New York Times on the Web Forums Science

Missile Defense

Russian military leaders have expressed concern about US plans for a national missile defense system. Will defense technology be limited by possibilities for a strategic imbalance? Is this just SDI all over again?

rshowalter - 08:31pm Jul 16, 2001 EST (#7099 of 7107)
Robert Showalter showalte@macc.wisc.edu

jimmyz211a 7/16/01 3:59pm ... as long as the US and Russia have the nuclear weapons we do, we're on each other's "hit list" -- as pointed out Sunday in Nuclear Arms Still Keep the Peace by ROBERT S. McNAMARA and THOMAS GRAHAM Jr. http://www.nytimes.com/2001/07/15/opinion/15MCNA.html

rshowalter - 08:41pm Jul 16, 2001 EST (#7100 of 7107)
Robert Showalter showalte@macc.wisc.edu

Directories, and a key story:
MD6965 rshowalter 7/12/01 9:22am ... MD6463 rshowalter 7/3/01 7:01am

I'd like to post links to a Guardian thread where I've said many of the most important things I'd like people to know. Psychwarfare, Casablanca -- and terror http://talk.guardian.co.uk/WebX?14@@.ee7a163/0

including the key story, #13.. http://talk.guardian.co.uk/WebX?7@@.ee7a163/13 ... to #23.. http://talk.guardian.co.uk/WebX?7@@.ee7a163/24

note #26 ... http://talk.guardian.co.uk/WebX?14@@.ee7a163/25

To see many references to this that thread, and to the movie Casablanca , search "casablanca" for this thread. Here are some of them:

MD3044 rshowalter 5/2/01 5:31pm ... MD3045 rshowalter 5/2/01 5:31pm

MD3831 rshowalter 5/14/01 12:09pm ... MD3523 rshowalter 5/8/01 4:12pm

Summaries and links to this Missile Defense thread are set out from #153 in http://talk.guardian.co.uk/WebX?14@@.ee7a163/159
MD4778 rshowalter 6/11/01 7:31pm

rshowalter - 08:42pm Jul 16, 2001 EST (#7101 of 7107)
Robert Showalter showalte@macc.wisc.edu

MD3046 rshowalter 5/2/01 5:32pm
In the Guardian Talk threads and in this Missile Defense thread, Dawn Riley and I have worked to focus patterns of human reasoning and persuasion, and problems with human reasoning and persuasion.

lunarchick - 08:57pm Jul 16, 2001 EST (#7102 of 7107)
lunarchick@www.com

Showalter did this open when you tried (it's temperamental): lunarchick 7/15/01 10:19am re longshot ~ http://www.tjhsst.edu/~bgoodman/supercomp/missile.html Go to home on this ... wow ... these guys have a SuperComputer ... anyone got a spare Cray to send my way? Seems there's a lot of calculations re
"USA shot a rocket in the air - it fell to earth they know not where"

lunarchick - 09:06pm Jul 16, 2001 EST (#7103 of 7107)
lunarchick@www.com

An aside here re computers. I heard that Fiji has a computer problem, they've had a spate of problems, but, a computer was stolen. Unfortunately for Fiji it contained their whole national accounting system. That's right! No means of 'looking' that's 'searching' into National Finances!

Now were the same thing to happen with MD !!!

rshowalter - 09:13pm Jul 16, 2001 EST (#7104 of 7107)
Robert Showalter showalte@macc.wisc.edu

From http://www.tjhsst.edu/~bgoodman/supercomp/missile.html

"Computers are slow. No matter how fast they get, we always manage to (rather easily) find problems for them to solve that would take several billion years to complete.

Comment: missile defense is full of problems of this kind -- sometimes one on top of the other -- which is why it is easy, using standard references such as Knuth, to show that the computations that count are impossible even if computers were a million times faster than they are.

"There are many physical laws, all of which involve many floating point calculations. It is thus normally impractical (and luckily unnecessary) to include all laws in all simulations.

Comment: Sometimes it is necessary to combine many physical laws - and when even two laws are in interaction over the same space -- the procedures are set up wrong. http://www.wisc.edu/rshowalt/klinerec But even with that problem fixed, intractably large computational problems still remain -- even if you had the data the computation needs -- and often you don't.

Most physical laws are continuum functions of continuum quantities (they can be measured to infinite precision given proper instruments and can take on any values within their domains). Many physical quantities are quantized (can only take on a set of discreet values within a domain), but the quantization of such quantities is normally on such a small scale that the quantities can often be treated as continuums.

While the mathematics of calculus are often useful in dealing with these continuums, programming a computer to truly (symbolically) evaluate calculus functions is difficult.

Comment: you betcha !

(3 following messages)

New York Times on the Web Forums Science Missile Defense