SPACE OF KNOWLEDGE

A lot of what is said here, about knowledge spaces and so, would have been impossible without computers — and, probably, there would have been no need for it.

Computers make it different. They open a new approach to knowledge. Not only does the quantity of available informations grow explosively, but also, by and by, a new quality of knowledge becomes visible, prospects of a land we are just going to explore — though we live there all the time.*(1)

Computers are part of what is nowadays usually called “new media”. As such they may be seen in contrast to traditional (or “old”) media. Both allow communication and conservation of knowledge, though in rather different manners.

Surely we are inclined to say that computers are much more capable than, for instance, books; for with the aid of computers it is easy to store and, if required, to recall thousands of books — together with uncountable additional informations, pictures, movies, learning programs and so on.

It is typical that, as soon as we get acquainted with something new, we cannot but see it standing head and shoulders above all the old. The scope of the new seems infinitely wider, enclosing the old at best as a somewhat primitive special case. A poor predecessor, maybe a toy for children…

Yet, not always and everywhere can computers replace a book really equivalently.

Maybe the most obvious distinction between computers and traditional media*(2) is denoted by activity.

For a start, computers must be switched on. They require energy. Processes are running inside of them that have to be maintained permanently. Under the surface there is ongoing activity.

But also on the surface, especially on that of the screen, there is motion — not only when movies are playing.

Such motion pictures preceded computers, thus heralding the age of new media, so to speak.*(3) Computers, however, are new and outstanding mostly due to their interactivity, which allows the user to launch and manipulate various kinds of actions.*(4)

Computational interactivity is enabled by continual breaks in the program flow. Everything proceeds in small steps, which may be strung together in almost any order. In this way, even running processes remain modifiable.

While seemingly static states turn out to be processes too, for they have to be renewed again and again; otherwise they would not take place at all.

Not only human users can interact with computer programs, but also other programs. This case occurs even much more frequently. Actually it happens all the time: a great deal of the programs running in a computer just control other programs or provide them with data or hardware resources or so.

Furthermore, every program is composed of smaller units, which themselves may be called “programs”. Each serves a certain function. When required, it gets initialized and interacts with other parts of the program by means of special channels or interfaces — which, after all, are again nothing but such sub-programs serving certain functions…

Most of the activity in a computer is started and controlled by programs, not directly by the user. These programs themselves are controlled by other programs, which again depend on others, and so on. There is interactivity between different levels of control, as well as between programs residing on the same level, so to speak.

In this sense “higher” programs are normally in no other way superior to those they control. Controlling is just a special function to be fulfilled, and often this is best done by programs as simple as possible. They neither have to know more than others, nor see everything happening elsewhere, usually they decide on the base of very few indices, following strict clear rules.

The same is true for programs that integrate others into one big one. Compared with the richness of those components, these central comprising parts are very often fairly simple, lacking every extra complexity that would harden the work of coordination. The more resources the core components consume for their own and for their task, the less is left for the rest, that means, above all: for the user.

On the one hand, every program should be as simple as possible, but on the other hand, every program is characterized by the ability to comprehend different possibilities of acting, which allows it to react on different circumstances, such as different inputs.

This is true even in case that it is just a matter of becoming active or not, a simple question of yes or no, of on or off, 1 or 0.

Different possibilities and the involved moment of decision are decisive for even the smallest elements of programs. They are the base of every computational task.

Decision changes everything. The whole thing — and especially its relations to the parts. So that these become much more a matter of coordination than of addition.

Of course we could say that a program existing of many other programs does contain all those and thus should comprise and possess all their possibilities too. But if this saying is associated with total control and perfect survey, than it is more than distracting. The association is definitely wrong.

One could argue that theoretically it should be possible to grasp all possible branches. And with very simple programs running in a highly reliable context this may eventually be practicable. If so, there is nothing to say against making use of them. Yet, trying to reduce the whole computing to such clearly determined programs is absolutely unrealistic. That would mean, on the one hand, closing one’s eyes to unpredictabilities arising time and time again, particularly when and where not expected; and on the other hand, it would mean restricting excessively the ways to use computers.

One great advantage of computers is definitely their ability to deal with situations that are not one hundred percent determined — thus with situations that amount in good approximation to one hundred percent of our real world.

The same is true for the usability of programs, the interactions with the users. The narrower the guidelines are those have to obey, the more the whole becomes unwieldy, nested and therefore intransparent. And it does not even end up providing more safety, but rather increasing the risk of incorrect use.

In general a program has to react to many different situations, some of them practically unpredictable. Sometimes it is better for a program not to become active at all, maybe even to crash down, than to do something that would have bad consequences. So doing nothing remains basically an alternative worth to be taken into account. It is part of each program’s spectrum of possible actions.

As we have noticed before, doing nothing is also fundamental in the sense of interrupting the flow for modulations or other actions to become effective.

In the end, both forms of inactivity are equal. They constitute the space between periods of activity. So they allow that activity to do — and thus to be! — the right thing in the right place at the right time.

While computers process on huge amounts of data, the activity leaps to and fro, forwards and backwards, guided by programs, making sense of the raw material. All the activity ultimately consists of such leaps, every piece of data tells the computer what to do next, where to jump.

Some of these jumps leave the box, they affect the outer world, so to speak, initiate material leaps that move solid bodies, stimulate mental reactions, cause all kinds of activity. Some of these activities may produce new data — fresh food for the computer.

So the computers are part of greater circles, members of a wider community of actors and actions. Activity goes in loops, thus forming figures that are stable because they repeat again and again. These figures give the activity direction and form — and so may very well be regarded as kind of programs.*(5)

A push on the button — and straight off we find ourselves somewhere else. Brand new spaces open up. We click a certain word and immediately enter immeasurable collections of knowledge, libraries holding answers to never asked questions. Suddenly there is music in the air; a movie starts playing; or someone sends a message, we talk together.

Those buttons, icons, hyperlinks, they are everywhere around us — simple things popping up into a whole universe the moment they are touched.

This is a metaphor for the new world of knowledge and, furthermore, a method that has become indispensable. It is the way in which our knowledge is organized today — rather than in more or less clearly arranged shelves and cases containing books devided into chapters, paragraphs, sentences, words, letters — which, assembled to words, form meaningful phrases and express step by step more and more comprehensive knowledge.

The new method demands and constitutes a new theory. A theory of push buttons, based on a pop-up-logic.

Mechanical devices are, in the main, constructed linearly*(6). The force is transmitted from one component part to the next, wherefore an appropriate connection has to exist.

With computers, things are fundamentally different: the activity is allowed to make almost unlimited jumps. It is conducted and controlled by the program, to be sure, but the program itself can pick from an embarrassment of riches. For that very reason it is capable of being so exact, of reacting on even the finest nuances, of searching its way past all obstacles.*(7)

Each particular way may pretty well be called “linear” — its base, however, is not linear, but rather spatial.

It definitely makes sense to introduce space here as a fundamental notion. And it has far reaching implications.

“Space” refers to a principal indetermination and openness: Much more is possible than appears at first sight. And things change, through permanent (inter-)activity. Linear representations of this activity soon reach their limits, since everything gets much too complex.

In this situation the notion of space may simplify a lot. It comprises, it comprehends the whole.

But it is also logically simpler, coming before any linearity — if we understand space as the abstract embodiment of activity.

Particular linear (inter-)actions are manifestations of this potentiality, tracks rutted through continual repetition, dug into space, structuring it, channelling the activity.

The totality of possibilities provided by a program, the set of its potential processes, can be understood as a space. That space is exactly so defined.

By the program.

A program is a representation of a space. The space boiled down to a point, so to speak. Transportable and reproducible. Objectively –

– provided the proper context.

The plentitude of possibilities mentioned before does not primarily refer to a program’s inner complexity and long twisted ways. Much more important is that the program reacts flexibly on different inputs — which is ideally the case even with most simple programs.

In principle, it is always the same that is happening, it is one program. But it produces a wide range of appearances, a whole space.

In order to communicate with one another, two programs need a third one interconnecting them.

As a rule, the communication is partial. Neither program is aware of every activity of the other. Parts of program A interact with parts of program B, thus establishing a connection. In this way a new program C is formed. With its own space.

The program-spaces penetrate one another, their superposition is a space of its own. Although there is no sharp distinction. Localization in one space does not exclude localization in any other. But of course there are many things occuring only in one of those interpenetrating spaces. Each space is in its way unique.

Programs create continuity where otherwise there are only gaps. They interconnect data far away from each other, making them lie instantly side by side. They rearrange the things, restructure space — and thus create a new space.