Monday, October 27, 2008

A sonic history of computing

I have been reading the book The Language of New Media by Lev Manovich (MIT Press, 2001). In the early part of the book, Manovich has some scattered comments about the history of computing, and he tries very hard to find connections with visual media.

Of course Manovich refers to the punched cards used to control the Jacquard loom, which is certainly a connection with visual media. But he goes on to discuss Turing's abstract machines, and to say: "[a Turing machine's] diagram looks suspiciously like a film projector. Is this a coincidence?" Manovich then discusses Conrad Zuse's use of old cinema film to make punched tape to control his machine. Manovich's two examples of computing machines in these comments are Babbage's Analytical Engine, which was never built, and Zuse's machines, which as far as I know had minimal influence on subsequent developments. Von Neumann misses out completely.

(And, by the way, I don't think Turing's original paper had a diagram. The paper is "On computable numbers, with an application to the Entscheidungsproblem", Proceedings of the London Mathematical Society, Ser. 2, Vol. 42, 1937. There is what I believe to be a photographic reproduction of it in the book The Undecidable, ed. Martin Davis, Raven Press, Hewlett, NY, 1965. There is no diagram in this reproduction.)

Let me make a case along Manovich’s lines for the importance of sound culture in the history of computing.

In functionality the tape part of Turing’s abstract machine is like an audio tape recorder; is this a coincidence? The BBC began using audio recorders in 1932; they used a steel tape. Turing submitted his paper in 1936.

The modern stored-program digital computer originated from several lines of investigation in the period 1937–1950. The first design that fully implemented the features of modern machines was EDVAC (initial design 1945, came into operation in 1951); this machine was far more important for subsequent developments than any of the examples Manovich refers to. The design used magnetic wire for input and output, a system based on audio magnetic wire recorders. For memory it used a mercury acoustic delay line, where data was stored in the form of sound pulses circulating in a tube of mercury.

Thus sound culture could be said to play an important part in the history of computing. Of course this has no more validity than Manovich’s statements about visual culture and computing.