Re: What is your philosophy? Tracking or Static?
> From: "Nate Hayes" <nh@xxxxxxxxxxxxxxxxx>
> To: "Dan Zuras Intervals" <intervals08@xxxxxxxxxxxxxx>
> Cc: <stds-1788@xxxxxxxxxxxxxxxxx>
> Subject: Re: What is your philosophy? Tracking or Static?
> Date: Wed, 25 May 2011 10:25:33 -0500
>
> Dan,
>
> It is easy to "wish upon a star," as they say.
>
> Feel free to present an implementable method for this
> magical solution... I should very much like to see it!
>
> Nate
Not at all.
Now watch carefully.
I have nothing up my sleeves.
The decoration for a function describes the meaning
of that function. If the operands are outside the
domain of the function it is marked as such. If the
operands include a discontinuous portion of the
domain, it is marked discontinuous. If the operands
are contained within the domain of that function then
it is marked with the decorations of its operands
(min or max, as the case may be) if & only if those
decorations are still relavant to the interpretation
of this function. For example, if an operand to an
add is marked discontinuous, the add is so marked
because add does not change that fact. If an operand
to a max is marked discontinuous, the result is so
marked if that operand (or some portion of it) is
selected for the result. Otherwise, if a function is
such that its operands are within the domain of the
function but their decorations have no meaning for
the result, then the result is marked
defined&continuous.
In all cases, and this is ABSOLUTELY REQUIRED for an
FTDIA, if xx \subset yy then the decoration for f(xx)
is no worse than the decoration for f(yy) FOR ALL f().
I'm sure I have badly stated it & missed dotting some
i's or crossing some t's but that's what we have John
for.
That's the magic, Nate.
Not very magic now is it?
At least, Aurther Clarke would not be impressed I think.
Dan