Marion Kalmus:
Q. 1.
I had a conversation recently with a programmer who is working with Neural Networks. He explained the beauty of Neural Networks something like this:

Traditional Artificial Intelligence design starts with a set of rules which operate on a rich data structure. This systemic process has been criticised for being 'brittle', ie. Where the programmer hasn't anticipated a permutation, the system will collapse when confronted with that permutation.

Neural Networks address data and ask 'What are the salient pieces of information?', so NN's are appropriate for dealing with complex, noisy situations: such as vision and speech recognition or motor control. For situations where it is not easy or appropriate to generate specific rules in advance.

One advantage of Neural Nets is their lack of brittleness and their functionality of 'graceful degeneration'. That is, if you pull out a neuron from a NN the network will cope with the loss better than a classic AI system from which a piece of statistical data has been removed.

One of the great advantages (and disadvantages) of NN is that they can spot things that you have missed as a programmer. A Neural Network is not wholly dependant on the antecedent imagination, intelligence and knowledge of the programmer. As a programmer, you are building a systems that can potentially learn and 'adapt' in relation to human interaction.

If you come from a philosophical perspective which sees art not as a discrete object which acts upon others, but as a process in which the viewer and artist alike take part in; then this perspective on NN is potentially an interesting idea when applied to art production.

ie: A programme (an artwork) that is not exclusively dependant on the antecedent imagination, intelligence and knowledge of the programmer (the artist). A programme (an artwork) that changes inherently in relation to the input data (in relation to the viewers interaction).

Work like this would seem to be ideally Postmodern in that the author's perspective is only an initial point from which the work itself soon departs. It would effectively make a nonsense of concepts like 'intention' and 'authorship', (and about time too).

I'm interested in the workshops views on this. Is it exciting? How will 'self reflexive' artworks effect how artists see themselves? And how does this place us in relation to our ethical 'responsibility' for our productions?

Q. 2.
This is a question about tendency of form in computing. The 'tendency' being the inherent mode of articulation that an art form seems to move towards.and which it is often impossible to not to reference when working in that form. For instance; if you are into the effect of space on the emotions, it might be better to work with Installation rather than 2D Photography. If you are into investigating the perception of colour, it might be a bit dumb to work exclusively with an HB pencil.
I have worked a lot with Film and what used to be called Time-based mediums because I think they tend towards Narrative.

I'm trying to work out what the tendency of of digital work is? I havn't got much further than that it is probably something to do with Logic?
Now that we increasingly get most of our data in 'bit' rather than 'atomic' form, is the concept of form in art becoming redundant?

Q. 3.
How can Interactive art acknowledge and assuage the fear factor in the the viewer?. (For viewers over 30 years old only).

Q. 4..
I have been interested in Epistemology, .and the conditioning precepts of cognitive systems for some time. I have come to a (perhaps Nihilistic) position where it seems that all systems are arbitrary and dependant on consistency. That any system can potentially work as long as consistency is observed.

I have recently been introduced to the basics of programming using Lingo in a Object Orientated Format, and have started to write collaborative scripts with artist Ben Evans.

Programming fascinates me because it tells me a lot about the logic of language and absolutely nothing about the world. In Lingo you invent a concept, and as long as you use define that concept clearly and always use the same naming convention in relation to it, it will work. Like all languages, as a user you don't really need to know how the language works. You must remember some handling conventions and learn some arbitrary names for things, but if you do so, you can start talking. When you are programming, it is impossible t fail to recognise how arbitrary your language is. (Two programmers writing script to perform identical tasks would write scripts that looked quite different.)

The success of the code you write is to do with the success of your logic. It has nothing to do with the names you invent. Nothing to do with the world, or objects in the world that you might want to imagine those names describe, signify or point towards.

I'm trying to learn to programme because it strikes me that eloquence in programming allows artists to relate to technology as a language rather than a tool.

I'm interested in looking at programming conceptually because it shows us what is inherent in all linguistic systems, and perhaps our own thought. Every time you sit down to programme you are involved in the creation of language in the most inescapably tangible way. For the beginner things move very fast, so the kinds of shifts that happen over centuries in inherited languages can happen over hours in one's own programming experience. Concepts and the names for those concepts emerge, are refined, work successfully, become redundant and are succeeded by newer concept and newer names.

As an artist,I have been trying to make artwork which unpicks and criticises causal (linear, consequential) systems. In relation to my epistemological interests, I recently started seeing causality as the form in which the propositions of thought are cast. As a form upon which the majority of Western Logic is dependent

I guess this is really a philosophical question about thinking: but in relation programming being a linguistic form and to artists programming interactive systems; Is it possible to think without causality? Is it possible to think (to programme) contingently, incidentally, uncertainly: that is in a non-linear, non-causal way?

Q. 5.
About ten months ago I suffered a head injury which has resulted in low level brain damage. The main area effected has been my vision. I have no ability to interpret information from my left eye and have a narrowed channel of vision in my right eye.

One result of this has been that I now see with monocular rather than binocular vision which has profoundly effected my understanding of space. In addition I often experience what I can only describe as 'strange' occurances with my right eye. I have sensate experiences which it is absolutely impossible for me to characterise forttwo reasons. Firstly the sensations themselves fall outside of any actuality that I have prior experience of and secondly it appears that there is no existing communal language to describe such occurances. (And why should there be language to describe what is essentially a neural confusion in a partially damaged system?)

Inherited language describes the limits of 'typical' experience, and tends to fail beyond those limits. Language often fails, there are whole areans of experience where we are used to the failure of language, (like spirituality and love and art), but the everyday description of space, and spatial navigation are not typical areas of language collapse.

Phenomenologists would probably argue that since the penomemology of my perception has been altered the world is acutally a different place for me, not that it appears different, it actually is different. And indeed I have found my 'unusual' vision to be a rich sourch of data for my existing interest in phenomenology and the conditions of cognition. Nevertheless, since my injury, if I want to attempt to describe or articulate my discrete sensate experience I need to invent new language.

As an artist I believe that art is inextricably bound with a certain kind of failure. Art escape the sytemising protocools of viewer and artist alike. Art emerges at the moment when; despite their intentions; the (comforting, protective) organising protocools of the artists and viewer alike collapse. So the job of the artist is to build constructs that have the potentiallity to fail, that failure being the moment when space is opened up with the viewer.

Programming is perhaps the most exacting systemising protoocool imaginable, yet it appears to be an area which unpicks this apparent dilema for artists. How can interactive art, which is computer based, escape its systemising protocools?

Marion Kalmus