An ironic young man ... may be viewed as a pest to society.

- Carlyle

Thursday, April 8, 2010

Myths and Legends of the Cold War

I begin this sad tale with a heavy heart. Being of sound mind and firm memory, I sense that it would be safer to yell "Intel forever!" in front of a crowd of AMD fans on a Friday evening than it would be to get into the thick of things and make sense of the subject that interests me. Up till now I was aware of the possible consequences and thus tried to pretend that I didn't know of any such question, as if it did not exist; I made it clear that everything that will be discussed below was but an insignificant episode in our history, which should be mentioned in passing, if at all.

Just to describe it in two or three sentences, half made up of interjections and punctuation marks. And yet... An evil fate leads me onward, and I cannot delay much longer. I hereby inform everyone that, if my breathless body is found in the near future at the bottom of the Reichenbach waterfall, this will be no accident, and I had no thoughts of leaving this vain world. But enough vague hints. At last I am embarking on an attempt to make sense of the question of how and why the USSR introduced the "Single System" of computers. The Rubicon is crossed, the bridges are burned. All rise; court is in session.
- "Ликбез," UPgrade magazine (Russia), May 1o, 2006
I don't blame you for being confused. What is this piece doing on this blog? What is with the columnist's mock-portentous tone? And what in God's name could he possibly be talking about?

The truth is, I didn't need to pick this particular article. There are any number of them with a similar message. And they all boil down to something like this: "In the late 1960s, the Soviet Union sold out its native computer industry in order to copy IBM's System 360 computers, and this constituted an enormous moral and technological defeat in the context of the Cold War. This is why we lost the computer race!" They all quote Edsger Dijkstra--and the quote seems to vary from publication to publication--saying something like "The Soviet Union's adoption of IBM's System 360 was the United States's greatest victory in the Cold War."

The backstory is this. In 1950, the Soviets built the first stored-program computer in Europe; many new types of computers were developed throughout the next two decades, and software development was occasionally on a highly advanced level. But by 1969 it was crystal-clear that the Soviet Union was falling further and further behind the United States. Computer utilization was low where it needed to be high and high where it needed to be low; reliability was poor; there were not enough programmers; the transition to integrated circuits was far from complete. So the Soviet government decided, together with the Warsaw Pact countries, to cancel all new development on its native-born computers and transition to clones of the landmark IBM 360 machines, which would consolidate all hardware and software development in the Eastern Bloc. Needless to say, despite the vast quantities of Soviet cash that flowed into the coffers of European and American technology companies, it didn't work: by 1983 leading computer scientists were explicitly announcing that the USSR would never catch up.

In and of itself, the story doesn't matter much. But it's part of a much broader complex of such stories, all of which, in the end, have a similar moral. The AK-47 was a superior weapon to the M-16 because it was cheaper and didn't jam when you breathed on it; those good old MiGs and T-series tanks were always superior to equivalent US equipment; Soviet doctors dreamed up therapies that are still considered bleeding-edge in the West. The moral, of course, is "We did so much with so little, and we still lost"; by extension, it's also "we deserved to win, and it's all the fucking government's fault." With the Single Series, it's even more poignant: "we were great with computers, and the fucking government sold us out to the Americans."

Are the stories true? The one about computers probably isn't. The Ural series (which the above columnist eulogizes with great tenderness as a superior native alternative to IBM) never cracked 100,000 operations per second, and never, as far as I can tell, transitioned to integrated circuits; at any rate, by the early 70s there were still a couple hundred Urals left in operation, and presumably they could have demonstrated their superiority, but failed. The BESM-6, the finest Soviet machine before the Single Series, was two years behind its American equivalent. Either way, it would have made little difference: the Soviets simply could not match the manufacturing and personnel-training rate of the United States. The Soviet economy was held together with toothpicks and gum; this worked okay for tanks and tractors, which have a limited number of components, but the high-precision fabrication needed for modern computing was simply not consistently available.

That isn't very romantic--but the USSR did gproduce something unique in this field: the world's only (I think) ternary computer, the Setun. Alas, like the native computer industry, it was not long for this world. It wasn't an intriguing evolutionary path with great, but wasted, potential; it just didn't work very well.

The point, I think, is that it doesn't matter if these stories are true. They are not made up with reference to the facts. No amount of technical documentation is likely to convince a Russian that they are false; the semi-mythical secret defense sector is always there to be cited as a source of mislaid innovations. These stories are simply myths, tinted, perhaps, with a great deal of defensive projection.

Of course, one need only look at all the movies and video games about supposed secret Nazi projects for raising the dead or colonizing Mars to see that such myths are not unique to the Russians. (Although maybe the Americans are unique in making them up about other countries.) In fact, the endless mythological churn of losers who try to explain away their loss goes back deep into history, at least back to King Arthur, maybe to Lascaux. What The Romance of the Single Series can teach us is that modernity--or even postmodernity--is no prophylactic against them. Alligators still inhabit New York City's sewers, just as third-generation Ural-32s still gather dust somewhere in the labs of Omsk-9.

3 comments:

  1. Dijkstra had a tendency to repeat himself and not all of his papers have been transcribed, but EWD 506 from 1975 could be the source of the quote:

    "The German adoption of ALGOL 68 has had a similarly paralyzing effect upon the Germans as the Russian decision in the late sixties to develop as their next national computer series a bit-compatible copy of the IBM 360—the greatest American victory in the cold war!"

    ReplyDelete
  2. Fascinating, and I like the moral you draw at the end (though not the reflex "it doesn't matter if these stories are true" — of course it matters, it's just not relevant to your point at the moment). I wonder what stories Americans will tell themselves when their empire is one with Nineveh and Tyre? I think of R. A. Lafferty's wonderful "Land of the Great Horses," about the wandering Angelenos and their myths of the city they have lost.

    (Переводческое: "Stand" should be "All rise." Also, something's gone wrong with "the USSR did give produce something unique in this field.")

    ReplyDelete
  3. Paul--that sounds like it's the one. I hardly think 6 years would be enough time for appropriate critical distance. (Although, unfortunately, ALGOL was popular enough for paralysis in Russia, too.)

    Hat--Fair enough, on all counts. The other obvious point of reference would be something like A Canticle for Leibowitz.

    ReplyDelete