Re: Bisection methods: a use-case example

["Nate Hayes" <nh@xxxxxxxxxxxxxxxxx>]

Michel Hack wrote:
> Nate Hayes wrote (about using a "pragmatic" midpoint2() that returns a
> finite result for an unbounded interval):
> > I think everyone agrees Sally did something really pragmatic, but one
> > might ask the question: is Sally really following the standard or is
> > she just working around it?
>
> Nate seems to have a different interpretation of what a standard does,
> compared to my understanding.
>
> In my opionion at least, a standard defines precisely how the things
> it covers are supposed to behave, and what things must be provided.
> It DOES NOT specify any particular programming style!  The standard
> does not apply to a program, it applies to a programming environment.

I think we have the same understanding, Michel.

For example, the moral of the story is how in both cases the midpoint 
operation is provided, but in Case A the standard environment requires the 
behavior:
   midpoint([1,5]) = 3
   midpoint([1,+Inf]) = NaN
as opposed to Case B where the standard environment requires the behavior:
   midpoint([1,5]) = 3
   midpoint([1,+omega]) = REALMAX
and hence:

Michel Hack wrote:
> The quality of the standard may decide how easy it is for Sally to
> write a correct, efficient and clear program.

Nate