Before would seemingly agree with some follies going on here: I
believe, like I've written for solid reasons, that the normal
Information Theory that led to a theoretical underpinning of various
interesting EE activities since long ago, is solidly understood by it's
makers, and when rightly applied is correct in the sense it was intended.
Think of it that a lot of statistics are done under the supposition that
the users of the formulas are done with the right assumptions, like that
the law of big numbers applies, in case there are statistical variables
within some form of well known statistical definition.
So it isn't right to say " so and so many bits contain so much
information ", or "don't contain so and so much" in the sense of the
"normal" formal I.T. because for the normal law there is the assumption
of the repeated experiment (as is normal and required) it isn't decent
to ADD ASSUMPTIONS to the theory, which would lead to statistics with a
P(A | ("given") B) type of formulation, because that leads to erroneous
results, just like assuming something is meaningless by some factor
outside of the theory's definition.
Maybe Vampiers or sociopaths (seriously, some should take a simple test
in that direction....) want fudder that allows them to talk all kinds of
directions as their mind sees fit, for some clearly opportunistic and
wobbly purposes, but abusing formulas and theories in such manner cannot
turn out successful in science without a great many eyebrows getting
raised, me thinks. Like why would professionally self-respecting
scientists need to worry about colleagues as to use 20 character
passwords based on analog random data? What a waste of usually state
funding, isn't it ?
So once more, it is possible to reason along the lines proposed here by
some, but the statistics that should be utilized to stay away from
incorrect or downright ridiculous conclusions are *given* probability
variables/densities/reasoning rules, or the whole thing becomes a laugh.
T.V.
--
dupswapdrop -- the music-dsp mailing list and website:
subscription info, FAQ, source code archive, list archive, book reviews, dsp
links
http://music.columbia.edu/cmc/music-dsp
http://music.columbia.edu/mailman/listinfo/music-dsp