Cheaper than printing it out: buy the paperback book.

I live on computer networks. The network of networks -- the Internet -- links several millions of personal computers around the world. No one knows exactly how many millions are connected, or even how many intermediate nodes there are. The Internet Society made an educated guess in August 1993 that the Net was made up of 1.7 million host computers and 17 million users. No one controls the Net, no one is in charge. The U.S. government, which indirectly subsidizes the Net, woke up one day to find that a Net had spun itself, without much administration or oversight, among the terminals of the techno-elite. The Internet is, as its users are proud to boast, the largest functioning anarchy in the world. Every day hundreds of millions of messages are passed between its members, without the benefit of a central authority. I personally receive or send about 50 messages per day. In addition to the vast flow in individual letters, there exist between its wires that disembodied cyberspace where messages interact, a shared space of written public conversations. Every day authors all over the word add millions of words to an uncountable number of overlapping conversations. They daily build an immense distributed document, one that is under eternal construction, constant flux, and fleeting permanence. "Elements in the electronic writing space are not simply chaotic," Bolter wrote, "they are instead in a perpetual state of reorganization."

The result is far different from a printed book, or even a chat around a table. The text is a sane conversation with millions of participants. The type of thought encouraged by the Internet hyperspace tends toward nurturing the nondogmatic, the experimental idea, the quip, the global perspective, the interdisciplinary synthesis, and the uninhibited, often emotional, response. Many participants prefer the quality of writing on the Net to book writing because Net-writing is of a conversational peer-to-peer style, frank and communicative, rather than precise and overwritten.

A distributed dynamic text, such as the Net and a number of new books in hypertext, is an entirely new space of ideas, thought, and knowledge. Knowledge shaped by the age of print birthed the very idea of a canon, which in turn implied a core set of fundamental truths -- fixed in ink and perfectly duplicated -- from which knowledge progressed but never retreated. The job of every generation of readers was to find the canonical truth in texts.

Distributed text, or hypertext, on the other hand supplies a new role for readers -- every reader codetermines the meaning of a text. This relationship is the fundamental idea of postmodern literary criticism. For the postmodernists, there is no canon. They say hypertext allows "the reader to engage the author for control of the writing space." The truth of a work changes with each reading, no one of which is exhaustive or more valid then another. Meaning is multiple, a swarm of interpretations. In order to decipher a text it must be viewed as a network of idea -- threads, some threads of which are owned by the author, some belonging to the reader and her historical context and others belonging to the greater context of the author's time. "The reader calls forth his or her own text out of the network, and each such text belongs to one reader and one particular act of reading," says Bolter.

This fragmentation of a work is called "deconstruction." Jacques Derrida, the father of deconstructionism, calls a text (and a text could be any complex thing) "a differential network, a fabric of traces referring endlessly to something other than itself, to other differential traces," or in Bolter's words "a texture of signs that point to other signs." This image of symbols referring to other symbols is, of course, the archetypal image of the infinite regress and the tangled recursive logic of a distributed swarm; the banner of the Net and the emblem of everything connected to everything.

The total summation we call knowledge or science is a web of ideas pointing to, and reciprocally educating each other. Hypertext and electronic writing accelerate that reciprocity. Networks rearrange the writing space of the printed book into a writing space many orders larger and many ways more complex than of ink on paper. The entire instrumentation of our lives can be seen as part of that "writing space." As data from weather sensors, demographic surveys, traffic recorders, cash registers, and all the millions of electronic information generators pour their "words" or representation into the Net, they enlarge the writing space. Their information becomes part of what we know, part of what we talk about, part of our meaning.

At the same time the very shape of this network space shapes us. It is no coincidence that the postmodernists arose in tandem as the space of networks formed. In the last half-century a uniform mass market -- the result of the industrial thrust -- has collapsed into a network of small niches -- the result of the information tide. An aggregation of fragments is the only kind of whole we now have. The fragmentation of business markets, of social mores, of spiritual beliefs, of ethnicity, and of truth itself into tinier and tinier shards is the hallmark of this era. Our society is a working pandemonium of fragments. That's almost the definition of a distributed network. Bolter again: "Our culture is itself a vast writing space, a complex of symbolic structures....Just as our culture is moving from the printed book to the computer, it is also in the final stages of the transition from a hierarchical social order to what we might call a 'network culture.'"

There is no central keeper of knowledge in a network, only curators of particular views. People in a highly connected yet deeply fragmented society can no longer rely on a central canon for guidance. They are forced into the modern existential blackness of creating their own culture, beliefs, markets, and identity from a sticky mess of interdependent pieces. The industrial icon of a grand central or a hidden "I am" becomes hollow. Distributed, headless, emergent wholeness becomes the social ideal.

The ever insightful Bolter writes, "Critics accuse the computer of promoting homogeneity in our society, of producing uniformity through automation, but electronic reading and writing have just the opposite effect." Computers promote heterogeneity, individualization, and autonomy.

No one has been more wrong about computerization than George Orwell in 1984. So far, nearly everything about the actual possibility-space which computers have created indicates they are the end of authority and not its beginning.

Swarm-works have opened up not only a new writing space for us, but a new thinking space. If parallel supercomputers and online computer networks can do this, what kind of new thinking spaces will future technologies -- such as bioengineering -- offer us? One thing bioengineering could do for the space of our thinking is shift our time scale. We moderns think in a bubble of about ten years. Our history extends into the past five years and our future runs ahead five years, but no further. We don't have a structured way, a cultural tool, for thinking in terms of decades or centuries. Tools for thinking about genes and evolution might change this. Pharmaceuticals that increase access to our own minds would, of course, also remake our thinking space.

One last question that stumped me, and halted my writing: How large is the space of possible ways of thinking? How many, or how few, of all types of logic have we found so far in the Library of thinking and knowledge?

Thinking space may be vast. The number of ways to overcome a problem, or to explore a notion, or to prove a statement, or to create a new idea, may be as large as the number of ideas itself. Contrarily, thinking space may be as small and narrow as the Greek philosophers thought it was. My bet is that artificial intelligence, when it comes, will be intelligent but not very humanlike. It will be one of many nonhuman methods of thought that will probably fill the library of thinking space. This space will also hold types of thinking that we simply cannot understand at all. But still we will use them. Nonhuman cognitive methods will provide us wonderful results beyond and out of our control.

Or we may surprise ourselves. We may have a brain that, like a Kauffman machine, is able to generate all types of thinking and never-seen-before complexity from a small finite set of instructions. Perhaps the space of possible cognition is our space. We could then climb into whatever kind of logic we can make, evolve, or find. If we can travel anywhere in cognitive space, we would be capable of an open-ended universe of thoughts.

I think we'll surprise ourselves.