<nettime> WebTracer interview

[matthew fuller <matt@axia.demon.co.uk>]

Interview with  Tom Betts /NullPointer

NullPointer has recently released a beta-version of a new web
visualisation application, WebTracer.  Downloadable from:

http://www.nullpointer.co.uk/-/tracer.htm



MF:  What are the questions you are asking about the structure of the web,
and about the software that is being developed to use it that suggest the
approaches to it deployed in WebTracer?

NP:  Well aware of the legacy of webmapping as a supposed demystifying
device and fetishised formalistic perversion of form I do not intend to
decorate this project with too much hypothesis of cultural and social
intent. (there are others who could grace it much better than myself)
However I cannot deny that the intentions of the application are not
primarily to aid webmasters in their analysis and development of their own
sites but to, as i hope is obvious, repurpose the information that
comprises hypertext and the web into another plane of perspective and
interaction.

The application deals with sites and pages as molecules and atoms, the
resulting cellular structures reflect the information structures of the
web. I find that the representation of the many shells and layers that
guide our exploration and expliotation of cyberspace can help to reinforce
the awareness that all information systems are guided by a great number of
defining elements. The Hardware used, the Operating System, the Software,
the Network Protocols and finally the File Structures themselves all mould
the way that users interact with dataspaces and the way that they can
create them.


MF:  When you use the software it is clear that the arrangement of the
relations between the nodes carries information in terms of the length of
the linking line.  What determines the magnitude of displacement from one
node to the next, ie, how can a user 'read' the information that the
software displays spatially?

NP:  The molecular structures created by the application are arranged
spacially in terms of several different modifiers. The program uses both
the order of links as they appear on a page and the relative depth of
links within the host webserver's html docs directory. The closer a node
is to the base of a WebTracer structure the closer that page lies to the
index page of the whole site, additional subdirectories create distinct
planes that are positioned up across the vertical axis. Hence sites with
strict and deep heirarchical file structures will create tall objects,
where'as sites with flat or database driven structures will result in a
flatter series of planes or plateaus of information. The order that these
levels are built is dependent on the order of their appearance to the
user, and each distinct directory path occupies it's own horizontal plane.
The color and length of any linking strand represents the direction and
distance of that link within the structure that is being established.


MF:  On your web-site, in the text accompanying some screen-shots of the
software in action, you use particular terms to discribe these spatial
arrangements such as 'plateau', 'crown', 'tree' and so on.  How much are
these ways of describing the links a result of the way the WebTracer
software spatially organises the display of links and how much are they
structures that are inherent to the structure of the particular web-sites
that it hits?

NP:  The particular structural forms that result from a WebTracer run on a
site; as 'plateau', 'crown', 'tree' are a combination of both the order in
which the program 'sees' the links and their paths and locations on the
remote webserver. Although the display routines can be configured
differently, the molecular model resulting from a 'trace' reflects very
closely the information structure of the target site, both on a file
structure level and on an information design level.


MF:  We already have as commonplace the phenomenon of art and other
websites being made to be only viewable through certain configurations of
software and access speed, that only make themeslves visible through
certain very narrowly configured sets of software devices. The arguments
for and against this, echo of course, some of those considered at the
inception of the web and are ongoing, with the destinction between
pyhsical and logical mark-up of text etc.(oldskool!)

 For these sites, the import and export filters of software already
constitute a hidden micropolitics of which file formats are accepted or
are interpretable and which not, based around alliances between the
different forms of organisation that generate these protocols and
standards. And obviously these systems of gating and reading, of coding
and decoding, operate at many different scales - including cultural ones -
during any particular period of use of a piece of software. One other
related thing that occurs on the web frequently is people blocking
spiders, from search engines etc. from their sites - that is to say from
people / machines reading their data in certain ways. I wonder, given a
perhaps increased emphasis on 'using' or perceiving the data on a site in
the 'correct' way, how you perceive the WebTracer operating in this
context?



NP:  Well, there's quite a range of issues you have highlighted here, but
as you point out they all stem from the same old internet (or hypertext)
argument of freedom of form/media versus control of form/media. As I
touched upon, in answer to a previous question, the nature of the internet
and associated technological media has meant that different parties see
different means to different ends. The ongoing process of encoding the
theoretically open system of the web is an inevitable development of it's
popularisation and commodification.

Reducing information to a series of eight.dot.three file formats and
locking those formats into the development and distribution of software
applications, serves to create a language that is both arcane and
specific. Such frames placed around the dataspace of the net have a dual
purpose; On the one hand they contextualise and compartmentalise the
medium into bite sized chunks, which users can familiarise themselves with
and reflect already existing metaphors or schema; On the other hand they
tie up data and medium to statements about ownership and intellectual
property.

With the definition of a system comes the ability to quantify it and
commodify it. A natural extension of this practice is the concern over
infringement of these definitions or alternative readings and systems
(hence the blocking of autonomous agents e.t.c.). The web has gone from a
very open media which grew because of it's inherent qualities of 'openess'
into a system overloaded with the imposed frameworks and metaphors of
commercialising agencies. There becomes an "official" way to browse,
syndicated by whoever has the largest presence in the definition of the
term. I'm not saying that applications such as webtracer are in any way
countering that trend (in a sense they are providing further reworkings)
but perhaps they will make people aware that there are still different
ways of viewing any system.


MF:  You mention the difference between flatter, or database driven sites
and those that have a more hierarchically ordered structure. Would you say
that one of the things that WebTracer and other pieces of software that
map links between sites is to effectively flatten all sites into a
'plateau'?


In a sense, yes, but the action is of course not a physical/dimensional
flattening but rather a psychlogical reduction of the intricacies of data
into one specific analysis. Webmapping software is concerned with certain
features or issues in hypertext, the rest it can ignore from it's
resulting output. Obviously there are many factors which affect and
dictate the production of a web site, but most webmapping software is
reductive and formalistic.


MF:  Following on from this, how do you see people using the software? How
do you use it?

I would like to see people using it in an almost sculptural way, there is
a certain aesthetic kick of of revealing the inherent structure of a site
which I think appeals to a lot of people. I would also like to think that
it could be used practically as well as an information design analysis
tool, but i suspect that it would need more commercial development for
this. I have used it for both these purposes, but I think that what I
enjoy most about it is the pseudo filmic way you can move from node to
node across a mapped site as if it were a medical examination. I have
already had many suggestions from users of some very varied and creative
ways of using the application from both the designer and the user point of
view.



MF:  There's a bundle of other material on the nullpointer site, from the
relocated material of dividebyzero.org to sound generation software in
which you seem to be exploring other potential spaces for software to go.
What are the key ways in which software can be developed that mainstream
software is missing out on at the moment?


I think that developing software is a real double-edged sword. As you
write new software, you become acutely aware that you will be continually
restricting aspects of it's functionality, to suit your needs. You can't
help then but reflect on the way this process occurs in all the other
software you use and even in the tools you write your own software with.
One of the few ways to counter this trend is the open source movement.

Open source isn't just about code either, it relates to a whole set of
attitudes that can benefit the resulting software. The video games
industry thrives on the developer community and is one of the most cutting
edge sectors of the industry. There is also a less visual but equally
important area within academic developer community (IRCAM,MIT e.t.c.) Each
area of the developer community has skills that can benefit the others. In
my own work I try not to restrict myself to working only in one community
or with one programming environment and I will use code or approaches that
are already available and then warp them to my own personal ambitions. I
would like to see simpler products coming from the mainstream software
market, but with a much greater facility for mods and patches to be
developed by the user community. If it wasn't such a janky program, I'd
love to see the Quake modmakers get to work on Microsoft Word;)






#  distributed via <nettime>: no commercial use without permission
#  <nettime> is a moderated mailing list for net criticism,
#  collaborative text filtering and cultural politics of the nets
#  more info: majordomo@bbs.thing.net and "info nettime-l" in the msg body
#  archive: http://www.nettime.org contact: nettime@bbs.thing.net