On Dec 30, 2007, at 4:03 PM, Bill Wilkie wrote:
Howard:
You bring up a very good point. I was thinking as I followed this
thread, that when I supported a large group of Cobol programmers
that all of the posters thus far had a valid point. At one point we
too said there was no need to use the display, or something we saw
once in a while, Exihibit-Named to print the error data. But we
also realized that there ARE some conditions that perhaps cannot be
easily detected in an edit and decided that it was better to
capture the data when it occurs rather than to have to re-create it
later since enough of the information may already be available to
resolve the problem. Of course it should not be used
indiscriminently. One condition I remember seeing was assembling a
record from multiple indices of a highly normalized database which
would be difficult to detect in an edit.
Just my .02 since I am partied out and decided to check emails.
Just a thought
Bill
Bill (and others):
I think we are pretty much agreed that *an* exception Might be OK to
be displayed but I think we are all in agreement that the console is
NOT the place to display it. Between SDSF (or another product) or
viewing it from archival process is the accepted way to do it. When I
say "AN" that means to me ONCE (but not in the console).
Ed
----------------------------------------------------------------------
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
