Liste des Groupes | Revenir à theory |
On 05/02/2024 07:58 PM, Keith Thompson wrote:The infinite expressions and completions and closureswij <wyniijj5@gmail.com> writes:>On Thu, 2024-05-02 at 18:02 -0700, Keith Thompson wrote:>wij <wyniijj5@gmail.com> writes:>
[snip]Nothing is different from the math. you understand (except several>
corner cases
which you will never need to worry about).
So there's nothing novel in your notation, and I needn't waste any more
time asking questions about it that you're unwilling and/or unable to
answer.
>
Is that a fair summary?
>
(I'll ignore the "corner cases" you allude to.)
No, nothing novel there changed the usual usage of 'fixed point number'.
I have no idea whether that was an attempt to answer my question. If it
was such an attempt, it failed.
>
The term "fixed point number" is quite distinct from both "rational
number" and "real number". I don't know which you're trying to define.
Fixed point numbers, in my experience, are a computer representation,
not a mathematical abstraction. But your subject header talks about
real numbers.
>
Feel free to attempt to clarify if you're so inclined. Or don't.
>
https://en.wikipedia.org/wiki/Numerical_tower
>
The line-reals are naturals n/d, 0 <= n <= d, d goes to infinity,
it has extent, density, completeness, measure [0,1], measure 1.0.
>
The field-reals are the equivalence classes of sequences that
are Cauchy, and that's the standard definition of the complete
ordered field, including when Dedekind cuts (of rationals) won't do.
>
The signal-reals are as that the rationals are huge,
when doubling them results a continuous domain.
>
>
It gets involved doubling and halving measures and spaces
and real non-standard analytical character and new results
in numerical series and methods in the Cantor space or
"2 ^ omega" of each of these different models of real numbers.
Then standard ordinary set theory is left consistent by just
making another result in set theory in function theory.
>
That _always_ exists.
>
>
Les messages affichés proviennent d'usenet.