2 replies to this topic

[Bruce Klein]

The Atkinson-Phoenix Nanotech Debate
Started in June 2003. Last update: July 17, 2003.

William Illsey (Bill) Atkinson wrote a book, , about nanotechnology. In it he was quite critical of Eric Drexler's approach to nanotech, and of Drexler himself. Chris Phoenix (CRN) wrote a review of the book, quite critical of Bill's understanding of the topic.

Bill responded. This touched off an email discussion, presented here for your enjoyment and information. Bill's response and subsequent emails are included in this file. This is the unedited dialog, hot off the keyboard. Corrections, where necessary, have been clearly marked. [Like this --CJP] [or this --WIA]

Email questions or comments to Rocky Rawstern: rocky at future-is-here dot com, and/or Chris Phoenix: cphoenix at CRNano dot org. (Bill asked that his email address not be posted. Chris will stay in communication with Bill and pass on messages if appropriate.)

First, pointers to a few of the highlights. Following that is a chronological list of the discussion. And after that, the emails themselves.

More: http://nanotech-now....tech-Debate.htm

[Discarnate]

Awesome. Thanks for sharing this, BJ - truely illuminating, IMO.

-Discarnate

[Jay the Avenger]

It's interesting to watch two guys fight over the possibility of an assembler, but I would not go as far as to call it illuminating though.


My opinion:

Bill brings up the issue of the fact that there is no such thing as a 'factory that can produce anything'. I do not think this is a good argument to use when you're trying to convince people that an assembler is not possible.

First of all: every manufacturer wants his or her vacuumcleaners, hi-fi-sets, cars, etc, to look a certain (fashionable) way. A factory of everything would be extremely hard to build, since there are so many possibilities to make a product look a certain way. I think it would be more realistic to compare a universal assembler to a software compiler. The compiler turns sourcecode into programs. When it comes to programs, you don't care how they look on the inside (the sequence of machine instructions), you only care about the working of it. This makes things a lot simpler, and that's why a compiler is a very feasible thing: it only works with so many bricks (= instructions) to produce a limitless amount of software programs.

The compiler itself is also software. You use software to build software. Seems like an chicken/egg-problem, doesn't it? Yet, if you go back in time about 50 years, you get to an age where computers took up soccerfields, and their programs were cartboard cards with holes punched in them. Somehow, by scraping together lots of tiny bits of technology, a compiler was built. From then on, the PC-era seriously started to get off the ground. Suddenly, everybody could build software. The compiler has indeed shown to be one of the more powerful tools to make possible exponential growth.

My gutfeeling tells me nanotech will fare the same way.


Second of all: I read about a factory of everything a few months ago. Somebody was planning on building some sort of santa claus-machine, that would be able to build anything.

Who says factories of everything aren't feasible? They've just never been built before, that's all.