This long-lived thread has, as usual, become a vehicle for the espousal of
hobby-horse views that have little pertinence to its notional subject.
The difference between the capacity of a halfword-length prefix and the
by-fiat maximal length of 100 for PARM= values has always been ugly, and it
should be eliminated.
Some bad code will in consequence cease to work correctly, but that is no
bad thing.
I doubt that making it 2**15 - 1 = 32767 will bring great trouble in train
except in old HLASM routines---I assume that any change will be coordinated
with changes in the support for retrieving PARM= values that is provided by
the statement-level languages---and I should not really object to making it
equal to the '21st century' unsigned value 2**16 - 1 = 65535, although the
argument that unsigned values are uniformly better than signed ones strikes
me as bizarre, even unseriosche.
Any such change should be made to a value that is larger than life, one that
will not have to be revised upward again before all of the documentation
changes it makes necessary have been made.
John Gilmore
Ashland, MA 01721
U.S.A.
_________________________________________________________________
Don�t just search. Find. Check out the new MSN Search!
http://search.msn.click-url.com/go/onm00200636ave/direct/01/
----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO
Search the archives at http://bama.ua.edu/archives/ibm-main.html
