When I was in college at MIT in the late 1970's, it was a strange time for computers. I worked in a part of the Lab for Computer Science that had close ties with the Artificial Intelligence Lab. Between the two labs, we shared four giant PDP-10’s— giant as in physically large, like dinosaurs, filling whole rooms, but not giant like fast or rich in memory. The iPod in your pocket is probably a thousand times faster and stores a hundred times more data than any one of the four machines that was shared by several dozen of us at one time. Computers have come a long way in that time, but not just in terms of their speed. And yet, for all their primitive nature, we had some advanced concepts back then you might have assumed are more recent innovations.

These particular four machines ran a unique operating system, aptly named the Incompatible Time-sharing System (ITS), that was at the time available only on those four machines. But understand, too, that this was at a time when the total number of machines on “the net” was tiny. My recollection is that, in particular, it was fewer than 256—the computer-equivalent of my saying “more than you can count on one hand.” (Computers have slightly larger “counting hands,” metaphorically speaking, since they use 8 bits in a byte instead of 5 fingers on a hand.) We knew the machines we used by both name and number.

ITS was fascinating as a time-sharing system (a machine that was simultaneously used by dozens of users) for a variety of reasons, but the one I wanted to mention here is that it had really no security whatsoever. It was as close as the computer community will ever get to a bunch of people living together in the wild, where the only rules are the ones that people voluntarily obey and the only protection is the shared desire of the participants to work together for a common good they all agree on. It was a place where at any moment the actions of any individual could have brought down the entire system, both figuratively and literally. (It took only a few keystrokes to stop the computer and everyone had the power to take that action. Not everyone knew the specific key sequence, but it was available in documentation for anyone who cared to look.)

It was open to the net in a day when really no one was around to cause problems. In fact, the system wasn't even passworded, so nothing kept anyone from logging in as anyone else, actually. People tended to use their own name just because there was little reason not to, and because it probably didn't occur to them ever to be anyone but themselves. I'm sure people must have sometimes played pranks, but the more common case was that someone would ask someone else to be them just to help them debug something. When freedom works, it's an amazing thing.

And everyone could read each others' files, including mail. When one connected, it was common to read the messages people had been sending to one another to get caught up. (This may seem weird, but again it's all a matter of perception. No one thinks anything of catching up in comments on a blog post that came while they were asleep, even if those messages are sometimes a back-and-forth conversation between two people with others merely “lurking.” The mail reading command on this system took an argument of whose mail you wanted to read, and people who sent or read mail on this system generally knew that it was readable by others. So again you have to evaluate these things in the context of the societal norm.)

Of course, we ran regular backups of all the files, just in case of a problem, but we didn't perceive a likelihood of problems even in that open world because we trusted that those we encountered would behave kindly to us. And, for a while, they did. Anyone could just connect and make themselves accounts. And they were quickly assimilated into the way we did things.

The spying mechanism allowed us to see what newcomers were doing, and so we could help them keep from making mistakes. If a new user was on the system, usually one and sometimes several people would watch them for a while to make sure they weren't seeming lost or confused. If they were making syntax errors, they'd receive messages from someone saying what to type. They quickly learned to just type questions at their keyboard and see them answered.

So it was all very social, and although I had used other computers that were otherwise before, I quickly forgot how unusual it all was and didn't remember until years later when I moved to another system and I thought suddenly how lonely it was. I felt like Hugh in the “I, Borg” episode of Star Trek: The Next Generation when he complains how quiet the Federation ship is, and how he no longer hears in his head all the voices of the Borg.

Over time, as more people came into this world, there were increasing incidents of nastiness. (See my remarks about the rise in crackers and the co-opting of the term hacker in a blog post last week.)

I have no way to formally validate this claim, it's just a personal impression, but my sense was that this happened when there were so many people that no one bothered learning each other's name any more. Once a community is not made up of people one knows personally, and might answer to later, it's easier to do things that are either potentially or actively negative. And so mischief started to creep in.

At first we added “optional passwords.” That is, most people could still be themselves without passwords, but for trust purposes we created a way for people who wanted to give passwords to identify themselves so that others could know they were really them. That way the new crop of bad guys wouldn't spoof the administrators.

(If it were Open Salon, it would be as if one could use whatever screen name you wanted, as long as it wasn't Joan or Kerry... just so that there was some confidence that the “powers that be,” in an otherwise relatively lawless place, were recognizable when they chose to speak.)

After a while, that didn't work. So we added passwords for everyone, and we began turning off the accounts of people who were causing trouble. (See a recent blog post of mine about being quoted out of context for a few random additional details on how this went.) Once inside, there was still no security, but it was harder to get in. A kind of gated community.

In the end, we finally gave up that operating system and moved to a commercial one that had conventional protection systems. It was almost another decade before social environments like the early virtual realities and the web were re-invented in other forms.

If you found this post interesting, please "rate" it.