________________________________ Dspace-devel mailing list Dspace-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/dspace-devel
