Replies to post #20420 on Advanced Micro Devices Inc (AMD)

[dougSF30]

spokeshave, where to begin...

Well, as someone who *has* taken nonlinear dynamics courses,I can assure that there are indeed patterns in chaotic systems. They are called "attractors", and some are quite bizarre, beautiful and unexpected. However, the patterns are generated by the randomness of the nonlinear systems, and they in no way predict the behavior of the system.

Well, I wouldn't say "patterns are attractors", though trajectories near attractors may certainly trace out 'patterns', whatever that vague term means. (Did you really take a course, or did you just read 'Chaos Theory' or some other popular math book?)

But that's beside the point. This was the original claim:

Assertion: "chaos theory suggests that there are patterns in just about everything, no matter how random they may seem."

I am refuting this statement.

I did not claim, "There are no 'patterns' (which is a rather non-technical, vague term to begin with) in chaotic systems."

This only underlines my second point about basic logical reasoning skills.

Yeah, see, that's where you are missing the point. To understand why TA can be an effective indicator, you need to be a student of psychology and sociology, not the real sciences.

I'm afraid you're not reading very closely, again, or you don't understand what the scientific method and experimental design are all about. It's nothing to do with "hard science" versus "soft science". Yes, *of course* it could be a self-fulfilling prophesy, if indeed the various 'practitioners' agreed closely enough on their religion. I've previously stated such.

My point was, REGARDLESS of any explanation one could dream up (human psychology, ... ), one needs to TEST the hypothesis, to see if it is actually predictive. How to conduct that sort of validation to an appropriate level of confidence is precisely what experimental design (and the scientific method) are all about. These folks pick random "method X" from a book they bought at Borders, and lecture on about its use, when neither they, nor the author has even bothered to properly assess its predictive value.

Does that make it clear to you?

Doug

[kpf]

spokeshave

...you need to be a student of psychology and sociology, not the real sciences.

Agree on the first part. Freudian slip on the second? ;-)

K.