[1596] in Discussion of MIT-community interests
[Mit-talk] Re: Fwd: [Jabber Project Announcement]
daemon@ATHENA.MIT.EDU (David Z Maze)
Mon Jul 18 15:30:50 2005
From: David Z Maze <dmaze@mit.edu>
To: mit-talk@mit.edu
Date: Mon, 18 Jul 2005 11:58:20 -0400
In-Reply-To: <20050718153300.GG24650@mit.edu> (Natan Cliffer's message of
"Mon, 18 Jul 2005 11:33:00 -0400")
Errors-To: mit-talk-bounces@mit.edu
Natan Cliffer <natan@MIT.EDU> writes:
> With the right amount of fudging, is there any reason why a
> jabber-based system couldn't have a wrapper that looked just like
> zephyr? I mean, you'd need to give jabber the concept of "instance"
> along with "chatroom==class", but to my totally uneducated sight it
> looks vaguely possible.
I don't know a whole lot about Jabber, beyond it being another IM
system that isn't AIM and has some traction outside MIT and CMU. I
think of the distinguishing features of (MIT) zephyr as:
-- Class/instance system allows selective (un)subscription by,
loosely, "group" and "topic" (and, locally, -c message -i 1.234
convention for academic usage);
-- Authentication of senders, along with technical barrier-to-entry of
having a Kerberos principal;
-- Multiple clients (zwgc, owl, vt, ...);
-- IS&T-supported logging of prominent classes (-c message and -c
help, in particular, in the zlogs locker).
There are other details of it that are annoying (hex-encoded numbers
in wire protocol, krb4 only), social conventions (zcrypt, personal
classes, spkng englsh lol ;)), and interesting technicalities (can't
get a list of subscribers to a triple; can subscribe to "personal"
zephyrs beyond <message,personal,%me%>; interrealm).
I'm pretty agnostic to what technology MIT uses for this sort of
thing, provided it's accessible from Athena; current political reality
probably means it's open source with Windows, *nix, and MacOS
clients. BITD I found -c message -i 6.170 and -i 6.035 very useful
resources in getting through those respective classes, and tried to
provide support when I TA'd them later on. I'd like to see, in
particular, the technical barrier-to-entry strengthened; I think the
signal-to-noise ratio is improved by having primarily MIT people
around talking about MIT-related things. Having support in the system
for "private" classes would help; that's typically done now through
tools like zcrypt, which are fairly hackish.
--dzm
_______________________________________________
MIT-talk mailing list
MIT-talk@mit.edu
http://mailman.mit.edu/mailman/listinfo/mit-talk
