Alan Turing - Life and Legacy

Along with many other anniversaries, 2012 is the centenary of the birth of Alan Turing. The Science Museum in London has an exhibition about his life and work, which I went to see yesterday.

Most people who've heard of Turing will tell you that he cracked the German Enigma cipher during the Second World War. They might tell you something about the Turing Test, which tries to determine whether a computer program is as intelligent as a human. (To date, no program subjected to a Turing Test has been consistently able to fool a human for more than about five minutes.) But to my way of thinking, Turing deserves to be remembered more as the father of computing and computer science. This exhibition addresses both his practical and theoretical work.

The exhibition is organised thematically rather than chronologically: the first thing you come across is almost the last thing Turing worked on, the Pilot ACE computer. (Turing didn't actually build this, but it was based on his designs. It was - briefly - the fastest computer in the world, being clocked at a blistering 1 megahertz.)

There are the inevitable Enigma machines, and a couple of rotors from a bombe, a machine used by the British codebreakers to help determine the daily key settings for Enigma. The latter are apparently originals, not reconstructions, which is surprising, as nearly all of them were destroyed after the war.

I had rather hoped that there might be a page or two of the manuscript or galleys of Turing's landmark 1936 paper, On Computable Numbers, with an Application to the Entscheidungsproblem, but I didn't see any. There is a video of someone talking about it and why it's significant, which is probably more useful for anyone who doesn't have a degree in computer science.

The Entscheidungsproblem ("decision problem") is a question posed by the mathematician David Hilbert (who was German, ironically enough) which essentially asks, within the language of mathematics, can every statement be proven true or false? Turing's paper showed that this was not so, closing off an entire field of mathematical enquiry. But in doing this, he proposed a "universal machine," now known as a Turing machine, which could calculate the answer to any and every mathematical problem that could be solved. The machine I'm using to write this blog post, and the machine you're (probably) using to read it, are essentially Turing machines. The same paper presents a program for a Turing machine that allows it to emulate another Turing machine, thus anticipating the concept of a virtual machine by a good thirty years. (Yes - VMs have been around since the 1960s. They just weren't practical or desirable for most uses until recently.)

The exhibition concludes with three interactive displays that teach basic concepts in programming - decision, looping and... I don't know what the third one was, as it was broken when I visited. I didn't think these were as successful as the rest of the exhibition, as they bear little resemblance to a real computer or real programs, and it took me a few minutes to figure out how each one worked. Perhaps my computer science degree and my career in software development are making me overthink things. Most people will never need to write a real computer program, so it doesn't matter (much) if these displays aren't like the real thing.

Overall, I think the exhibition gives a good account of the man and his work, and would recommend it to anyone who's interested in the Second World War, codebreaking or computing. Admission is free (always a bonus!), and it runs until the end of July 2013.