<nettime> Interview with I/O/D, the Makers of Web Stalker

[Geert Lovink <geert@xs4all.nl>]

E-Interview with the makers of the Web Stalker browser
Simon Pope, Colin Green and Matthew Fuller
By Geert Lovink

Made for the First International Browserday, Amsterdam, april 17th, 1998
For more information: http://www.waag.org/
The Web Stalker: http://www.backspace.org/iod

GL: 'Everybody is a browser designer' - but it is not everyman's hobby to
build one (yet). Where does the idea, to create one's own browser, come
from? Normally, designers are working with content and have to make it
look nice. But now there is the new profession of the 'interaction
designer'.  Are you one of those? Are you techno determinists, who believe
that the shape of the interfaces is determining the actual information? 

Matthew Fuller:  Hmm, this is one of those statements along the lines of
'Jederman ist ein kunstler'. (Joseph Beuys) These statements sound
democratic, but actually have the subtext of meaning *Everyone wants to be
like me - the great man!*
No, not everyone is a browser designer for sure. And certainly it would be
unwise to want to be like us. People should actually have aspirations
right? The idea of making another piece of software to use the web with
came about for a few reasons. First of all, I/O/D had been working with
different ideas of interface and a general praxis around speculative
reinvention of the computer anyway. Secondly, we were bored by all the
hype. Thirdly, we knew it could be done, but didn't have the skills of the
knowledge to do it properly - so we had to do it. As for the normal
behaviour of designers I reckon I'll leave that part of the question for
Simon or Colin to answer with a firmer grip on the handle of the knife that
needs twisting.
As for being techno determinists? I guess we are interested in finding
this out. What comes into play using the web? The material on the URL
being used, which encompasses the programs, skills and materials used to
put it together as well as the specific items of data; then the actual
hard infrastructure - computers, servers, telephone lines, modems and of
course the software running on them, (in short, bandwidth considerations); 
then the software being used to access the web - a great big pile on top
of which sits the Browser, terminal viewer or whatever. All of these
elements and how they mix determine to some extent the nature of the
interaction.
For instance, try using a web site packed full of java-scripts, frames and
vrml with a browser from a couple of years back. You'll find that the
type of interaction available to you is pretty much fully determined by
the technology you have. You're locked out.  On the other hand, just
looking at all of this misses out on the key piece of equipment in the
relationship - the user. One of the things that drove us to make the Web
Stalker was that we, and pretty much everyone else don't really use
web-sites in the way that they are suposed to be used. Whether it's
switching off gifs or blocking cookies or whatever there's an element of
street knowledge that you use to get to the stuff that you really want. We
made the Web Stalker to work in the same kind of way. It's designed to be
predatory and boredom-intolerant. At the same time though, we hope that as
a piece of *speculative software* it just encourages people to treat the
net as a space for re-invention.

Geert Lovink: Web Stalker is showing us the backstage of the browers. 
Could you explain us how it actually works? What kind of code do we get to
see? Is it just HTML or hidden directories of the servers? What do
webmasters and sysops try to hide for us and what can we learn from it?
Web Stalker as a hackers tool for extra-governmental gangs that are trying
to undermine the effeciency of global capitalism? 

Simon Pope:  the web stalker moves only within the limits of html space.  any
co-conspirators needs to be fore-armed with at least one URL which refers
to an html document. give this to the 'crawler', and the stalker begins its
process of parsing, hungrily searching for links to other html resources.
initiating a 'map' window, opens a channel onto this  process, through
which urls are graphically represented as circles and links as lines. the
stalker will thrive on known links and resources - as long as each html
document contains a link to another html document, the stalker will live.
pitch it into a netscape, microsoft, macromedia or java-only space and it
will soon perish.

Colin Green: When we began to use the stalker as our primary web-access
software, we became aware of the extent to which html has become a site of
commercial contention. Browsers made by the two best-know players frame
most peoples' experience of the web. This is a literal framing. whatever
happens within the window of explorer, for instance, is the limit of
possibility. HTML is, after-all, a mark-up language which indicates
structure and intention of a document. There is no imperative to interpret
<cite> as <italic>, as there are none which demand the use of 'forward' or
'back' to define a spatial metaphor. 

Matthew Fuller:  We've had reports from users that amongst other things,
if you use the Web Stalker on a site with extra content being added to it
every few hours, such as some news services for instance, you can start to
find files whilst they're still in the queue - before the news 'happens'.

Simon Pope: Commercial interests have tried to exploit the web by
controlling the velocity of browsing. the stalker subverts this - it
confounds the faux-melodrama of the click-thru by automatically making the
link for you. Suspense is ridiculed and fluidity is returned to a realm
where processes of delay and damming are recognized advertising
opportunities. It is here that the convention of the "web page" helps to
solidify html, presenting each document as the potential apex of the
user's experience. A leaf-node rather than link. 

Geert Lovink: but is the web stalker not also a bit protestant, in the
sense of anti-image - pro code? HTML and the WWW are being presented to us
as the big step forward for the normal user, to have an easy-to-use
interface. what is so disgusting about all these fancy websites, funny
graphics and sexy buttons? isn't the stalker a bit step back, very male
and hackerlike in its approach? i don't say that the explorer is female...

Matthew Fuller: The Web Stalker establishes that there are other potential
cultures of use for the web.  The aesthetic conventions of current Browsers
are based on the discipline of Human Computer Interface Design. To
describe the predelictions of this approach to interface you only have to
note that the default background colour in page-construction programs is
grey.  Progress is marked by the incremental increase of fake drop-shadow
on windows.  Here, the normal user is only ever the normalised user.  Time
to mutate.
For us, software must also develop some kind of relationship to
beauty. This can in one sense be taken as something that only happens in
the eyes. But it is also something that happens at a level that is also
profoundly interwoven with politics in the development of these potential
cultures of use. It is in this sense that we call The Web Stalker
'speculative' software.  It is not setting itself as a universal device, a
proprietary switching system for the general intelligence, but a sensorium
- a mode of sensing, knowing and doing on the web that makes its
propensities  - and as importantly, some at least of those 'of the web'
that were hitherto hidden - clear.
Rather than taking an ascetic view we see that a key problem with
the Browsers is that they don't allow the Spew to manifest itself *enough*.
This software is a call for the voluptuation of the nets and everything
they connect to.  As the union leader Big Bill Heywood used to say,
stroking his belly and sucking on a tasty dog-shit-sized cigar: Nothing's
too good for the proletariat.

Geert Lovink: After having done Web Stalker, what is the relation between
the small, arty, conceptuals anti-browsers and a perhaps more serious one
that will be free public domain software? It is maybe hard to estimate how
influencial marginal autonomous software production actually this. There
are many different estimations about this. How do you see the Amsterdam
effort of the 'International Browser Day' in all this? 

Mattew Fuller:  The Web Stalker proposes another model alongside the two
other main models of radical software production. The first is obviously
that of Free Software.  The second is that of programmers working in
collaboration with specific client groups whose needs are not met by the
programs developed in a 'free' market.  A good example of this is the
icon-based email program being put together in de Waag. Both of these
models are based on a specific or wider consensus. The Web Stalker
proposes a complementary model, one that is interventional. That is
designed specifically to make a far reaching breach into the material and
imaginal space of the technical and social context in which it is placed. 

Simon Pope: Until recently,there were few points in the development of pc
software where source code was opened-up to end-users where applications
could be modified or extended. With Netscape's recent announcement, at
least there is now an awareness of the existence of this type of
development, even if the take-up by end-users (rather than developers)
might not be that widespread.

Colin Green: We develop software from a very specific position: Lingo has
been our language of choice and from necessity for the past 5 years.
During that time, there has been a gradual shift in the method of
programming, from proceedural to object-oriented approaches. This change
happened as much through an ad hoc engagement with Lingo by frustrated
users than from the imposition of methodology from another programming
language. The result has been that there is no standard way to deal with
Lingo, so it's not been practical to share sourcecode - it takes too much
time & effort to decipher someone elses scripting. The days of being able
to get away with cut & paste of other peoples scripts are over - nothing
interesting came out of that approach anyhow... 

Simon Pope: Also, there has been no real percieved benefit in giving away
Lingo scripts. If you can write good enough code to be able to give it
away, there's probably very little out there you actually WANT in return.
This is changing. Once-novice coders are now gaining in confidence and
turning-out software with the intention for others to use it, tear it
apart and rebuild it according to their own design.  We'll open-up the
back of next software project to expose it to this kind of developoment. 

Matthew Fuller: For us, the Browserday is a very useful initiative.  Once
the breach has been made, proving that the net can be used and develeoped
in ways largely at variance with the proprietary browsers and the
interests they maintain, the floodgates can - potentially - open.  A
thousand different net sensoriums can be launched.  The Browserday is
important because it was done in a way that was at once informed by both
technique and theory without priviliging either and done in a populist
celebratory manner.  It's not just done to force the didactic proof that
software can be -exciting- but also that people can make actual, rather
than virtual, reconfigurations of ways of seeing, knowing and doing.  And
some of the wild stuff that the students came up with!!!  In this alone it
went beyond the usual dreary parade of technoculture events that people
have become accustomed to. 

---
#  distributed via nettime-l : no commercial use without permission
#  <nettime> is a closed moderated mailinglist for net criticism,
#  collaborative text filtering and cultural politics of the nets
#  more info: majordomo@desk.nl and "info nettime-l" in the msg body
#  URL: http://www.desk.nl/~nettime/  contact: nettime-owner@desk.nl