Personal Freedom and the Internet

There are a lot of personal reminiscences around about the origin of the modern Internet and about how it came to be. This is another, although it predates anything called the Internet by a fair bit. In fact, it even preceded DARPANET.

I post it with a bit of reluctance. After all, democracy is still the surest guide we have to escape evil, and the government the modern Internet was designed to evade was, and still is, far closer to democratic than most. But, at the time, we felt we had no choice. And, as long as we are human, I doubt that the choices we were forced to face can ever go away.

I grew up under the shadow of John Foster Dulles - if the "commies" do anything I don't like, I can push a button and the whole world will go up in a fireball. Goodie gumdrops stuff, especially since Canada is between the USA and Russia so everything that didn't work would fall on us. Not to mention that Dulles worked for an army general turned president. Today, "Area 51", Annie Jacobsen, documents how out of control American bomb people were at the time.

My life continued with Senator McCarthy and his destruction of so many of my fellow musicians south of the border, General McArthur who blew the Korean War up to bigtime, President Johnson who faked the Tonkin incident to create the Vietnam war, and the American CIA.

And so it was that I quietly joined in the effort to attempt to preserve personal freedom in the face of foreign domination and statism.

My first encounter with computers was at McGill University in 1956 - check out this new-fangled language called FORTRAN. McGill didn't have a computer then for general use, so my hookup was a 110 baud line - I believe to DARPA thence to Dartmouth College in the USA but I stand to be corrected; an ASR Teletype was all I saw. Batch only, of course - FORTRAN didn't go interactive until years later - but the usual error messages generally came back within 10 minutes. Snide comments typed by the computer operators came back faster. (My recommendation: skip FORTRAN - assembly language is faster. It stood up too; I could program faster and more reliably in HP 2116 non-macro assembly at NRC until the late 1970s.)

My experience continued with postgrad courses in the new Shannon information theory, much later with BASIC (definitely at Dartmouth), then by running state-of-the-art computers at the National Research Council of Canada such as a 1968 HP 2116 that ran at a blazing 300 kHz with 8 kB core and 300 kB drum. (I ran a whole building full of world-class experiments with that system, 16 direct-digital-control multiprogrammed experiments no less. I could even edit and recompile its operating system with 15 experiments running.)

But, how could we exchange scientific data with our fellows without the CIA eavesdropping? Sure, we could use letter mail hard printouts, as long as we scotch-taped the envelope corners so we could see if the contents had been read enroute. (That was to keep tabs on our own RCMP, who were so dumb then that they opened them anyway, then stuck a on a note supposedly from the Post Office, "Damaged in Transit".) And, the "bandwidth of a Beetle full of mag tapes" was mighty high for the day. But, surely there should be a better way...

At the time, a number of hackers were setting up what were called BBS's - bulletin board systems. (The term hacker was a compliment then - you had to be an electronics expert too, not just a programmer. A remarkable number of us were 'hams', licensed amateur radio operators - I was VE2ARH.) BBS's interfaced to the telephone network by means of modems - 110 baud, masquerading as Western Union Teletype ASR33's. (In Canada, 300 baud then required that you rent an expensive modem from Bell Canada, along with an equally-expensive specially configured telephone line.) Very quickly, most BBS's set up a long distance charge avoidance network - each BBS had a list of all other BBS's within local-dial range, so we in Ottawa could send messages to Toronto in half a dozen relays or so without having to pay for long-distance connections. Canadian day-time long-distance charges then were hundreds of times a professional salary. But, unlike many parts of the world, in Canada local telephone calls have always been without limits. But, the dumbest exchange monitor could detect the two frequencies used by modems then and switch on a recorder, and we knew it.

And so it was that I joined the world of international intrigue: I proposed a system called Jaberwocky, after the Lewis Carroll rhyme

'Twas brillig, and the slithy toves
Did gyre and gimble in the wabe;
All mimsy were the borogoves,
And the mome raths outgrabe.

Each letter (upper case only, of course - this was the 1960's) was represented by a phoneme. The commonest English letters were represented by the commonest English phonemes, so the result approximated the spectrum of an innocuous English speaker. Pauses were put in that matched human breathing. Do remember that no computer then could even come close to analysing voice waveforms in real-time, so we had to design the "phonemes" so that discrete-component audio filters could sort them out and output DTL (diode-transistor-logic) digital signals.

It was a mess of wire-wrap, of course - only a dozen or so were ever made. And, it seems laughable now - anyone scanning with a modern PC could detect Jaberwocky's output within seconds. But at that time, I doubt very much that any automatic system eavesdropping on the telephone lines of that time could have detected the difference between Jaberwocky-digital communications and those of the average English speaker.

Jaberwocky was local, and slow: 3 characters/second. Others of my colleagues had wider ideas. Coast-to-coast television was just taking off. (The first TV set where I grew up was bought specifically to watch the coronation of the current Queen Elizabeth. Movie film of the proceedings was shot in London, each reel loaded as soon as it was exposed into a series of propeller-driven airplanes and developed enroute to Newfoundland, where a specially setup national microwave hookup had been rented by the Canadian government for the occasion. The whole neighbourhood jammed the livingroom and staircase, and peered in through the windows from the porch to see that ca.14" black&white screen. A week later, we all visited the local movie theatre to see it in colour.) The bandwidth used by TV was then astronomical - four megabaud for the video signal alone.

Well, to skip technical details, we worked out that a deterministic 50 kilobaud signal with a proper spectrum could be hidden in a standard TV signal of the time with no possibility then of detection that it was there except by someone who knew the code. And so, the spirit of Jaberwocky went coast to coast, on a TV network that I won't name even to this day.

Paranoiac? Not if you grew up in the world I did.

Of course, by 1983 and TCP/IP, 110 baud was laughable, 50 kilobaud trunk lines even more so. But, when you read the accounts of the network pioneers who came after me, who did their best to design the modern Internet so that no one could ever control it, remember where we all began - with egomaniacs who were determined to dominate the entire world or destroy it if they couldn't. ("We shall never capitulate, no, never. We may be destroyed but if we are, we shall drag a world with us, a world in flames." Adolf Hitler, the democratically selected chancellor of Germany, 1934.)

So, if you don't like the ability of people to send you untraceable junk mail on today's Internet (and I'm sure you don't), remember the alternatives we faced. Those alternatives will pop up tomorrow if we ever decide that freedom isn't worth the hassle.

John Sankey
other notes on computing