[-empyre-] building better list software

["Jim Andrews" <jim@vispo.com>]

have been enjoying the discussion of lists; i'd like to introduce an
additional thread in the discussion about lists that can coexist while the
other threads are going on.

occassionally over the last couple of years i've wondered about how listserv
technology could be improved. my own experience is mostly on digital art
related lists. so it's really that sort of forum i'm mainly thinking of.

listserve software has been around for a long time. pre-web. which indicates
it has served well. lists are remarkable in the way they facilitate group
communication. before lists, i imagine that only 'letters to the editor' in
print publications had a similar function. and that sort of forum is
typically low volume, typically edited for space, and completely controlled
by the publisher. or radio talk-shows, say, but everyone has to talk with
the host; they don't have the opportunity to talk with one another.

it's also remarkable how little listserv software has changed over the
years.

the question occurs to me what sort of room there is to create better
listserv software. which prompts the question of what the drawbacks of lists
are.

several come to mind. of course feel free to add/revise.

the lists on the net devoted to discussion of digital art-related issues are
not legion, really. there are probably about two to six dozen, i imagine.
though very few of us would be able to name four dozen lists. so one of the
problems is simply knowing what lists exist and what they're about and the
qualities (and quantities) of them.

of course this sort of information could be got from an html page. but it
could also be got via an XML feed and displayed in a software/art
application, much like the list of available newsgroups is downloadable from
a newsreader client. as new newsgroups are created, the list increases in
size. together with links to information about the lists.

this application could be a plugin to existing email clients, so that people
would not have to change their email client. writing plugins to some email
clients such as Outlook is well-documented and the tools are available. Not
sure about other email clients.

what else could this application do?

one of the main problems with lists is that as they grow in
volume/membership and diversity of subject matter, there are few options for
managing the positive possibilities and also the problems that naturally
arise owing to the growth of a successful forum. the size of the 'room'
doesn't change as the members pile in and, eventually, the list becomes more
or less unreadable, unfollowable in an unmoderated, previously successful
forum.

also, in digital art, different types of artists vie for attention to their
particular type of art or their own particular art, and so we sometimes see
subgroups on lists vieing for control of lists if, by no other means, sheer
volume of posts.

part of the problem is that lists are, architecturally, limited to one
folder. everyone is always in the same room or folder, which can be
claustrophobic, as it were. some web-based forums allow for the creation of
new folders (and related features), but these are web-based forums. i would
say the feature-set of such forums is admirable. but the interfaces are
clunky and you have to go visit the forum rather than it being integrated in
the existing email client. they are too 'pull'-oriented. you have to pull
too hard to get the information and reply etc. the apps are slow and clunky
compared to email clients. but the feature-set is admirable.

sometimes it happens that when non-web-based lists grow, they spawn other
lists. necessarily separate lists. a multiple-folder architecture would
allow the lists to coexist simply via the creation of another folder to the
list structure.

another list problem is that it is currently too hard to filter out messages
from people you don't want to hear from. it's easy to do this when receiving
email from individuals, but not when they post to lists. it should be easier
to filter people out you don't want to read.

if this were the case, list managers of unmoderated lists would have less
occassion to worry about the signal-to-noise ratio. the easier it is for the
members to set their own signal-to-noise ratio, the less the list owners
have to worry about it.

another problem with lists is that promising and important discussions peter
out. rarely are topics pursued with sustained vigor to the duration the
subject requires. this is partly owing to the time-based nature of the
forum. things are half-hashed out and then people are on to another subject.

part of the solution may be to implement features that make it possible to
cultivate the *publication* while not killing the time-based aspect of
listserv forums. publications need more time to be developed. it would be
cool to have features available to the list owners and the membership that
allow them to create folders for publications. when a topic proves to be
popular and there are thoughtful posts on the matter, they could be copied
to a new folder the purpose of which is to develop something like a
publication on the subject. so a 'thread' can develop over a longer time.

this also addresses another problem of lists. the time-based nature of lists
makes posts pretty evanescent. but lots of people put real effort into their
posts and such posts deserve not to disappear so quickly into the archive
(which usually isn't really nicely searchable). if there was a *single*
interface by which all of the above sorts of issues could be addressed, as a
plugin to existing email clients, i think that would make for better forums.

the above sort of project strikes me as quite a useful one as a project in
"social software", though it is daunting in its scope. best to regard it
hypothetically. there are, of course, many other issues such an app could
address. other features occur to me but i've written more than enough in
this post. it might be interesting to see if people could suggest useful
features to such an app/plugin to existing email clients. what are the main
problems with lists, as you see them, and do you think there are software
solutions (or if not 'solutions', then 'obviators' or whatever)? and what do
you think of the multiple-folder idea?

ja
http://vispo.com