More interesting than this loss have been the reactions to it.
They have included the usual chestnuts: floating-point arithmetic yields
rounded values that are unusable in business environments; UNIX has an
unusably low MTBF and z/OS MVS a commendably high one; the care and feeding
of 10,000 chickens is likely to be more costly than that of 10 cows; etc.,
etc.
There are cointexts in which all but the first of these notions has merit,
but what has gone without discussion is that the great technical strengths
of the mainframe are little exploited, even actively rejected in most
mainframe COBOL shops.
This old-COBOL subculture has a synchronous, move-orient[at]ed, compile-time
bound world view that judges the asynchronous, list- and
pointer-orient[at]ed, execution-time bound technology of z/OS irrelevant to
its needs.
In the upshot radical change---outsourcing, conversions to UNIX or LINUX,
whatever---is sometimes seen to be and used as a verhicle for dispensing
with/destroying an old, dysfunctional IT organization that is heroically
obstructive/reactionary and risk-aversive.
Mainframes and z/OS thus suffer not because of their own proper deficiencies
but because of their historical association with an obsolete episteme. They
must be dispensed with in order to dispose of their caretakers.
John Gilmore
Ashland, MA 01721-1817
USA
_________________________________________________________________
Call friends with PC-to-PC calling -- FREE
http://imagine-msn.com/messenger/launch80/default.aspx?locale=en-us&source=wlmailtagline
----------------------------------------------------------------------
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
