Windows: a disorder riding machine


Matthew Fuller


This text is a section of a longer one: Eating Disorder. It is available to

download from http://www.ctheory.com/



The reverse fax effect


In discussions of computer interface design it has become commonplace to note that the standard QWERTY keyboard and its international variants is ananachronism, designed for the pace of nineteenth century machinetypewriters. Having become the standard, it is a design that apparently proves impossible to break1. There needs to be some other kind of text input device. One which matches the processing potential of the computer -

and, with the increasing incidence of repetitive strain injury - one which is kinder on the human frame. Nevertheless, there are as many factors closing this change down as there are pushing for its occurrence.

The fax effect is that situation when, for the first few people to own a fax the machine isn't very useful. As more and more and more faxes are attached to the networks each machine becomes more and more useful.2 The reverse fax effect is that situation when, as more and more people use a piece of technology, and it becomes a standard, the possibilities of that technology being improved or of a radically new approach to it being developed are incrementally closed down.3 The reverse fax effect is the process currently occurring to interface design for personal computers.


Mark Dery gives voice to a common apprehension when he says that:

"the computer, resists representation. Its smooth, generic casing is too inscrutable and its inner workings too complex, too changeable for the imagination to gain purchase on them"4 To embed computers in a shroud of inscrutability turns them into a place where the imagination fears to tread - a representation that itself masks the fact that computers are machines for absolutely specific representations, a place where concepts become materialised, sprayed onto a wafer, soldered into position, plugged into a grid. The workings of computers, however complex, can always be reduced to a mathematical table.

Both Babbage's Analytical Engine and the Turing Machine existed for several years at first as sets of mathematical specifications. Several generations of machine on, this utter predictability is still the case. It is not the entire case though. The usefulness and beauty of digital technology is exactly that it does also resist representation, and is always finding itself as a new way to make itself other. This slipperiness is not something that is only possible once the hard graft of locking this table into place has been achieved, but is also a driving force in its construction: 'Desire never stops making a machine in the machine.'5 As Manuel DeLanda writes "Interactivity, the passing of the machinic phylum between humans and computers was developed both as an intellectual goal by visionary scientists, and "conquered in battle" by the Hackers at MIT. It was scientists like Englebart and Kay who transformed the computer screen into a place where a partnership between humans and machines could be developed. But it was hackers like Steve Wozniak and Steve Jobs who out of sheer desire assembled these ideas into a machine that could compete in the marketplace "6 In 1982, Jacques Vallee a French computer scientist based in the States published a prescient, cooky, book, 'The Network Revolution'. Towards the end of it he imagines a psychedelically enhanced meeting with the Washington research funding bureaucrat of his dreams. In the reverie that ensues he utters the blasphemy that then Office of The Future, a square room full of machines, will never achieve what it promises. Why? "I just don't think it's that simple. An office has people in it, you see. They have lives and ambitions and needs and emotions. They interact. Most of the 'productivity' everybody talks about comes from the interaction, not from the ability to write and process pieces of paper. And those machines will do nothing for the people interaction. "7 This desire for interaction, in or out of an office context, coupled with a joyously declamatory revelation of the largely unspoken truth that computer technology is, "the most absurd, irrational thing ever devised."8 signals the immanence of the then being formed personal computer. We are now at a point where we must ask what form does this desire take once this hole has been punched into history, how it is embodied, how does it flow or solidify?

Perverse Engineering

The window device is not an abstract machine but a shell of it, another layer in the interface to it, that whilst it is at once 'a certain way of regulating and constructing discourses that in their turn define a field of objects and determine at the same time the ideal subject destined to know them'9 also provides a machinery for working on discourse - a self-reflexive tool. Government simulation programmes such as Sim City have strict though largely covert boundaries that force the player to maintain a steady 'realistic' middle of the road between the too-much fun poles of anarchy and despotism in order to sustain the game for any length of time. In the case of games such as these, the boundaries are necessarily relatively hard to define in order to keep the player guessing and to enjoy playing for long enough. In the case of operating systems where the actual code is rabidly guarded there is no real necessity, and no real possibility either for the conceptual machines running underneath the interface to be so veiled. Nevertheless, the 'Mystique of the Macintosh' is closely maintained with only clues to it revealed up front, for instance in the 'Religion', the Human Interface Guidelines manual. For the machine unconscious, there are always virtualities pulling the strings. From Turing's schematics for the universal machine to a corporate vision of what might be the hot new thing five years down the line, there are always attractors being put into place now, pulling things into the future. (In a very real sense for people in research labs or design workshops, the future is now. Scientists, designers, managers, often running themselves as guinea pigs but also running potential futures through an obstacle course of concepts, heuristics, profit/loss projections etc.).

Virtual futures that have already receded far over the dateline can also function as attractors into the past. The continued dominance of the QWERTY keyboard, despite its widely acknowledged technological redundancy is of course just one example. Chances are currently not against the attractors of the future being as half baked and packed with dumb compromise as most of those whose results have accreted into the present.

A retro-melancholy Toni Negri, speaking in another context - of factory automation and struggles over the length of the working day - suggests that: "Automation is freely invented by the knowledge that springs from the rejection of work but is, on the other hand, applied in order to break and mystify the generality of this proletarian and labouring need."10 If contemporarily it is hard to imagine a personal computer otherwise than as an aggregate of windows, bit-mapping, icons, pop-up menus, the mouse, of routines and subroutines spiralling in and out of each other, one use and effect of cyberculture is to recognise and open up dialogue with this machine embedded desire for the abolition of work. To stop this dream image congealing into a fetish through the engineering of a 'childlike wonder'11 in the 'subtle dance of mode and modeless, menu bars and trash cans and mouse buttons'.12

The membrane of the screen reflects and puts into place events occurring at the very deepest levels of the computer. But just as the window catches and commands the attention of the computer, it provides an axiomatic - a forcing device for the behaviour of the user.

Whilst, 'The first principle of human interface design, whether for a doorknob or a computer, is to keep in mind the human being who wants to use it.'13 it is also, 'an exciting challenge precisely because the goal is to change the world by improving the way real people accomplish their tasks in classrooms, offices, factories, homes'14 According to Stephen Levy, what made the 'Leading edge yet compassionately designed'15 Macintosh so insanely great to the development teams that worked on it was that it would change the way people thought and interacted with each other. It would be a discourse machine that produced its operator.

Operating Systems are designed to be abstract, to be general. They must be non-specific enough to allow ease of use and consistency, across a multitude of tasks and data types. Currently, various strands in the computer industry are converging on breaking the rigidity of separate applications where, "users must break their work up into separate tasks and perform each within its associated environment - taking the tasks to the tools.'16 This is being achieved through the development of small specific software tools designed to be used in clusters and taken to the task, often regardless of the data-type. However, whilst increasingly unwieldy applications are fracturing to meet the upsurge of data-types, the conceptual models that construct operating systems and their interfaces remain largely the same.

suspect device

"The concept of windows as a means of looking at part of a large sheet ofpaper was developed by Doug Englebart at his NLS laboratory at Stanford Research Institute in the late 1960s. His system used two tiled windows on the screen. Alan Kay's group at Xerox PARC extended this concept to overlapping windows in the 1970s."17 Paul Heckel, the author of one of the few key texts on interface design, goes on to suggest that once a software device has gone beyond the rear-view mirror of a superficially familiar metaphor and into the actuality of a device, it becomes more useful. For Heckel, scroll bars are the most dynamic and radical aspect of the window - hence their uptake across many disparate applications.

Indeed, the redevelopment of the Macintosh OS for instance has pretty much been that of packing the apple menu with more and more ways to slice across the arboreal directory structure as the hierarchy of windows has continually had incursions made against it in the shape of finder, desktop manager, recent files, recent applications, icon aliasing etcetera. Nevertheless, the layered windows device is not so quite cumbersome that it has had to be done away with.

The powerful impetus behind its introduction was to create a way aroundthe constraints obviously imposed by the size of the display screen, and thus to get as much on screen and hence manipulatable as possible. For the computer as hardwired epistemology, everything depends on the machine's position within wider dynamics; what it is able to come into composition with, and how; the ensemble of conditions which it precipitates, and the devices moving through it. As I have suggested earlier in terms of money, all devices have axiomatics that force what they can and cannot do and that these restrictions last only when they are productive of certain effects that come into composition with wider formations. This doesn't mean that they aren't frustrating. A problem found in the production of general tools - where the general is often mistaken for the generic - carries over into the lazy reiteration of devices that are at best not necessarilyapplicable in the development of other programs.

Whilst in this manner the dominance of a device accrues as a convention - the reverse fax effect - we can also make some observations starting from the question. 'What realities, what domains of objects and rituals of truth does the window produce?' The little animations, frames zooming out from an icon - visual rewards exploding across a 3d but ultra-shallow space; the promise of 'an infinite amount of folders'18 that Apple used early on to coax in the custom of stationary fans who found its potentially hypertext-like neurotic vastness appealing; the perpetual unfolding, all

perhaps provide something of an answer. Windows provides a system through which systems of categorisation operate. In this it is inducive to classification rather than circulation. This fixity though is complicated when, peering into the flicker of its flexible, repetitive grid, the user is encouraged to view windows as, ' your view into information' - a synthetic space where you can actually 'see' your documents. In this alloy of dynamics, the user is the disorder to which and by which this device to negotiate alterity with choice trees, directories - a determinedly neutral and homely version of Foucault's 'capillaries of power'19 - is applied. The windows device is a disorder riding machine, turning mess into a straight line.


© Matthew Fuller

1 The repeated enculturation of QWERTY is bolstered by the costs of staff

retraining, etc behavioural grooving

2 "Roughly speaking, the relations of this sort go up as the square of the

number of elements in a system. This means, for instance, that the

difficulty of running something like a telephone exchange increases not in

proportion to the number of subscribers, but more nearly in proportion to

the square of the number - hence the installation of electronic switching

apparatus in place of the village postmistress." C.H. Waddington, Tools

for Thought, Paladin, London, 1977

3 A variant of the reverse fax effect is applicable to cars - the more

people have them, the more useless they become.

4 Mark Dery, interview by Geert Lovink, NetTime mailing list, 1/6/96

5 Gilles Deleuze and Felix Guattari, Kafka, towards a minor literature,

trans. Dana Polan, University of Minnesota Press, Minneapolis 1986, p. 82

6 Manuel De Landa, War in the Age of Intelligent Machines, Zone

Books,1991, p. 226

7 Jacques Vallee, The Network Revolution, confessions of a computer

scientist, And/Or Press, Berkeley, 1982, p. 208

8 Ibid

9Michel Foucault, Remarks on Marx, conversations with Duccio Trombadori,

Semiotext(e), New York, 1991, p. 101

10 Antonio Negri, Letter to Felix Guattari on 'social practice', in, The

Politics of Subversion, a manifesto for the twenty-first century, Polity

Press, Cambridge 1989, trans. James Newell, p.157

11 Insanely Great, p.131

12 Ibid p165

13 Howard Rheingold, An Interview with Don Norman, The Art of

Human-Computer Interface Design, ed. Brenda Laurel, Addison Wesley,

Wokingham, 1990, p.8

14 Ibid, p.8

15 Levy, p.147

16 S. Joy Mountford, Tools and Techniques for Creative Design, in Laurel,

1990 p.26

17 Paul Heckel, The Elements of Friendly Software Design, Sybex, Alameda,

1991, p.191

18 Mac Plus introduction tape, Apple, 1985

19 Michel Foucault, Truth and Power, in Power/Knowledge, selected

interviews and other writings, ed. Colin Gordon, Pantheon, New York, 1980