From: Rakesh Bhandari (rakeshb@STANFORD.EDU)
Date: Fri Jan 16 2004 - 14:02:46 EST
>Hello Rakesh > >I didn't get the overall direction of your questions, but I was reeled in to >try to answer: > >>Doesn't the movement of mental laboring >>activities, e.g. writing code, assume the application of the Babbage >>principle to mental operations ... ? > >Programming labour can be divided into specialities just like any other >labour. So there's no difference in kind in this respect. Additionally, the >cost of computers and the requisite software to build new software is very >low. Ian, it will take me some time to think through the rest of your message; it's been a couple of years since I read Harel's Computers Ltd. What Computers Can't Do, which after all is just an introduction. Would you mind my asking which specialities in the writing of code are likely to be outsourced? I remember twenty years ago my father's excitement about how CAD had revolutionized microprocessor design. I would imagine that the tasks turned over to computer checking first had to be simplified? Will much of what is being outsourced to foreign software engineers eventually be automated? I'll talk to my father tomorrow to try to get some idea about this past revolution. Thanks for the as always stimulating message. Yours, Rakesh >It's easy to move software development to other low-wage geographical >areas, on condition that the local workers are well trained. Plus, any >software capital, such as a large and complex software infrastructure, can >be compressed and transmitted to the new location at almost no cost. A more >philosophical point is that, according to the Church-Turing thesis, the >reduction of human mental activity to the simplest operations has been >achieved, which completes Babbage's program. The Church-Turing thesis is >informally the hypothesis that all human mental activity is replicable by a >particular class of machines, of which an ordinary personal computer is a >finite example. This means that all mental labour can in principle be >automated. As you know in practice lots of virtual machines have already >automated many simple mental tasks (e.g., calculators), and some more >sophisicated ones (e.g., machine translation of languages). Turing's concept >of the Universal Turing Machine (that is a machine able to simulate the >operation of all others) bears an uncanny resemblance to a Henry Ford >production line: there is a tape containing bits of information (the >production line) and a head that moves along the tape altering the >arrangement of information (a worker transforming the raw materials into an >end product). I don't think this is a concidence, although apparently Turing >was inspired by the typewriter. Turing's reduction of mental tasks to the >simplest possible operations is not only amazingly abstract it is also >amazingly powerful, a truly revolutionary advance in science. > >This probably doesn't address your question, but talk of Babbage, Turing >etc. presses my buttons! >(As an aside, I seem to recall that Babbage wrote a book on the division of >labour in factories). > >ATB, > >-Ian. > >_________________________________________________________________ >The new MSN 8: smart spam protection and 2 months FREE* >http://join.msn.com/?page=features/junkmail
This archive was generated by hypermail 2.1.5 : Sat Jan 17 2004 - 00:00:00 EST