[2170] in Commercialization & Privatization of the Internet
Re: Low end users
daemon@ATHENA.MIT.EDU (Richard Mandelbaum)
Wed Jan 29 08:19:47 1992
To: Dave Crocker <dcrocker@mordor.stanford.edu>
Cc: rma@tsar.cc.rochester.edu, HORN%athena@leia.polaroid.com, com-priv@psi.com
In-Reply-To: Your message of Tue, 28 Jan 92 21:40:43 -0800.
Date: Wed, 29 Jan 92 08:15:09 -0500
From: Richard Mandelbaum <rma@tsar.cc.rochester.edu>
Right. This also has some bearing on the mainframe vs. workstation
debate that was going on a while back. PC's and workstations
are enjoying part of the runaway success that they are enjoying
because (as yet) there seems to be no additional infrastructure
costs. I venture to guess that if one really did abolish all of
the dinosaur computer centers that every one so devoutly wishes
would go out of existence then in a short time one would find that
the support structure for the PC's/workstations would have expanded
immensely. It is very very hard to optimize support services, especially
in University environments where faculty expect immediate service
as soon as they have a problem. Given that it is even difficult to
staff for average load conditions rather than peak load conditions.
Richard
____________________
Let me stress the importance of Richard's explanation about the
source of cost for a service, especially an infrastructure service.
Support and administration are killers. They eat away at operating
margins at a phenomenal rate that simply is not imaginable to anyone
(or at least THIS one) until they participate in developing or
operating one.
While not trivial, the actual costs of the relevant technology usually
are highly manageable and subject to reasonable styles of optimization
.
Optimizing (i.e., minimizing the cost of) support is hugely difficult.
Dave
