random

[Alex Martelli]

"Nick Perkins" <nperkins7 at home.com> wrote in message
news:A3iS6.121059$eK2.28202517 at news4.rdc1.on.home.com...
    ...
> > """
> > A physical RNG can have the property that one cannot make
> > any prediction about the value of the next bit even given
> > _complete_ information about how the numbers are being
> > generated.
> > """
    ...
> This is not a contradiction.
>
> if we define predicates:
> I : have complete information
> P : can make prediction
>
> then the first statement is:
> I implies (not P)

Ah, there's the rub of our disagreement: I read the 1st stmt as:
    not (I implies P)

That is, in natural language, I don't read a statement "the fact
of having complete information implies I cannot make a
prediction" (which would afford some inference, the way
natural language is commonly used, that maybe NOT having
complete information might make things better!), but rather
"it's not the case that having complete information implies
I can make a prediction", i.e., "DESPITE having complete
information, I _still_ could not make a prediction".

"Even given (a powerful army) I could not (defeat Napoleon)".
Do you read this as "having a powerful army IMPLIES I
cannot defeat Napoleon"?  This doesn't sound right to me.
Surely it's "NOT (having a powerful army IMPLIES I can
defeat Napoleon)"?

"Even (using Python) I cannot (make my GUI pretty)".  Do
you read this as it being the fact of using python that IMPLIES
my inability to make GUIs pretty?  Surely the 'Even' removes
all doubt -- if you *didn't* have it in the original sentence
your reading would be quite sensible.  But what is that 'even'
doing there if not EXACTLY impliying the reading is NOT:
    (using Python) IMPLIES (no pretty GUI)
but rather
    NOT ( (using Python) IMPLIES (pretty GUI) )
...?


Alex