I guess I'll introduce myself here: I'm Francis Hwang, the Director of
Technology at Rhizome.org, which means I'm largely responsible for our
technical strategy and implementation. My work at Rhizome intersects
with this month's topic in regards to the ArtBase, which is Rhizome's
online archive of new media art. It contains more than 1400 works
collected over the course of about 7 years, and as such is the largest
archive of its kind.

Kevin McGarry is also in this discussion, and he deals with the ArtBase
on a more day-to-day basis, but I'll throw in two thoughts here that
may or may not be interesting to others:

1. As an arts institution without a physical presence, Rhizome has
always run on a fairly small budget and as such we've continually tried
to make ArtBase's daily operation cheaper as it grows. Originally much
of the information gathering was done by a Rhizome employee or intern,
but these days a lot of it is overseen directly by the artist herself,
with somebody inside of Rhizome approving and shepherding the process
at certain steps. We're currently looking into other ways to make the
ArtBase even less dependent on staff resources. Of course, this has
significant up-front costs. Your software has to get more robust as you
involve more people, and you have to spend a lot of time thinking about
other people and who might find this sort of volunteer work in their
own self-interest.

2. Continuing in the "cheaper is better" theme, I've been watching the
world of social software is it discusses and experiments with
"folksonomies". This is one of the most fascinating mini-trends in the
social software world right now; if you want to see it in action you
might consult http://del.icio.us/ or http://flickr.com/ . Roughly
speaking, folksonomies are systems that allow users to tag resources
(bookmarks, for del.icio.us, and photos for flickr) with their own tags
for their own use, and then allow others to share, compare, or collate
tags. These are really sloppy, of course, since New York City might
show up as "newyorkcity" or "nyc" or "new_york_city" or whatever, and
there's no automatic way to make sure you get them all. On the other
hand, such systems are easy and fast, so they produce a lot of
metadata, sloppy as it may be: as of this writing it looks like
del.icio.us users have posted 27 links in the past 60 seconds. This is
a pretty stark contrast with the more traditional, archival view of
categorization; that it involves a lot of forethought and would never
be done for free. Whether such systems can be of use in more focused
fields such as fine arts remains to be seen, but I predict we're going
to see some pretty interesting experiments in the next few years.

Francis Hwang
Director of Technology
Rhizome.org
phone: 212-219-1288x202
AIM: francisrhizome
+ + +