Sex and drugs and serious thought

John Naughton, an Open University lecturer, used to write an excellent television column for The Observer . He now writes an internet column in which his wonderment at the new medium alternates with scorn for those who,because they do not understand the net, think they can abuse it, exploit it or dominate it.

A recent target has been Jack Straw, the home secretary whose Regulation of Investigatory Powers Bill (RIP) could have meant jail for anyone who lost the key to their encrypted files. The protest was belatedly taken up by other national papers when the bill was already in the House of Lords. Naughton's anger over RIP is easier to understand when you have read his book. It is the story of the imagination, invention and sheer hard work that went into this greatest of 20th-century social inventions: an open, transnational communication medium whose core values are freedom of speech and the gift economy.

Democratic it is not. The internet in its early days was ruthlessly meritocratic. Naughton tells how Tim Berners-Lee, who invented the worldwide web as a tool for furthering serious research communications, had a row with Marc Andreessen in 1993, over his Mosaic web browser, which could show pictures. Berners-Lee regarded this as desperately frivolous. Yet it was precisely this feature that caught the world's imagination and initiated the web's phenomenal growth.

This was not the first clash between elitist and populist views of the net. In the previous decade, the net's snobs tried and failed to control the anarchic, nonstop, global electronic discussions known as Usenet newsgroups. In the early days of Usenet, messages were propagated between computers running the popular Unix operating system, using dial-up phone connections. But control of Usenet fell into the hands of a group of big computer sites with superior communications, nicknamed the Backbone Cabal. When Brian Reid created the alt.sex and alt.drugs newsgroups on April 3 1988, he cleverly ensured that the messages would find alternative routes. The Backbone Cabal, which did not think sex and drugs were suitable topics for discussion, could do nothing. "At the time," Reid later recalled, "I didn't yet realise that alt groups were immortal and could not be killed by anyone."

Equally immortal is Fidonet. Launched in California in 1983, Fidonet was even more grassroots than Usenet since it used personal computers rather than professional Unix systems - it still has 3 million users. Naughton comments: "At a time when governments and multinational corporations are itching to get the internet under (their) control, it's deeply reassuring to know that the framework for an alternative, free communications system not only exists, but thrives."

But dialled calls between computers are a poor way to route information from millions of sources to millions of destinations. The way the internet now works is to chop information up into standard-length "packets", label each packet with a destination address and let the packets find their own way there.

Paul Baran proposed this idea in the United States in 1964, and Donald Davies had the same idea independently in the United Kingdom in 1965. Davies's team forged ahead with what they called "packet switching" and, ironically, it was the British engineers who persuaded the Americans to use packets on their new research network, the Arpanet. The internet's TCP/IP protocols were originally developed for "internetworking" between the Arpanet and other networks. The US defence department placed the Arpanet contract with Bolt, Beranek and Newman, a much more laid-back outfit than the typical defence contractor.

The sites on the proposed network were mostly universities. Their representatives on the Network Working Group were mainly graduate students.The group, which included Vint Cerf, Steve Crocker and Jon Postel, floated ideas in papers, called requests for comments (RFCs). Internet technologies are still developed in the same collaborative, non-authoritarian style. Nothing was secret, solutions emerged iteratively and the results were in the public domain. This was, Naughton argues, the genesis of the open source movement, the gift economy that would drive the development of the Linux operating system and the Apache web server in the 1990s.

The idea of the web did not come out of the blue to Berners-Lee. As long ago as 1980 he had created a hypertext program called Enquire to help him cope with information that was too complicated to remember. Naughton probes deeper. The Talmud is, perhaps, a hypertext. By 1933, Vannevar Bush had his first inklings of an instant document-retrieval machine. His famous article "As we may think", although written in 1939, was not published until 1945.

Then came Doug Engelbart, Ted Nelson (who planned a fully commercial global "docuverse" but lost out to the web's gift economy) and Bill Atkinson, whose brilliant 1987 Hypercard program got one thing wrong: it "assumed that all the connections worth making resided on your hard disk". Big mistake.

Naughton writes a particular kind of history. It stresses decisive moments, inimitable geniuses, oaks from acorns and crucial chance events. All of which makes a very good yarn.

What is missing is any account of the false starts and failed projects, which could perhaps have given us a global network of a very different kind. In the 1980s, TCP/IP still had to see off IBM's proprietary Systems Network Architecture and the overblown Open Systems Interconnection, an endless series of international standards that failed to standardise anything. As late as 1988 there was a threat that the Arpanet would be cut off from other networks and the Communications of the ACM asked: "Can the internet survive?"

Katie Hafner and Matthew Lyon did much of the basic research on the internet's early history, and Naughton gives generous credit to their 1996 book When Wizards Stay Up Late: The Origins of the Internet . One reason for reading Naughton's book is that Hafner and Lyon end their account before the web was invented. Another is Naughton's brilliant advocacy in distilling the ethos of the internet and persuading us that it is precious and worth defending. He spells out the message of a new medium more convinc-ingly than Marshall McLuhan ever did, and in snappier prose.

Tony Durham is web editor, The THES .