home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!stanford.edu!morrow.stanford.edu!sumex-aim!rice
- From: rice@sumex-aim.Stanford.EDU (James Rice)
- Newsgroups: comp.ai.genetic
- Subject: Re: Kozan Book Order INfo
- Date: 21 Jan 93 16:13:45
- Organization: Knowledge Systems Lab, Stanford University
- Lines: 94
- Distribution: usa
- Message-ID: <RICE.93Jan21161345@hpp-ss10-1.stanford.edu>
- References: <g6f3zcg@rpi.edu> <GRAHAM.93Jan18133302@Xenon.Stanford.EDU>
- <RICE.93Jan18180101@hpp-ss10-1.stanford.edu>
- <1993Jan21.213429.27459@VFL.Paramax.COM>
- NNTP-Posting-Host: hpp-ss10-1.stanford.edu
- In-reply-to: dave@vfl.paramax.com's message of Thu, 21 Jan 1993 21:34:29 GMT
-
- In article <1993Jan21.213429.27459@VFL.Paramax.COM> dave@vfl.paramax.com (David Lee Matuszek) writes:
- In regard to
- Genetic Programming: On the Programming of Computers by Means
- of Natural Selection, by John R. Koza.
-
-
- In article <RICE.93Jan18180101@hpp-ss10-1.stanford.edu>,
- rice@sumex-aim.Stanford.EDU (James Rice) writes:
-
- |> I'm biased, but I'd suggest reading the book and watching the movie.
- |>
- |> Rice.
-
-
- In what way are you biased?
-
- I helped out a fair bit inthe research that went into the book.
- I am a co-author (movie-maker?) of the movie.
-
- I'm thinking seriously about spending $55 for the book (though I wish
- someone would tell us a little more about it), but $35 for a "movie"
- seems awfully steep. What's the point of the movie? What makes it
- worth this much?
-
- --dave
-
- As I've said, I'm biased in that I've been involved in
- these projects I don't get any sort of royalty out of
- either so I have no financial motive to advertise them,
- but we would all be foolish if we thought that every time
- an author cites his own work he is doing so for solely
- altruistic reasons. (I'm evangelical about GP too)
-
- Having said this, I genuinely believe that the book is a
- major work. Some even call it a landmark work. To quote
- John Holland on the cover: "The research reported in this
- book is a tour de force. For the first time since the
- idea was bandied about in the 1940s and the early 1950s,
- we have a non-trivial, non-tailored set of examples of
- automatic programming."
-
- Stewart Wilson said "John Koza has discovered a general
- and robust method of evolving computer programs that is
- effective over a breathtaking range of problems in applied
- mathematics, control engineerin, and artificial
- intelligence."
-
- At $55 for a >800 page book, I call that pretty good value
- for money in this field.
-
- I haven't seen a review for the movie, so I'm unable to
- quote any hyperbole for you. However, I can describe the
- movie a little.
-
- It's an hour long and shows animated visualisations of a
- number (23, I think) of problems discussed in the book.
- These problems include things like the wall-following
- robot, whose behviour is easier to appreciate when
- animated.
-
- The movie was professionally shot, all of the graphics
- were captured with a scan-line converter and are all at
- good resulution (no cameras pointed at the screen). The
- movie took over a week of continuous professional
- editting. I (bias again) believe it to be the best
- researchoid movie of its kind I've seen by about two
- orders of magnitude. Obviously, the average NOVA
- documentary has higher production values (though not much
- higher), but such a documentary has about 30 seconds of
- content. The movie is all content. The length was chosen
- specifically so that it can easily be shown at brown-bag
- lunch events. I have heard tell of lecturers getting a
- copy also to fill in a college lecture slot.
-
- $35 for a way to save you from having to prepare a lecture
- sounds like a bargain too.
-
- Koza's no Paul Newmann, but then he's not on screen much.
-
- Frankly, I'd be more comfortable if someone else gave a
- less biased review, but you understandably wanted to know
- my bias.
-
-
-
- Rice - still trying not to sound like a broken record.
- P.S. Please remember that's genetic-programming-request@cs.stanford.edu
- ^^^^^^^^
- to get on the genetic-programming@cs.stanford.edu mailing list.
-
-
-
-
-
-