Programmation Oriente Art, colloque organised by David-Olivier Lartigaud & Anne-Marie Duguet, CRECA (le Centre de Recherches d'Esthtique di CinŽma et des Arts Audiovisuals), UniversitŽ Paris, Sorbonne, 19/20 March 2004.

David-Olivier Lartigaud, 'Introduction'

Many comparisons are made between 'computer art' of the 1960-70s but clearly the conditions were very different - artists worked with programming out of necessity at that point in time. With today's 'software art' the emphasis is on programming in a quite different sense - to Lartigaud, it is 'a wish to go against standards and formats ever more massively spread'. This distinction is made explicit in Andreas Brogger's work in the comparison of the exhibition 'Software' and the contemporary term software art. He insists on the need to operate a 'more dialectical way of understanding the status of these objects' in whether to accept software art as art - however, if you applied the formulation of Benjamin here, you would simply conclude that whether it is art or not is the wrong question, as it changes the nature of art. For Lartigaud, it is a question of judgement and a term such as 'artistic programming' clearly means something different to the different constituencies of art and programming, but also probably makes little sense in itself. I would simply say look at a term like this dialectically - decidedly in terms of its contradictory nature. A similar mistake would be made in separating technical and artistic registers (note: in the case of ascii art, the programming itself has achieved an aesthetic dimension).

'Programming' is defined as a 'set of utterances describing a forthcoming action [...] a set of operations to be implemented in order to get a result'. Computing programming clearly makes sense in relation to this. It etymology can be traced back to the Greek 'programma' as 'what is in advance written' - as set of instructions to be executed that are fixed beforehand. Unlike a score that is followed but interpreted, a computer follows the instructions without interpretation. Thus, is a different kind of criticism required - a 'hybrid criticism' necessitating expertise and critical understanding in art and programming. He thinks that 'open source' is a particularly interesting area for criticism in that it opens itself up for scrutiny whereas if a work is not open source how can anything other than its execution be appreciated. Here strategies like 'deprogramming' and 'dŽtournement' of technology are of relevance too in calling to attention the structures and standard formats of software.

There is a confusion between art and aesthetics here - just because something, like programming, can be considered for its aesthetic qualities that does not necessarily make it art. Audience-address is part of this issue in how much a viewer understands the programme behind the work (this was the debate over the CODeDOC exhibition of course).

--

Olga Goriunova, 'Runme.org Repository: What you believe is what you get'

With a vested interest in the category 'software art', Olga Goruinova is unconvinced of the usefulness of debates over the terms themselves. She simply states that software art and criticism is justified in its usage; in the fact that people find it a useful focus for discussion and to grant exposure to particular practices. In the case of the Read_Me festival, the intention was to draw attention to works that lie outside the mainstream festival culture and thus build an alternative structure of power and alternative curatorial strategies - one in which people could submit works to an 'open' repository that was not selected or juried (this is the Run_Me.org site). This is essentially an open database based on a software repository model.

This is not without its difficulties particularly around categorisation. However, the festival probably attracts a wider constituency that includes demo-coders for instance, but still suffers from a confusion of codes. For instance, anecdotally, in the case of a work exhibited at the 2003 festival, the programmer was not interested in attending, having any say in the works presentation or collecting an allocated fee. The clash of cultural registers is illustrated in this example where the expectations are built upon the conventions of the media art festival. Cramer puts it thus:

'Another problem is the association of software art with the "media art" system, with the side effect that artistically interesting computer programs - like those which emerge in the field of GNU/Linux and Free Software, for example - do not reach the software art competations, festivals and exhibition' (Cramer, 2003 'ten theses') Software art is clearly not just media art (simply something between sender and receiver that is mediated) but is a process too.

Artists and no-artists might be seen to make things that could be considered art, but whether software art is art or not is the wrong question as I have stated elsewhere.

To Goruinova, an interest in code goes back to classical philosophy and the absolutist fantasy and belief that absolute truth might be discoverable using mathematical logic. Code is a leap of faith and an act of worship in such a scenario. She emphasises that software criticism uses theory not as an abstract system but as a tool kit (this is paraphrasing Foucault).

Andreas Broeckmann, 'Questioning Software Art'

Andreas suggests the term 'autistic computing' to describe work that is relatively autonomous and self-sufficient. In this case, reflexivity is embedded in issues of representation and he cites a number of examples that exemplify the issue of the 'narrativisation of the code' in some way: Rainer Mandl's work where the program executes the drama of Oedipus; Maurizio Bolognini's 'sealed computers' where the computers are processing data and sending it through a local network but we are denied access to the detail of this (the monitor connections are even sealed with wax); and, Harwood's 'london.pl', where the relationship of the code and its representation are somewhat intentionally confused. These examples are not autonomous but reveal machinic processes in relatively open systems. The question remains as to aesthetics - and Broeckmann suggests one possible way of assessing the value of this work in terms of aesthetics through a Kantian approach and the idea of the sublime (note: interestingly Fuller refers to the 'geek sublime' in his essay 'Freaks of Number').

Inke Arns, 'Read_Me, Run_Me, Execute_Me: Software Art and its Discontents'

The title taken from Freud's 'Civilisation and its Discontents' is a neat way of drawing attention to what lies hidden - although doesn't explicitly talk of the return of the repressed in relation to code. Using this analogy, open source might be advocated to avoid repression and to put the programmer in touch with the culture's sublimated desires to be open and free (repressed under capitalism).

Arns is keen to compare and contrast the terms 'software art' with 'generative art'. Using Phillip Galanter's definition, she regards 'generative art' as far too general and inclusive (applied to music, architecture, design, etc), that tends towards a discussion around the negation of predictability and intentionality. It also tends towards a focus on end-products in her opinion as opposed to process. On the other hand, software art, for her, reveals hidden structures that are potentially political in nature. This is a spurious distinction for me as there is clearer a more developed critical discourse around software art (thanks to transmediale and the read_me festival).

I would disagree strongly with this comparison. For me, generative art is more evocative of process as an adjective that describes an activity - it describes a performance and is perfomative. One might mention 'generative grammar' from Chomsky or 'generative matrix' from Zizek to describe ideology if one wanted to be more explicit about the potential for a political dimension. Its inclusivity also allows for a history to be registered and cultural practices that do not necessarily involve computation - which doesn't just made it open-ended but allows for broader cultural production to be seen as 'generative' - generative of change to my mind, even of social change. These terms are in flux all the same, and very open to debate and contestation.

Florian Cramer, 'Ten Theses About Software Art' (2003)

To Cramer, all digital art is software art in that it relies on or is assisted by other software to run (be it a browser, operating system, network protocols).

[add to bit about the relationship of writing to the tool - in code notes]

In 'Ten Theses About Software Art' (2003), Florian Cramer explains that artists make software in a way that writers use language: 'No literary writer can use language merely as a stopgap device with which to compose an artwork that is not in itself language - so, like in a recursive loop, literature writes its own instrumentation.'

The connection between software art and conceptual art is made apparent in taking the concept like code as the material fro the artwork. In this, for Cramer, notation and execution of a concept or of code are collapsed into one piece (in Goruinova & Shulgin, 2003: 54) closest to our understanding of what consititutes language. His example is La Monte Young's 'Composition 1961', a piece of paper with the instruction: 'draw a straight line and follow it' (what might be considered as a precursor for socialfiction.org's '.walk'). However, software art is also not synonymous with conceptual art as dematerialised art in the sense that Lucy Lippard describes the privileging of concept or idea in her book Six Years. For software art is decidedly material, if not immaterial of course (this is a longer discussion). To illustrate his point, he points to a more recent history in describing the development of JODI's work from browser manipulations, to hacks of game engines, to current work that uses (retro) BASIC sourcecode. In this last example, minimalism is certainly evoked but any conclusion that this is simply conceptualist in the older sense should be avoided. However, JODI's work is indicative of a tendency in software art practice to reflect upon its conditions, its materiality and to reject the industrial gloss of work influenced by the aesthetic and technical determinants (and limits) of proprietary software (Lev Manovitch's discussion around 'flash aesthetics' is relevant here).

Much of the discussion here revolves around the separation of user and programmer - with the development of so-called user-friendly interfaces closing off an understanding of the apparatus as a controlling strategy (Apple and Microsoft are part of this deskilling monopoly). The problem with the figure of the artist-programmer in this connection is that the processes of programming have correspondingly become closed off, mystified, based on elitist knowledge and hence contributes to the return of a romanticised myth of creative genius embodied in the programmer or hacker class. (link to Author as Producer notes) These tendencies once deconstructed by post-structuralist theory remain present in certain discourses and practices - therefore are ripe for software criticism.

In the connection to algorithmic and generative art, Cramer is also keen to make a separation from software art. Here, it is claimed that the influence of cybernetic and systems theory informs the Galanter definition in emphasising the issue of autonomy. Again, I think Cramer confuses autonomy in terms of artistic practice and political discourse - ultimately the theorisation itself isn't inclusive enough. He tries to point to neglected human and social aspects but these issues are embedded in a discussion of autonomy in the work of the Frankfurt School or indeed in autonomous Marxism - to name but two obvious examples. The problem for Cramer in Galanter's definition is that the generative aspect is seen to result in a completed work of art rather than being a work of art in itself - execution is foregrounded (which is the legitimate criticism of our aesthetics essay in part). This is an important point and clearly reveals a limitation on the generative art definition itself but not the term that remains open to further and improved definition. Cramer would add that software art emphasises software as material for play: 'it no longer reads it - as in conceptual and generative art - as pure syntax, but as something semantic, something that is aesthetically, culturally and politically charged'.

Software is clearly thoroughly political (think of patents, licenses, rights, monopolies, etc) but so is generative art - and generative art suggests, at east for me, the possibility of not just deconstruction but of an active involvement in transformation.

Matthew Fuller, 'Freaks of Number' (also posted to nettime, March 22, 2004).

Fuller is keen to historicise software art in terms of calculation, and sees Maurice d'Ocagne's 'Le Calcul SimplifiŽ par les ProcŽdŽs Mecaniques et graphiques' (1893) as an early example of computer criticism. The standardisation of objects for Fuller typical of industrial production follows this same numerical logic. Programmers are contemporary 'freaks' of numbers in this sense, as individuals able to assemble their thoughts in an appropriate way for the purposes of calculation. They have 'power lodged in their heads'. This is not hard it should be pointed out - he quotes Joseph Weizenbaum (programmer of Eliza) in this connection: 'Almost anyone with a reasonably orderly mind can become a good programmer with just a little instruction and practice.'

It is almost as if programmers exhibit a numerical disorder (that as an example he expressed in contrast to the numerical disease of being in a hetereosexual couple). In this way, there is a politics of numbers that is exemplified by Benjamin (quoted by Fuller):

'... the ability of machines to duplicate words and writing outstrips human needs. The energies that technology develops beyond this threshold are destructive. First of all, they advance the technology of war and its propagandist preparation' (note: from the essay 'Edward Fuchs, Collector and Historian' but the Artwork essay makes more of this connection of technology to war as a direct result of an aestheticisation of politics that might be extended perhaps, in this context, to an aestheticisation of code). Software art in this sense is a mechanism for a reverse tendency, a critical means for the exploration of how software propagates the standard object.

He says: 'On the scale of numbers, post-industrial society is perhaps something that occurs when the 'avalanche of numbers' of Hacking, an enormous and self-generating torrent of factualisation, tabulation and recording meshes with numericalised labour, mechanisation and product and informational standardisation and variation.'

Mainstream computing is locked into a neo-Platonism for Fuller that finds aesthetic value in the most simple, purely expressed of formal resolutions to a given problem.

In contrast is Harwood's London.pl (found on the run_me.org software repository) based on, or plagarising, William Blake's poem 'London' written in the last decade of the Eighteenth century. Here, in both works, statistics and the modulation of populations are used for social comment. In the Harwood version, arguably, the contemporary 'arithmetico-material' conditions are doubly registered both in content and form. For instance, one line of the program comments reads:

# Find and calculate the gross lung-capacity of the children screaming from 1792 to the present.

The issue is made public as well as actualised in the present to evoke contemporary suffrage.