Re: IBM driving mainframe systems programmers into the ground
Anne & Lynn Wheeler
ps2...@YAHOO.COM (Ed Gould) writes:
> Scott:
> I agree with you and other posters but I think I have some evidence
> that indicates IBM is essentially attempting to get rid of a lot of
> sysprogs. IIRC the year was 1995 (6? it was early in the SERVPAC
> years). I was attending an IBM class for SERVPAC's in the IBM Chicago
> education center. One of the 2-3 instructors was talking and giving
> the outline of how SERVPAC's work. After his talk he made a statement
> that SERVPAC was the attempt of IBM to either eliminate or minimize
> the systems programmer. I was surprised but it dawned on me that a
> *LOT* (perhaps 20 percent of the class) consisted of application type
> programmers(!!).As the class dragged on it was clear that the people
> that were asking the questions did not have a clue as what
> SYS1.LINKLIB (and other types of system data sets were). The class was
> getting hopeless bogged down as 20 percent of the class kept asking
> less that typical questions to the point of almost not knowing what
> MVS was and how it
early 80s saw a big explosion in mid-range market ... huge numbers of
vm/43xx and huge numbers of vax/vms. there was some SHARE study that
while vm/43xx had edge in price/performance and other issues ... vax/vms
supposedly had significant advantage in staff hrs & skill level
.... that skill level & amount of effort for vm/370 (still significantly
less than mvs) was market inhibitor (& competitive issue vis-a-vis
vax/vms).
there had actually been a proposal for the 43xx percusor (138/148) to
effectively make vm370 appear as part of the hardware (somewhat
analogous to current day LPAR support) ... but it was vetoed by
corporate hdqtrs ... since corporate was going thru one of those phases
to kill off vm370 completely. of course this was just following the
future system project being canceled and their was mad rush to
repopulate 370 product pipeline ... XA & MVS/XA was going to take yrs
yrs starting from scratch ... in fact MVS/XA had succesfully made the
argument to corporate that it was necessary to shutdown the vm370
development group and move all the people to POK in order to be able to
make MVS/XA first-customer-ship schedule.
Endicott eventually managed to save the vm370 product mission ... but
they effectively had to recreate the vm370 product group from scratch.
In any case, starting in the mid-80s ... hardware and operating systems
were starting to become commodities ... and apps & data were starting to
leak off the mainframe ... and MVS ... which had viewed the enormous
staff & skill level for support ... as a positive attribute ... was
starting to view it as a non-competitive cost-of-ownership issue.
at the time, i bucked quite a bit of the conventional corporate wisdom
... by pointing out that hardware and operating systems was becoming
commoditized (and what happens to profit margins as that happens).
--
40+yrs virtualization experience (since Jan68), online at home since Mar1970