On 7/16/2013 11:15 AM, R.S. wrote:
W dniu 2013-07-16 19:22, John Gilmore pisze:
R S is right. Constantly accumulating detritus of this sort 1) makes
the mainframe look bad, 2) provokes repeated discussions of this sort,
3) impairs performance marginally on a case-by-case basis but
cumulatively, and 4) is ugly.
And 5. IMHO very important: some day you will HAVE TO DO IT.
I seriously doubt that's true!
There are many installations with IMBED/REPLICATE in their catalogs.
Some of the larger ones asked IBM to explain why they had abandoned
their commitment to upward compatibility and platform investment
protection i.e., what was the _business case_ for customers to expend
time, effort and $ to redefine these catalogs?
[Aside: my unofficial understanding is that IBM wanted these attributes
removed primarily so they could clean up their code to make future
development easier.]
Whatever the "official" justification turned out to be, it was deemed
insufficient by customers and IBM executives alike, and this led to the
"retrenchment" to which Skip referred in an earlier post.
Since then, IBM catalog development has implemented *numerous*
substantive enhancements e.g., RLS catalogs. Would those enhancements
have been implemented sooner if every customer in the world had been
forced to remove IBMED/REPLICATE from every such catalog in the world?
Perhaps. But, we will never know...
--
Edward E Jaffe
Phoenix Software International, Inc
831 Parkview Drive North
El Segundo, CA 90245
http://www.phoenixsoftware.com/
----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [email protected] with the message: INFO IBM-MAIN
