By conceptually and functionally positioning itself in a manner that
closely relates to "The Stillman Projects," the "STARRYNIGHT" project
[http://www.rhizome.org/starrynight] operates within a fairly unexplored
(but not new) area of net.art that is concerned with data mapping and
collaborative information filtering. While corporations are manipulating
the strength of the internet by employing analyses of user interaction
to shape and target internet navigation, advertisement and sales
(amazon.com, netscape/alexa - related links etc), the net.art community,
for the most part, has yet to consider the advantages of the web as a
public and smart space, yet to recognize the benefits of surveillance.
"STARRYNIGHT" hopefully represents a wider acknowledgment of these
issues.
The first "Stillman Project" was hosted by Switch in the Summer of 1997.
(http://switch.sjsu.edu/web/art.online2/stillman.links/questionsEntry.html).
It allowed the user to visually experience the selections of the other
users on the site and to base their selection on that experience. The
trace left by the user moving through the text was colored depending on
the user's answer to a question. The user's navigation showed up as an
immediate change of color of the selected link.
The next "Stillman Project" was hosted by the Art and The Military issue
of Switch (http://switch.sjsu.edu/web/v3n3/militarytoc.html). As a play
on geographical vs. networked territory, the user was assigned a color
depending on which country or type of institution they were navigating
the site from.
In November 1998 "A Stillman Project for the Walker Art Center"
[http://www.walkerart.org/stillmanIndex.html] was launched. It was
commissioned by Steve Dietz and hosted by the Walker and the Shock of
the View web sites. As a parasitic/symbiotic system, it was not allowed
to change the look of the sites by displaying the traces on each page,
but instead showed the traces on a dynamic site map. The map contained
links to all pages on the site that had been visited. The links were
organized depending on what traces of color that had been left on them
i.e. what "type" of users that had been requesting the page.
One of the interesting and important issues raised in relation to these
kinds of projects ("Stillman", "STARRYNIGHT") could be summarized by a
question Steve Dietz asked when he interviewed me in November last year:
+ + +
SD: One of the obvious aspects of collaborative filtering is that while
it is an alternative means of navigation, it can also create a kind of
self-fulfilling prophecy in which the most-visited pages are… visited
the most.
LJ: A good conversation on an interesting topic hopefully influences the
participants. It should make us change what we believe in and what we
want. We need to get used to interpreting this kind of information and
to adjusting our agendas in appropriate ways. Every day we are in
situations where we have to weigh information about what others think
and want with our own agendas and beliefs. These technologies have more
to do with consensus than with democracy or autocracy. They might make
us think and react in more similar ways. I don't think there is
necessarily any harm in that. It is again a question of the individual
versus the collective, and the approach to it is going to vary greatly
even within the cultures of the United States and Europe.
(The full interview can be found at
http://www.walkerart.org/gallery9/jevbratt/index.html)
+ + +
I decided with the first "Stillman project," in an attempt to go around
this issue, to avoid a quantitative method for generating traces of
visitors. The prototype made in 1995 was utilizing a binary scale (white
to red) to indicate requests of pages and I noticed how users tended to
select the most "hot" (red) links. This might be interesting in itself,
if interpreted as some kind of flocking behavior, but quantity is too
easily mistaken for quality.
It would be interesting to hear how the creators of "STARRYNIGHT" have
been thinking about these issues.
