🚀 go-pugleaf

In article <1992Dec26.110116.1258@netcom.com> strnlght@netcom.com (David Sternlight) writes:

>Steve Gardner writes:
>"Imagine having to pay Newton for the 
>        algorithm for calculating the derivative of a polynomial."

>Actually, Sir R.A. Fisher did something somewhat analogous to that
>(the publication/ academic equivalent). He invented the method of
>maximum likelihood, a powerful general method for deriving all sorts
>of statistical tests, and kept it a secret. He then derived one test
>after another, published each one separately in the journals, and got
>payment for each usage of the method, in effect, in the form of kudos
>if not money.

This is the utter nonsense put out by the admirers of Sir Ronald who
keep blissfully unaware of what everyone else knew and did.  I do not
doubt that he rediscovered maximum likelihood, and that mathematical
statistic was in the doldrums at that time, but there were those who
knew the at least century old method, and were using it quietly.

He wrote mainly in journals not read by those with mathematical knowledge,
and especially by the few who were aware of what had been done to apply
mathematics to statistical analysis at the time.

RSA may seem miraculous to those who do not know mathematics, but not to
those who do; I suspect the Patent Office to be staffed by those in the
former class.  A BS in mathematics does not mean knowing much, if any,
non-computational mathematics, and certainly not the number theory and
knowledge of computational complexity needed to appreciate what has been
done by the promoters.  As a professor for a long time, I must question if
many are really "skilled in the art", as the patent laws put it.
-- 
Herman Rubin, Dept. of Statistics, Purdue Univ., West Lafayette IN47907-1399
Phone: (317)494-6054
hrubin@snap.stat.purdue.edu (Internet, bitnet)  
{purdue,pur-ee}!snap.stat!hrubin(UUCP)