All posts made in Sep. 2007:

I had some recommendations for more seminal texts in computing:

  • Rod McLaren pointed me at Tracy Kidder's The Soul of a New Machine, on the creation of a new computer at Data General in the 1970s. It appears to be a story of how objects are not invented but argued, cajoled and bullied into life. This very human account reminds me of Constructing a Social Science for Postwar America (Steve Joshua Heims) on the collaborative construction of cybernetics and what it influenced.
  • robertogreco sent me Jason Kottke's review of Glut: Mastering Information Through the Ages (Alex Wright), in which he highlights Paul Otlet's 1934 description of the radiated library: Over there, in an immense edifice, are all the books and information. From there, the page to be read, in order to know the answer to the question asked by telephone, is made to appear on the screen. The screen could be divided in half, by four, or even ten if multiple texts and documents had to be consulted simultaneously. Sounds like a good reference to chase down. And I don't know why I haven't read Glut yet.
  • Andrew Otwell mentioned two books: ESR's The Art of Unix Programming (1/2 programming and 1/2 cultural history) and The Pattern on the Stone by Danny Hillis. The latter I have been recommended so many times I need to go order it right now.
  • Danny O'Brien said simply: HAKMEM? It's a 1972 "memo" (technical report) of the MIT AI Lab that describes a wide variety of hacks, primarily useful and clever algorithms for mathematical computation. There are also some schematic diagrams for hardware. Aha! But it's odd--it feels like extracts from lab books and answers, and I don't know enough of its history to feel its exciting newness. More research required.
  • Michael Dewar brought up Stafford Beer, inventor of the 1970s pre-internet socialist cybernetic economy system Cybersyn. Now I've previously looked at Cybersyn, but didn't realise (as Michael told me) that Beer's book Platform for Change has an essay on it. That's definitely one for the book pile.

Thanks all; loads of reading there. Now I can't find the email - maybe it was IM - but Tom Armitage mentioned The Essential Turing to me. Man, that looks like it's right to the heart of the matter. Maybe I should read it to warm up to Feynmann's Lectures on Physics.

A bunch of texts about computing:

My friend David Smith has ferried to me a first edition of Ted Nelson's Computer Lib/Dream Machines (1974). This is extremely good. Computer Lib established the computer as something with which people could be creative; something with which people could create art (oh, and make their lives better).

I also have Doug Engelbart's Mother of All Demos (1968) on DVD. He had it at a conference I was at, and was copying it for a friend... I was nearby and happened to have a DVD burner... I asked if I could snag a copy. It's pretty good quality; I'm very pleased.

Both Engelbart and Nelson read Vannevar Bush's 1945 essay in the Atlantic Monthly, As We May Think. Now that'd be an issue to own. Bush put the US scientists on a war footing, and in this article gave something back in the form of the memex: a kind of hypertext, knowledge-management device for linking and sharing articles and pictures, based on microfiche and cameras. Engelbart read this article as a radar technician in the Philippines, put the ideas together with the radar screen he used, and realised that computers didn't need to be used just as calculators to figure out ballistics--they could be used as personal helpers, in collaboration with people in an interactive way. The 1968 demo included co-working, hypertext, links, outlines, cursors, video conferencing, and the mouse.

Another book I have is Lion's Commentary on Unix 6th Edition (1996 reprint). It circulated illegally for some time, only being published 20 years after it was written. The whole operating system is short - less than 9,000 lines - and it rewards reading. Want to know what a process is? Here you go. And a file? It's in there. Magical.

And so three questions:

  • What other seminal texts are there?
  • What should I investigate further? I should take time to look into Lettvin's What the frog's eye tells the frog's brain (1959) with with Maturana, McCulloch, and Pitts. The latter two mathematically modelled the neuron, and fed substantially into cybernetics. Cybernetics was heavily influenced by anthropology, and also by the feedback mechanisms in anti-aircraft guns in World War II (itself modelled on brains, perhaps?). The theoretical structure of the brain inspired the earliest computer architectures. It's all tangled up.
  • What is there, from the last 10 years or so, that will have this impact? Ben Russell's Headmap Manifesto (1999) is on my hard drive.