 |
 |
|
 |
||||||
Projects .LRN Project Site .LRN TAB TAB Chat 2003-01-21
We are planning to try to get out an OpenACS 4.6.1 maintenance release with a beta Monday (2003-01-27) and a final on Friday (2003-01-31). It's not really a TAB thing but mostly a convenient venue to hear if anyone had any issues would block it (and none of us really did).
We agreed that we should do some background reading and respond in the thread in which Michael Feldstein said:
What I am saying is that the three standards themselves are designed to overlap and interoperate. Therefore, our efforts to implement the three standards should also overlap and interoperate.
To my mind, this is just the sort of situation that the TAB is meant to facilitate. What we have is several simultaneous and independent projects where the whole could be greater than the sum of the parts. I suggest that the leads on the three projects at least touch base with Don to help ensure that their implementations are complementary. The benefit to doing so for all the various individual projects are (1) you get excellent technical guidance, (2) you get help in finding out what other people are building that could make your own development project easier, (3) you get higher visibility and therefore higher likelihood of attracting funding, and (4) you enhance your reputation as being good members of the community.
Dan Wickstrom quickly covered what he had done with uPortal and we had a short discussion about how we might move things forward there.
We talked about the OpenACS chat and Don mentioned he would turn up for this weeks (which he did and which went well again).
There is a forum thread for followup discussions
Jan 21 17:07:39 <donb>
I think we have two major agenda items ... one is 4.6.1 which we're discussing here because the Jeff and Don secret, insane cabal are the two main co-plotters behind this maintenance release and it's convenient to talk about it here
Jan 21 17:08:02 <jcdldn>
I have exchanged some mail with Bart and talked about what the baseline for testing should be (4.6.1 release if soon enough, beta if not).
Jan 21 17:08:45 <donb>
I liked your time line of trying to roll a quick 4.6.1 beta by the end of the week. I need to move on to other stuff so the sooner the better IMO
Jan 21 17:09:07 <jcdldn>
I did get the 7.3 patches committed but have not done an install. In an insanely ambitious too many moving parts effort I am going to attempt to install 4.6.1 with AOLServer 4.0 on 7.3.1.
Jan 21 17:11:11 <lars>
BTW, we're using automated testing for workflow, and it's *really* nice to see that everything still works after a refactoring
Jan 21 17:12:39 <donb>
Lars ... are your automated tests in CVS along with everyone else?
Jan 21 17:12:47 <lars>
Yes, they're in CVS
Jan 21 17:13:02 <lars>
(although we have a ton of uncommitted stuff right now)
Jan 21 17:16:50 <jcdldn>
I think that is the only critical thing, I do want to get the file-storage fs_maybe_create_new_mime_type fiasco cleaned up.
Jan 21 17:17:08 <jcdldn>
Since it comes up more than anything else.
Jan 21 17:17:31 <donb>
file storage is flat-out unscalable as it exists but that won't get fixed until 4.7
Jan 21 17:18:06 <jcdldn>
Oh, and the thing Lars ticketed about not invalidating perm cookies when passwords are updated seems like a reasonably serious issue.
Jan 21 17:18:55 <donb>
A sidebar mini-issue ... folks can have my "populate" mini-package if they want. I don't think it really belongs in our main CVS tree because it's not really meant to be general, I just picked some object types I was interested in and wrote scripts to insert them
Jan 21 17:19:36 <danw>
Maybe you could put it in file-storage.
Jan 21 17:19:39 <donb>
Jeff may want to add his "fill up the forums package" script to it, too ...
Jan 21 17:19:43 <donb>
I was thinking of that, Dan
Jan 21 17:20:24 <jcdldn>
I would like to put it into CVS, maybe in a "contrib" project to provide a home for the homeless.
Jan 21 17:20:25 <donb>
put it in file storage and post a pointer. what I should really do is get my own server organized like lars and Jeff have done
Jan 21 17:20:25 <lars>
who about putting it in the openacs-tools module
Jan 21 17:20:31 <lars>
there's no reason to not have it under CVS
Jan 21 17:21:08 <jcdldn>
I would put up some things there too (like the qa and cvs browser I have on my site).
Jan 21 17:21:10 <donb>
I'm OK with either of those. And I'd just put up a "contrib" download repository for folks to use except for the fact that I think download needs a little work first
Jan 21 17:21:58 <donb>
Maybe a "contrib" project for more general stuff like Jeff's mentioning and my "populate" package under openacs-tools since it's specific to oacs 4?
Jan 21 17:21:59 <jcdldn>
yeah. embarrassing bmomjian could not download the code because we suck.
Jan 21 17:22:18 <jcdldn>
(re download)
Jan 21 17:22:18 <donb>
neither dave bauer or I could reproduce that BTW
Jan 21 17:22:40 <donb>
we both got progress bars and successful downloads
Jan 21 17:22:47 <danw>
Might be an AOLServer problem.
Jan 21 17:23:09 <danw>
I think this used to come up from time to time for aD as well.
Jan 21 17:23:10 <jcdldn>
I've seen it before. It's flakiness with the ns_writefp I think.
Jan 21 17:23:42 <donb>
The log excerpt Dave passed me is exactly what you get when the socket dies in the middle of things
Jan 21 17:23:52 <jcdldn>
4.6 has ns_returnfile for file system stored stuff which works much better.
Jan 21 17:24:36 <donb>
I still need to check in my "file copy" changes to HEAD ...
Jan 21 17:24:40 <danw>
You would think the two procs would have the same implementation on the socket end of things.
Jan 21 17:25:11 <donb>
The socket stuff's generic
Jan 21 17:25:27 <jcdldn>
ns_returnfile mmaps and writes, ns_writefp does read, write.
Jan 21 17:25:43 <donb>
I think Dave was thinking Bruce may've hit "cancel" because he wasn't seeing a progress bar or any other indication that the download was working. That will kill the socket and cause the error
Jan 21 17:26:15 <danw>
ns_returnfile is more efficient then.
Jan 21 17:27:53 <jcdldn>
maybe we should cut the beta sooner (like Monday) and plan on 4.6.1 final Friday. There is only one data model change right now (that I know of).
Jan 21 17:28:30 <lars>
There'll be one more, small one, to acs-install.sh: Removing acs-workflow from 4.6 as well
Jan 21 17:29:49 <jcdldn>
ah right. Lars, will you make that change?
Jan 21 17:29:58 <lars>
Yes, should only take a few minutes
Jan 21 17:30:40 <donb>
I think we're set for 4.6.1 ...
Jan 21 17:31:09 <jcdldn>
excellent. I will post something on forums tonight and update the status page.
Jan 21 17:30:55 <donb>
SCORM ... sounds like some alien sci-fi creature
Jan 21 17:31:17 <danw>
or a rare tropical disease...
Jan 21 17:31:42 <jcdldn>
donb: do you know your way around this acronym hell?
Jan 21 17:32:07 <donb>
No, I just did a google search and haven't found the definitive fount of knowledge yet, either
Jan 21 17:32:36 <donb>
Time to do some research, though, that's clear ...
Jan 21 17:32:59 <donb>
Also it seems that Sloan's funding Caroline and/or friends to do some related stuff
Jan 21 17:33:23 <jcdldn>
We (Don?) should respond to the post MichaelF made...
Jan 21 17:33:26 <donb>
I plan to read the relevant forum threads closely today ... have any of you been following this stuff closely?
Jan 21 17:33:44 <donb>
Yes, I'm planning to respond ...
Jan 21 17:33:48 <lars>
I've read the threads, but none of the linked documentation
Jan 21 17:33:50 <jcdldn>
I have read it but it was all a maze of twisty passages to me.
Jan 21 17:34:03 <donb>
yeah, a maze ... that's about it
Jan 21 17:40:23 <jcdldn>
I think Staffan's point about funding smaller projects is a good, although not in our province. It would be nice if there were some more guidance from Al et al on possibilities for getting funding for small projects.
Jan 21 17:40:41 <danw>
yes that would be interesting.
Jan 21 17:41:45 <donb>
Well ... Al needs to get cracking on the EAB, the consortium, etc
Jan 21 17:42:17 <donb>
we (TAB) are actually in the lead as far as getting organized now, who woulda thunk it?
Jan 21 17:45:50 <donb>
I think we're all going to have to educate ourselves. I'll start digging through the reference docs. I do also need to poach the Jakarta decision making doc for a rough draft of our TAB decision making doc. So I don't think I'll be doing much programming for dotLRN/OpenACS this week
Jan 21 17:46:44 <donb>
My impression is that the SCORM-ish stuff is mostly of interest to the e-learning world vs. the "support the classroom" target that's been dotLRN's primary focus thus far?
Jan 21 17:47:01 <donb>
Does that sound right? (knowing that none of us really knows)
Jan 21 17:47:04 <jcdldn>
Is Michael the best to look to in terms of what the substantive issues are with the various standards?
Jan 21 17:47:16 <danw>
He seems to have written them.
Jan 21 17:47:40 <donb>
He's probably best in terms of understanding how the political worlds collide over them as opposed to understanding them technically
Jan 21 17:47:56 <donb>
Michael does high-end consulting in this realm
Jan 21 17:48:42 <jcdldn>
Is the EAB/UAB chosen and public somewhere? I just know a few names. [Ed: State is described at http://openacs.org/projects/dotlrn/governance]
Jan 21 17:49:13 <donb>
Partially chosen, not organized, not fleshed out
Jan 21 17:34:09 <jcdldn>
danw: you looked at the uPortal stuff right?
Jan 21 17:34:15 <danw>
Yes.
Jan 21 17:34:35 <danw>
I've been talking with Al about it.
Jan 21 17:35:03 <donb>
Can you summarize?
Jan 21 17:35:10 <danw>
sure.
Jan 21 17:35:21 <danw>
Al wants uPortal integration with dotLRN and I'm going to look into provide a soap service interface to uPortal. dotLRN side will be Tcl and the uPortal side will be java probably apache axis.
Jan 21 17:36:37 <danw>
Nothing is finalized yet.
Jan 21 17:37:49 <donb>
Interesting ... I'd poked Al some time ago about the desirability of getting you to get our soap act together, thinking ns_java at the time, didn't realize you guys were discussing in such detail.
Jan 21 17:38:15 <danw>
He just posted over the week-end, so I wrote him privately.
Jan 21 17:38:21 <donb>
Ah ...
Jan 21 17:38:42 <danw>
Back in Nov. I posted that I might work on it, but nobody seemed interested, so I moved on.
Jan 21 17:39:04 <donb>
I'll have to go look at that post, too, and give him encouragement.
Jan 21 17:39:09 <lars>
Well, I think there's significant interest in seeing that happen today.
Jan 21 17:39:19 <danw>
I guess Al only reads the bboards from time to time...
Jan 21 17:40:09 <donb>
He's just now getting back to having time for dotlrn
Jan 21 17:56:04 <jcdldn>
the openacs weekly chat went ok. I think they missed some of you though :)
Jan 21 17:56:20 <jcdldn>
I am still waiting to hear from Nathan on the email.
Jan 21 17:56:21 <lars>
Yeah, it was 9 pm!
Jan 21 17:56:39 <jcdldn>
I know. It works well for me since kids are asleep.
Jan 21 17:56:37 <danw>
how many people attended the chat?
Jan 21 17:56:45 <jcdldn>
I think about 25 or so.
Jan 21 17:57:04 <donb>
Who seems to be the primary organizer?
Jan 21 17:57:04 <lars>
ah, see, we were out drinking with Carl :)
Jan 21 17:57:12 <jcdldn>
rbm and davb in and out. Mostly Bart and I answering questions.
Jan 21 17:57:32 <jcdldn>
Bart made a real effort to lead (which worked well)
Jan 21 17:57:44 <donb>
Good, I like Bart.
Jan 21 17:58:22 <donb>
I'll try to take part this thursday
