Such a "history", however, overlooks the fact that interactive art is firmly rooted in the aesthetic upheavals of the 20th century. The questioning of the role of the artist, the work, the audience, the market and the relationship between art and society by the dadaists, the constructivists, the surrealists and others prepared the ground. In the 1960's Fluxus, happenings and "participation art" (Frank Popper), cybernetic art, the art & technology movement, environmental art and video art already provided many of the incredients of interactive art. The pre-digital work of pioneers like Shaw and Hershman bears evidence of this. Present day artists by no means ignore their parentage.

It is true that interactive artworks were first welcomed (and even commissioned) by science centers rather than art institutions. This was partly a consequence of the context in which pioneers like Myron Krueger had worked and first been appreciated. Interactive art is still most often seen in the context of the computer world (e.g. Siggraph art show) or the festivals and institutions dedicated to bridging the gap between art, technology and design (e.g. Ars Electronica in Linz, Austria). In the mass media, interactive art is often featured in the context of science and technology, rather than art and culture. C.P. Snow's old idea of the "two cultures" still seems to have some validity.

The situation reflects the customary inertia of the art world, which has hardly legitimized video art, just 25 years after its inception. What's more, video art has developed much closer to the existing art institutions, whereas computer art allegedly has a questionable background as the bastard son of the military-industrial complex. Interactivity is seen by many cultural elitists as a fad, inseparable from the all-pervasive banal discourses of technoculture. Or, maybe there is an unconscious fear of the potential of the interactive media (CD-ROM, WorldWideWeb) to make the museum obsolete? Sure enough, museums now have interactive touch-screen information kiosks, or CD-ROM catalogues for sale - should these be considered as mere marketing ploys, "vaccinations" against the I-infection, or as real symptoms of a new "attitude"?

Although there is no reason to try to fit interactive art into any traditional aesthetic canon, there is a need to differentiate it from other interactive applications. Most of the latter use interactivity as an operational strategy and aim at reducing it along the way, opting for a closure. Video games may be remarkably complex in their architecture, but they are a form of goal-oriented activity, whereas art is multi-layered and open-ended. There is no final "solution" to an interactive artwork, no way to exhaust its meanings.

Design (understood as rational planning) forms an important part of the realization of an interactive artwork. There are many design tasks (often distributed among co-creators or helpers): for example, designing an interface, or a flow-chart for a hypertext architecture. There is also need for engineering skills. However, neither a beautifully designed software code, nor an ingeniously engineered hydraulic platform is a work of art. An artwork requires something else, a kind of surplus of inspiration and signification which will transcend the rational assembly of the "machine parts", melt them together and give them a raison d'être on a higher level of abstraction. This is something different than creating an involving plot for a video game.

Some critics wish this was true. However, there is much less openness in most interactive artworks than they hope. Even though the presence of the artist is usually hidden behind the scenes, it can be inferred. There are few interactive artworks which purport to give the user the impression of being the (sole) creator. Kit Galloway's and Sherrie Rabinowitz's classic telematic work Hole in Space (1980) was one of them - the situation was set up in a public space, no learning process was involved, no announcements were made, no indicators of "art" or "signatures" exposed. The two-way "hole" highlighted the actions of the users in both ends - these were, however, hardly conscious of producing art.

Most artists have adopted a more "traditional" strategy, inscribing their presence in different ways. Many interactive artworks are "unique" artefacts, "sculptures" or installations which can be experienced in a certain time and space and are identified as somebody's creations. In most cases there is an implied presence of the author in the work, which can be felt e.g. as direct address to the user, but even in the restrictions introduced into the modes of interactivity, thus deliberately positioning a "distributed" will (of the artist) against the will of the user. The native Canadian artist Yuxweluptun has (in his VR installation Inherent Rights, Vision Rights) given such restrictions an ideological significance by superimposing strict rules of movement onto his "native" virtual world, thus demarkating his own territory from white man's territory of cyberspace.

As artists' Net sites will develop, the question about the death of the author will appear again. Some artists will probably prefer to disappear behind their sites, transforming themselves into invisible figures of the webmaster, while some will "die" as the activity around the site proliferates; some will do their best to impose their presence in the manner of the countless private home pages, but with a product that makes a difference.

I don't think the amount of interactivity should (always) be the main criterium when judging interactive art (although it is crucial when judging an industrial application). As stated before, there are works which deliberately restrict the possibilities of interaction as part of their artistic strategy. There are even works which terrorize the user with deliberately produced "malfunctions" and "bugs". Ken Feingold frequently deceives the user's expectations, e.g. by disturbing the principle of a real-time one-to-one relationship between the user and the work. The responses one gets may have been triggered by the previous user of his Surprising Spiral. The work could, of course, have worked just like a teller machine.

Works like these are highly self-reflective, reflecting also a consciousness about the historical pre-forms of interactivity. They could be characterized as meta-interactive art. Other artists appropriate prevailing interactive applications and turn them into new experiences, as exemplified by Perry Hoberman's way of using regular barcode readers in his Barcode Hotel. This kind of interactive art maintains a constant dialogue with the field of interactivity, questioning and deconstructing its achievements, and sometimes extending them to unexplored directions.

Beside meta-interactive works, there is also need for artworks which concentrate on building new systems, pushing the limits of interactivity by applying the latest scientific ideas, for example artificial life, intelligent agents or knowbots. These artistic experiments lead away from prevailing patterns of interaction towards more complex situations, which will eventually enlarge the user's range of possibilities. Christa Sommerer and Laurent Mignonneau belong to such artists.

The idea of intra activity is simplistic. The self-reflective situation should be seen as just one of the "conversational" options embedded in an interactive system. The basic situation is a "polylogue" rather that a monologue. The user of even a "local" (or "off-line") interactive artwork engages in conversation with multiple "partners" (in addition to oneself) simultaneously: the physical "frame" of the work, the fictional world it "contains" with all its elements, the software with its agents, the implied author(s) of the work.

In an artwork which also incorporates an on-line connection the situation gets even more complex: in addition, there is now the possibility of communication with real humans in remote locations as well as with manifold software agents and knowbots residing in the net. Sometimes it will be difficult to tell which is which. In the near future we will probably see more and more of these kind of hybrid artworks, with both a local and a global face, and providing the user the simultaneous experience of being present and faraway in some distant location. Such situation tend to reduce rather than increase the narcistic potential of the medium.

This idea, again, is too simplistic. Interactive technology isn't gendered from the outset; it becomes gendered when it is put into a social context, turned into a cultural form. Yet, interactivity has many different applications in a variety of contexts. Not all of these can be labeled as "masculine", even though the breakthrough of video games as the first interactive medium with a strong mass appeal certainly seems to highlight the masculine appeal. Video games provide a powerful outlet for suppressed frustration. They also re-enact modes of behaviour which can be easily adapted - when stripped off the notion of "talking back" - to acts aimed at causing destruction and imposing dominion.

By putting the user into the controls interactive technology could be claimed to have a strong liberating potential, as well, making it an effective means to analyze and deconstruct pre-existing ideological formations. Maybe this the reason why there are so many female artists working with interactive technology. All New Gen, a kind of reversed video game, meant to expose the masculine construction of the Nintendo ideology, by the Australian cyberfeminist group VNS Matrix is perhaps the most extreme, but by no means the only example. Lynn Hershman has used the gun, the main fetish of the male-dominated American society, as the interface in her America's Finest to reflect on representations of militarism and the consumer society. Here interactive art functions as a kind of philosophical instrument, enabling us to experience something familiar as if entering an alien territory, to investigate the world - and ourselves - from a fresh perspective.