The word came to me from several sources on December 12, 2001. Google Groups, a search engine company started by my friends Larry Page and Sergey Brin, had updated its archive of USENET postings to go back to mid-1981--more than 20 years, and almost back to its beginning in 1979.
USENET is the world's largest online community. And for a long time, perhaps even still, it was the soul of the Internet. It was the meeting place. The place where the important topics were discussed. The place where personalities met and things happened.
Google's new USENET archive brought back memories for me and for thousands of others who lived and made the dawn of the computer network age. A decade before there was a Web and 15 years before the dot-com bubble, we argued the technical and political issues of the day, met and made friends and rhetorical adversaries, and dreamed of the "WorldNet" yet to come.
USENET was created in 1979 when Tom Truscott and Jim Ellis at the University of North Carolina wanted to replace the tool that updated users on the latest computer center news with a general tool that could support contributions from anybody, a multiuser "bulletin board." There were a few of those around, but what made theirs different was that two machines could run the same software and exchange messages with one another, and these machines could in turn talk to others, until an entire network of machines would share the messages. Users would read by running a local program (rather than dialing out as you did to a typical BBS), but ended up communicating with people all over the place.
| A decade before there was a Web and 15 years before the dot-com bubble, we dreamed of the "WorldNet" yet to come. |
When they released the software on tape at the USENIX computer conference, it spread like wildfire. Dozens of systems ran it and started exchanging news with others. Some would call each other up with modems every few hours or every night to synchronize messages. Better-off sites exchanged messages using the ARPANET at any time of day. In addition, the same programs allowed email, in a cumbersome way, between any two people on the Net, the emails traveling hop by hop the way the USENET messages had spread.
It didn't take long for lots of discussion categories, called newsgroups, to spring up, and for more sites to join. New versions of the software and protocol were written and the growth continued. Soon USENET became the place to go to discuss the issues that mattered to the computer nerds of 1980. There was lots of focus on computer hardware and software, but also discussion of science fiction and political issues--and jokes.
The world's oldest electronic conferences, the ARPANET mailing lists, were quickly integrated into USENET. Gateways were made so newsgroups starting with "fa" (From ARPANET) contained the mailing list contents, and USENET users could post and participate. ARPANET mailing lists emerged around 1975, not long after network email linked ARPANET users together. The first one was started by network pioneer Dave Farber, who today is a professor at University of Pennsylvania, Electronic Frontier Foundation board member, and leading Internet pundit.
  | |
Anybody--at least anybody on a computer at a major company or university--could participate in a newsgroup, and they did. Many of the online trends and words, from emoticons to flaming, originated in USENET and the mailing lists. Similar experiences were happening on online services like CompuServe, and on a growing crowd of stand-alone BBS's around the world, but nothing was like USENET with its energetic and highly educated crowd.
I grew up in Canada, which wasn't connected to the ARPANET, but when I was consulting on the development of VisiCalc (the first spreadsheet) in 1979, I was given an account on MIT-MULTICS, and joined my first mailing list. I was hooked. Twenty months later, while working in California during a summer away from school in Canada, I found USENET.
I immediately began forwarding various messages on the network, and the whole SF-lovers list/newsgroup to my friends in Canada, who spread it around there. At the time, I was working for Personal Software, the leading PC software company, writing software for the IBM-PC before it would be released. Those were heady days, the very dawn of personal computing for the masses; and yet I was already distracted by an epiphany. This--letting people communicate with one another--this was what all these computers were for.
When I got back to the University of Waterloo to complete my last year there, I arranged to get the USENET software installed on my faculty's Unix machine, and (through the generous offer of Bill Shannon and Armando Stetner at DEC to handle long distance charges) the school joined USENET. Around the same time, Henry Spencer at the University of Toronto got his system connected, and the Net (as we usually referred to USENET before it became the common term for the Internet) crossed a national border for the first time. (Little did we know just how much the borderless nature of networks would affect the world.)
Even without having to pay long distance charges, the university's administrators were concerned about the resources the then-tiny network consumed. But I got other people hooked during my trial period and after a time there was no way they could take it away. The newsgroups and the email that came were too seductive, and too useful.
And the network took over my life. I loved participating in it, and my favorite topic was the network itself. Many of us loved to discuss its future and its politics, for we knew in some way (though none of us knew fully) that this was the future of the world. I helped maintain groups and wrote tiny pieces of software to help the Net grow. Others contributed much more software, all of it free and all of it source code, years before the free software movement would take full form.
It was the right place to be. By participating on a list that was discussing the future of email addresses, I "met" Jon Postel, who was the architect and custodian of those systems until his untimely death. In trying to see how to add top-level domains to email addresses, he had proposed writing them as user.domain@com. Though a minor participant, I put forward that user@domain.com read more naturally. This came to pass without further debate, and it amuses me to think my suggestion may have inspired the "dot" that became such a part of the world's vocabulary later. This contribution was a purely lucky accident, but it is symbolic of the nature of those days and why it was easy to get hooked.
O'Reilly & Associates is the premier source for information about technologies that change the world. In addition to authoritative publications, O'Reilly offers a range of conferences, including:
The O'Reilly Emerging Technology Conference, May 13-16, 2002, in Santa Clara, California, which explores the emergence of a new network--distributed, untethered, and adaptive.
I made many friends and came to know many others, though 99 percent of them I had not met in the flesh. Indeed, there were people at my university whom I interacted with on the Net but rarely in the physical world. We all discovered quickly how easy it was to depersonalize people when communicating purely by writing. Perfectly reasonable people would get quickly involved in "flame wars" and use invectives they would never say to a person's face.
As these archives return to the world, I have to admit some trepidation. Reading my old words I am sometimes painfully informed of how naive, and sometimes plain wrong, the younger me could be.
In those days I was one of many who worried that soon the Net would collapse under its own weight. The traffic kept doubling on a rapid scale (forget Moore's law), and at first it was all paid for out of hidden budgets. System administrators at large companies would hide the long distance charges in giant corporate phone bills. Luckier sites used the ARPANET, but many worried that all the nontechnical traffic would be shut down some day as a violation of the rules requiring that the ARPANET be used only for research and education.
People were doing a lot of that, but quickly the majority of the traffic became quite frivolous by that definition. And there were the dirty jokes and the sexual discussions to boot.
| The old Net is long gone . . . but the Net still thrives today. |
So like many, I occasionally would write a post suggesting the Net would fall if certain current trends continue. We worried what would happen each fall, when students returned to schools and started posting as "clueless newbies." We worried what would happen if there were too many newsgroups, or newsgroups about drug use. We worried about the large postings of software and other binaries. We worried when sites like John Little's "Portal" appeared, where any BBS user with a modem and a monthly fee could get on the Net. It all couldn't scale, could it?
For a while, in the early days, a person could read all of USENET. Later, it took a determined person with a lot of free time. Not long after, it became impossible. As groups filled with more and more noise, it became harder to be a full participant. Some groups split into subtopics; some did not. At each turn, something came to solve the problem. Modems got faster and long distance got cheaper. Permanent connections got more common. The news-reading tools improved to allow you to skim and browse more easily. Inter-networking and LANS became common.
And so it never died. Each time we predicted it would die, something came to save it. It was too valuable, too important to too many people. In the end I concluded that it did die, but each time it was quickly replaced in a phoenix-like way with something bigger. The old Internet is long gone (except in those archives), but the Net still thrives today.
Back then we also wondered what would happen when the world discovered what you could do with computer networks. Going back as far as the 1970s, this was a popular topic among network builders. We all spoke of the coming "WorldNet" and what it might mean to have everything connected.
We feared in particular that when the rest of the population discovered the Net, it would be as pleasant for us as it was for the Native Americans when the Europeans discovered the new world. Jumping ahead a decade, it's amusing to note that we were mostly wrong. The general population discovered the Net and immediately looked to its earliest denizens for guidance on how to act, live, and even do business. It was as if the Europeans had landed in America and eagerly asked Native Americans how they, too, could build tepees and commune with the great spirit.
Years later, in 1989, I decided to write down a short summary of some of the major events of Net history I had seen or participated in. I did it just as a series of bullet points, too busy to write it all but wanting to get it out of my brain before it vanished.
As a joke, making fun of myself and others, every ten lines or so in the history, I inserted the phrase "imminent death of the Net predicted," to remind people of all the bad predictions that had come and gone. It later became a cliché to respond to predictions of the dire fate of the Net with that phrase. Sometimes it's even been echoed back to me, which is an entirely appropriate irony.
(I am pleased to see that with the archive in place, it will be possible for historians and cultural anthropologists to expand my history and the history of others with pointers to the real events. I will do a bit of that myself some day.)
Some of my explorations in Google made me blanch, others made me smile. When I started my first software company and got my own Unix box, I had it join the Net. As was the custom in those days of a smaller Net, each new site announced itself to the community. Seeing my own small announcement again gave me a warm glow, reminding me of the excitement of having my own site.
I also found great hoax sites like Landon Noll's kremvax and my own nsa-vax.
Hoaxes were great fun in those days. The Net was small enough and slow enough that you could catch a lot of people. Most people took it well. These days hoaxers get lumped in with "trolls," people who are desperate for attention and who thus go "trolling" by making outrageous statements hoping for a response. Rob Pike of Bell Labs perpetrated the Mark V. Shaney and Elisabeth Bimmler hoaxes to great amusement, but he would be shouted off the Net today. (The archives are not yet complete, so the text of these hoaxes is not yet available.) My own favorite hoax, teasing SF fans about Battlestar Galactica, is also not yet there.
Much of the history of the Net revolved around arguments over group names. I've become convinced this was particularly counterproductive, and I wish we could have designed things to avoid that debate and get into more substantive issues.
The naming debate was so voluminous because the names of the groups (arranged in a hierarchical tree like a file system or Internet domains) sometimes conveyed their importance. In addition, in fewer cases than people imagined, some sites with limited budgets would only be fed the more "serious" groups in the serious hierarchies. This made people fear having a "talk" group and crave a name for their group, starting with "comp" because everybody got that.
In the beginning anybody could create a group just by typing in a new name when asked what group they wanted to post to. This was not just anarchic, it caused groups to be created when people made a typo. Later it required an explicit action, and much later a system was put in place to decide what groups would actually be universally accepted.
The debate came to a head when it came time for a "great renaming," the simultaneous renaming of all groups, putting them into seven broad hierarchies. In many ways that didn't stop the fighting, it only made it worse. The naming debates continue to this day.
| It's amazing how much is discussed on USENET that never makes it to the HTTP-based Web. |
There is an immense amount of Net history--now world history--buried in that archive. I'm certainly not done wasting my time exploring and reliving it.
It's worth noting that the USENET archives have been up for some time, because a dot-com company called "Deja News" put up an archive sometime in 1995. Around the same time DEC's Alta Vista also had a close to real-time USENET search archive. These archives contained only current articles posted after the archives began. Both companies hoped people would browse the archives just as they search the Web, and that they might get advertising dollars from this. That was not to be; but when Deja News reached death's door, they at least sold the archive business to Google.
Even the current archives were a tremendously valuable resource. It's amazing how much is discussed on USENET that never makes it to the HTTP-based Web. Particularly, discussions of products and companies. If you're ever searching for experience people have had with a particular firm, the USENET archive is your place to go.
But with the addition of the past, it becomes a fount of history. I've been doing some etymological research. For example, I have just written an essay on the origins of the term "spam" based on Google searches. I also include the origin of "net-surfing."
Sadly, many people are a little too embarrassed at some of the things they wrote in the distant past, and so they ask the archives to remove them. To stay copyright-safe, Google complies, as did Deja News. While I can understand the motives of those who are surprised to have their pasts come back to haunt them (Who truly wants it recorded in stone that you foamed at the mouth like an idiot about something a decade ago.), from a historical perspective it is a shame. Fortunately, nonpublic archives for use by historians and researchers will probably remain complete. Copyright law has exceptions built in to stop it from standing in the way of research.
It may make sense to allow postings to be made anonymous, so that personally identifiable information about the poster is removed, thus allowing the public to repeat most of the experience.
Google has produced a page listing pointers to some of the major events of Net history that you may find a good starting point. So even if you weren't around the first time, you can still live the nature of the digital beast as a spectator. Dig away.
