[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

RE: Two technical questions on IEEE Std 754-2008



-----Original Message-----
<snip>

        That you don't seem to be interested in using both
        encodings wastes that latter time.  That you don't
        seem to be willing to support decimal even to the
        point of changing the fundamentals of the arithmetic
        WRT NaNs & infinities wastes the rest.
Dan
-----End Original Message-----

Dan,
  I am curious, although COBOL is no longer "strictly" limited to "business"
applications (cf. the "B" in COBOL), this is certainly our historical (and
even today MAJOR) usage.

Can you give me some examples of how/or why a business application might
want to do "numeric" operations on a Nan or Infinity?  As chuck has
indicated, the current draft does let one detect these and to move them to
compatible data items.  What we don't let programs do is use them in
"arithmetic" statements such as the COBOL
  COMPUTE
  ADD, SUBTRACT, DIVIDE, MULTIPLY
  SUM (function and report handling)

This may (or may not) be a difference in what different languages call
"numeric" and what this implies.  If you could give me some examples of what
"functions" (actions, whatever) make sense (particularly in a business
environment) for NaN or INFINITY, this might help me to better understand
your objection.  furthermore, if there are "important" features for these
values that COBOL is missing, I do think the committees would want to know
about this sooner than later.

Remember, COBOL is a programming language that has never supported
imaginary/complex numbers (or arithmetic using or resulting in them).
Similarly, until 2002, if you defined a data item as
    PIC S9999 Usage Binary,
the maximum "valid" value would be 9999 and NOT a binary value that could
fit into the storage that could hold 9999.  In other words, COBOL has been
(and mostly continues to be) a "the data processing world is decimal, not
binary or any other base".

I think the inclusion of both the decimal and binary floating point
interchange formats and the possibility of arithmetic that follows the rules
for decimal floating point arithmetic (in most cases) from 754, is a
substantial improvement/enhancement.

However, I am quite serious in wanting to know what FEATURES you think we
are missing that our users will/would want.

P.S.  My personal guess is that if any of these features ARE of significant
interest to COBOL users, that implementors will provide them as extensions -
until the next revision of the Standard. It is only those features of
minimal or no use that tend NOT to get implemented - Especially if the
environments provide either software or hardware support for "fully
conforming" 754 support.


754 | revision | FAQ | references | list archive