Univ. charging $100/mo./person for mainframe access

Richard H. Miller rick at crick.ssctr.bcm.tmc.edu
Thu Mar 2 10:54:47 EST 1995


In article <3j2o89$anb at news.doit.wisc.edu>, dalehn at facstaff.wisc.edu (Donald A. Lehn, Ph.D.) writes:
|> In article <3j0qqc$8rc at news.doit.wisc.edu>, mtkloth at students.wisc.edu 
|> says...
|> >
|> >gold at astro.ocis.temple.edu (Bert Gold) wrote:
|> >>
|> >> One of the universities where I work has decided to charge $ 100 per
|> >> person per month for mainframe access.  As everyone in these newsgroups
|> >> knows, this kind of access is an absolute requirement for doing
|> >> molecular biology in 1995!
|> 
|> You are right.  Internet access is a must.  If the UW started to charge
|> $100.00/month I'd tell them to screw themselves since there are many
|> local net providers who will give you unlimited slip access for about
|> a third to a quarter of this amount.

I thing you misread Mr. Gold said; the charge is $100/month for access to the
university's mainframe system. There is nothing said about internet access. If
this $100/month represents a flat fee for unlimited access then it may be a 
bargin (compared to cost recovery where you pay for what you use) 15 years
ago when we still did cost recovery for scientific computing, many researchers
could easily exceed $100/month in those days when you consider CPU charge, I/O
charges and disk storage charges. Since some charge-backs include connect time
and memory surcharges I can see the cost could exceed that considerably. 

In this case, the university should be able to justify the cost structure and,
I would hope, be willing to share the model used to develop the charge. (We
had no problem in sharing this.)  

|> 
|> I would encourage you to drop a note to NIH about this.  I bet your
|> university is already billing computer access to internet as an
|> "overhead" expense for grants it administers.  I am willing to bet that
|> the govt. wouldn't take kindly to paying for something that isn't
|> delivered.

And I would encourage you to talk with the people first and see if they can
explain the charges and if they are unable/unwilling to explain them, discuss
the matter internally first. 



|> ps. A friend of mine who keeps track of such things estimates that
|> the real cost of providing internet services to everyone on the
|> net runs less than a $1.00/month (actually more like 50 cents/month).

I doubt it, I have done the cost models and unless you have a tremendous
user base, there is no way you can do it for $1.00/month (think about it; 
for say 1200 users you are only talking about $1,200/month and I doubt you
can even pay for the hardware and maintenance costs for $1,200/month much
less the people costs.)



-- 
Richard H. Miller                 Email: rick at bcm.tmc.edu
Asst. Dir. for Technical Support  Voice: (713)798-3532
Baylor College of Medicine        US Mail: One Baylor Plaza, 302H
                                           Houston, Texas 77030



More information about the Methods mailing list