To boldly go where no human has gone before; to seek out the basis of a new civilization in the realm of cyberspace: This (one can get away with claiming in so self-congratulatory a context as a commemorative anniversary issue of the premier software developer's magazine) has been the 15-year mission of the enterprise DDJ.
Fueled by several power sources, this month's "Programming Paradigms" touches down at three points along the electronic frontier. I am indebted to David Bushnell, Victoria Elder, and Howard Rheingold for insight into the origins of ARPANET; to Lee Felsenstein for his explication of the vision of community memory; and to Computer Professionals for Social Responsibility and John Perry Barlow for the working notes to the Code of the New West.
First, a loose definition and a strong claim. Cyberspace is so science fictiony a word that I almost shrink from using it as I want to use it. But it's the best word I can think of to refer to the place where you go when you go online. That's the loose definition, and here's the strong claim, which I will come back to twice more in this column: Although cyberspace is not associated with any physical location, it is a real place, or many places. Describing it in these terms is not metaphorical, but factual. This, I believe, is important to realize.
The story is told, by David Bushnell and Victoria Elder in Directions and Implications of Advanced Technology (Jonathan Jacky and Douglas Shuler, eds., Ablex Publishing, 1989) and by Howard Rheingold in his Tools for Thought (Simon & Schuster, 1985), of how the Department of Defense funded the development of a new technology that was inherently antiauthoritarian. It says something about my own prejudices that I find this surprising, but I do. That is, I am surprised that the DOD would bring into existence something that it could not control; I am not surprised that a technology could have political implications.
In 1960, the batch model of programming prevailed: You submitted your program in the form of a box of punched cards and hoped that there would be no syntax errors or typos in your code when it finally got its turn to run. Time sharing, when it came into its own in the 1960s, was an enormous advance in accessibility and productivity -- now many users could work simultaneously at terminals, interacting with the computer as though each were the only user. Then came remote hookups and the possibility of working at home on the campus or office mainframe.
Toward the end of the 1960s, some important ideas came together, one of them being the idea that, rather than just connecting terminals to central computers, it might be interesting to connect computers to computers. Where the ideas came together most effectively was at ARPA (later DARPA), the Department of Defense's Advanced Research Projects Agency.
The Rand Corporation put together an 11-volume classified report in 1964 that proposed developing a fully distributed communication system for all (data and voice) military communications, via what would later be called "packet switching." The reason for opting for a distributed system was military: The argument was that a system with no central control would be much harder to disable in a nuclear war. The proposal went nowhere, but a few years later, Robert Taylor, the director of ARPA, was considering something very similar. ARPA was supporting research in various universities and other research sites around the country, so Taylor was well situated to see the value of close communication among distant researchers. By 1968, ARPA was looking at a full proposal for a distributed, packet-switching network of the computers at ARPA research sites.
Such a network would demand significant processing power to maintain it, but the proposal did not use a large central computer to run the network. Rather, it gave each constituent computer a dedicated processor, which would come to be called an Interface Message Processor, or IMP, to share the job of running the network. The network would be distributed.
ARPA bought the idea and solicited bids. IBM, among other likely bidders, didn't bid. It was Bolt, Beranek, and Newman (BBN) that got the bid and that built ARPANET, and BBN still maintains it today.
Rheingold points out the subtle changes that the distributed nature of ARPANET wrought: "The controlling agent in a packet-switched network like ARPANET was not a central computer somewhere, not even the 'message processors' that mediated between computers, but the messages themselves." And: "The idea of a community that could be brought into existence by the construction of a new kind of computer system was perhaps the most radical proposal of the [original] paper."
A community. ARPANET really did create a community, a community whose gathering place was, and is, ARPANET. Even though it has no specific physical location, ARPANET is a real -- not a metaphorical -- place. In cyberspace, place can exist without physical location and community can exist without physical proximity.
The community that gathers on the ARPANET commons is a specialized community with built-in shared interests. As ARPANET was being implemented, another network was being designed that endeavored to create a kind of electronic commons for ordinary people.
Community Memory's chief engineer, Lee Felsenstein, describes the system in deliberately prosaic terms: "Imagine a telephone book in which you can list yourself as many different ways as you want, even under different names, and when you change your listing in your telephone book, it changes in everyone else's telephone book. Wouldn't that be nice to have. For everyone to have. That's essentially what Community Memory is."
More concretely, Community Memory was planned to be a network of public-access terminals in the San Francisco Bay Area. It was not initially designed as a distributed system such as ARPANET. While ARPANET had the resources of the DOD behind it, Community Memory had a hand-to-mouth existence as the brainchild of impecunious 1960s counterculture technophiles. It ran on a central computer that Felsenstein and the people he was working with happened to have, a 940 time-sharing system; Felsenstein characterized it as "the first machine designed and built for time-share. Big deal." The original purpose for which the computer had been acquired had been forgotten, supplanted, or voided somehow, but there sat the machine with a free-form query input database, a large hard disk, and time-sharing access. Community Memory was the use to which the group decided to put it.
Community Memory went online in 1973. "We were running the system with a terminal in Berkeley and one in San Francisco from August, 1973 to January, 1975," Felsenstein recalls. It was used: One of the first personal computer companies -- in one sense the first -- was started from a discussion on Community Memory. Processor Technology founder Bob Marsh connected with Felsenstein on Community Memory, and Felsenstein later designed the Sol computer that made Proc Tech, briefly the hottest of the early microcomputer companies. But the idea was that Community Memory was to be for nontechnical people. The success of that idea had to wait for the success of the technology and financing of the Community Memory project. It took over a decade before Community Memory went online for real. In mid-August, 1984, Community Memory put three terminals online in Berkeley. Today Community Memory supports a diversity of users, although it is still a small network.
That diversity is more important than it might seem and has a lot to do with the civilizing of cyberspace.
"As a result of the fact that we've got a group of people who are convinced that [information] needs to be free and another group of individuals who are convinced that they own it, we enter cyberspace already in a state of civil war. And that's not a good way to start out a new civilization."
That's John Perry Barlow, who, with Mitch Kapor, founded the Electronic Frontiers Foundation. Barlow was speaking at the annual meeting of the Computer Professionals for Social Responsibility last October. He may have been preaching to the choir, but he made a strong case that all computer professionals should give thought to the uses of the technology they create and, in particular, to the new world they are creating.
"Most of the people I talk to about this stuff say 'I'm a bus architect, not a social philosopher,' " Barlow said. "And I really understand that. But society is so completely perplexed by this technology that even a bus architect is better qualified to be a social philosopher than a lot of people."
Barlow outlined the questions that we social philosophers of cyberspace should be asking ourselves, and helping less technical people to ask themselves:
"We have to make it possible for human interaction to take place in a familiar format. Design is a very important function because what it is is the cultural mediation of the familiar. There is no design going on [in the computer field], with the exception of the Macintosh. If we designed cars this way, the engine would be in the front seat and you'd adjust your speed by diddling with the carburetor. It's primitive, folks. We have to make it possible for ordinary people to live here. We have to think about issues of esthetics and culture." And design can have serious cultural implications: "Jaron Lanier has said that there's nothing technologically advanced about the Mac: it's a cultural device. He's right.
"The second thing we have to do in civilizing cyberspace is to reduce the polarization that already exists. I go around telling security people that they have nothing to fear from the crackers, and telling crackers that security people are basically afraid and confused, and that if you treat them as though they were malevolent fascists they will be that. If you call somebody a pig he will oblige you by becoming one.
"But it's difficult, because everybody has already chosen up sides and drawn a deep line in the dirt, and they are already throwing rocks across it. So one of the responsibilities that you folks have," Barlow told the CPSR audience, "is to try to bring some light and some humanity and some decency to a situation that is already very deeply polarized.
"Finally, there are political solutions. I don't have a lot of faith in political solutions over the long term because I think of cyberspace as being basically an apolitical place. For starters, it's transnational. I get a lot of e-mail from abroad saying, 'How can we protect our civil liberties?' And I can't say, 'Rely on the First Amendment,' because they don't have one. This brings up a very important point:
"You can't rely on the law, because the law is for the 'real' world and is necessarily local."
If you can't count on the law, where can you turn? "You have to start relying on culture and community and shared ethics," Barlow says. "I hate to tell people that because they've become so reliant on lawyers that [they find it hard to grasp.] But I think people are going to have to treat one another like human beings, and that's the only way it's going to work.
"Besides, you've got a situation [in cyberspace] where guerrillas will always win. There is no better jungle to fade back into than the jungle of information -- which, as Stewart Brand says, wants to be free anyway. And will be, I think.
"In order to have this kind of cultural integrity, the first thing you have to do is to abolish fear. You have to assess the real risks [and confront them honestly]. It's difficult telling people that we have to abolish fear, because there's a lot of nameless dread kicking around, mostly as a result of alienation from this very stuff. People out on the end-user end of things feel like they're on the learning curve of Sisyphus. They're afraid. They feel like they're being dragged into a place where their children are natives and they will never be able to learn the language. And that's enough to give you a lot of nameless dread.
"Good work is being done on the political front, but I would not rest my faith for the long term in political solutions, particularly legal ones. I would rest it in the community of my brothers and sisters and the people in the computer community who can understand the technology and can make it inclusive of the ordinary folks out there who are afraid of it."
In the Q&A session after Barlow's talk, a representative of Community Memory stood and talked about how young urban African-American men are using CM's public access terminals to discuss events in their community and to invite each other to parties. The CM representative singled out this group because they are a group currently being stereotyped in the media as drug-crazed homicidal maniacs. "I hope that as [we] build a constituency for electronic freedom," the CM representative said, "we also do everything we can to build in a diversity of participants, or else it will be a very white, very bland, very government-controlled cyberspace." And very male, she might have added.
It is a tenet of the American mythology that we need a frontier. I doubt that this is as peculiarly American a need as we sometimes pretend. Doesn't the whole world need a frontier? America itself was once Europe's frontier, its New World, and a lot of Europeans seem to think that we're all still cowboys. Maybe they need to think that. Since the West was won, that is, civilized, various metaphorical frontiers have been put forth to substitute for our lost frontier, to try to give us whatever it is that we need from a frontier. I suggest that the Electronic Frontiers Foundation is doing something better than this. I suggest that the name of Kapor and Barlow's organization is not metaphorical, but literal, and that the electronic frontier is a real frontier because it is a real place, every bit as real as the physicist's universe of quarks and galaxies.
But much larger.
Copyright © 1991, Dr. Dobb's JournalThe Community of ARPANET
The Electronic Commons
The Civilizing of Cyberspace
In beginning to answer these questions, Barlow says, we begin the process of civilizing cyberspace. There are many ways to go about this, but "the first and most important of these," Barlow says, "is the same one that we used with the West, which was making it habitable for ordinary people. It is very difficult for an average person to go to the places I go whenever I log in. And the bandwidth is extremely thin. We have to change that bandwidth to make it possible for human interaction to take place there.