On Feb 13, 2008 11:37 AM, SJS <[EMAIL PROTECTED]> wrote: > begin quoting [EMAIL PROTECTED] as of Wed, Feb 13, 2008 at 10:33:10AM > -0800: > > On Tue, Feb 12, 2008 at 10:52:25PM -0800, Andrew Lentvorski wrote: > > > Now, to be fair, if the book really deals with non-linear mechanics, > > > then computer simulation is a pretty good way to go as *all* of our > > > non-linear analysis methods are approximate. However, that's a far > cry > > > from "precise and unambiguous". > > > > No. Sussman isn't claiming the physics theories he presents are perfect > or > > unable to be improved. He is claiming that by communicating said > > theories with software his *communication* of those potentially > > imperfect theories is precise and unambiguous. Big difference. > > I think the comparision to teaching is a better one. > > When you program the computer to solve a problem, you are teaching > a particularly stupid student to solve the problem.
I don't equate programming a computer to learning, the computer is being told what to do. Programming a computer is done to solve a very specific problem, and only that specific problem. The computer can't take what it has been programmed, apply it to something else that looks nearly the same, but is slightly different. I think higher order procedures is trying to accomplish this fact, but only partially. There's going to be some limit to the problem domain that you can't get generic enough and still have the computer solve slightly different problems. > > It's the act that assists in understanding, not the product. > > > Think about it for a minute. It grows on you. > > Like mold? Some mold is good for you... > > -- > A softly creeping slime > All over everyone. > Stewart Stremler > > -- > [email protected] > http://www.kernel-panic.org/cgi-bin/mailman/listinfo/kplug-lpsg > -- Mark Schoonover, CMDBA http://www.linkedin.com/in/markschoonover http://marksitblog.blogspot.com [EMAIL PROTECTED] -- [email protected] http://www.kernel-panic.org/cgi-bin/mailman/listinfo/kplug-lpsg
