FOM: What are the FOM issues in this?

[Martin Davis]

At 01:11 PM 1/31/00 -0500, in reply to a message in which I said in part:

> >For arithmetic operations the IEEE floating point standard is a beautiful
> >accommodation, implemented in most compilers.

Steve Stevenson wrote:

>The IEEE recommendations have have nothing to do with compilers: they're 
>machine
>specs. And check William Kahan's website at
>
>http://www.cs.berkeley.edu/~wkahan/ieee754status/
>to see how big a joke they really are.

I'm grateful to Professor Stevenson for pointing me to (Turing award 
winner) Kahan's site.  I saw no joke, but I did see the following on that site:

"Programming languages new ( Java ) and old ( Fortran ), and their 
compilers, still lack competent support for features of IEEE 754 so 
painstakingly provided by practically all hardware nowadays. ... 
Programmers seem unaware that IEEE 754 is a standard for their programming 
environment, not just for hardware."

So not everyone agrees that "The IEEE recommendations have nothing to do 
with compilers." In fact, isn't it evident that a hardware specification in 
itself will only be of use to a programmer who writes code at the hardware 
level? On the other hand, I was dismayed to learn that compiler writers 
don't automatically regard incorporation of the IEEE standard as necessary. 
My acquaintance with compilers comes from teaching elementary programming, 
and so I knew that the Borland Pascal and C compilers did incorporate the 
IEEE standard (but not in all the "real" types available to the programmer).

Martin



                           Martin Davis
                    Visiting Scholar UC Berkeley
                      Professor Emeritus, NYU
                          martin at eipye.com
                          (Add 1 and get 0)
                        http://www.eipye.com