Re: RE: [Micronet] Which hosts running older OSes will be blocked by min. stds?

On Fri, Jun 25, 2004 at 03:36:08PM -0700, Mike Hunter wrote:
> On Jun 24, "Tom Holub" wrote:
> >
> > First of all, academic departments don't have much flexibility in their
> > state funds. They are static from year to year, and the only way to
> > "budget" for a new expense is to stop spending money on something else.
> > In most cases, computing is competing with the teaching program, and
> > when departments have to decide whether to cut support for computers,
> > or cut support for teaching, they will almost always choose the former.
> > And even as a geek, I can't really argue that it shouldn't be so.
>
> I understand what you're saying and I agree that computers are a much
> better thing to cut than instruction. But there's still a disconnect with
> the *expectations* people have...if a department said "hey, we've got 30,000
> extra bucks, let's buy new workstations", they **still expect** them to
> work in 4 years with no central support, as evidenced by the desperation
> SNS sees every day from people who are infected and don't know what to do,
> and are expected to fix it (by virtue of no one being available to help
> them.)

I agree that there's a disconnect, but I think it's an understandable
one. Almost anything else departments buy, continues to work in 4
years. It's quite a paradigm shift for a department that's used to
purchasing, say, refrigerators, that to properly manage the equipment
you need to spend an amount as large as the equipment cost, every
year. Department managers aren't reading IT magazines which put forth
the idea of expensive centralized management as a fait accompli.

Even today, when I tell a department it will cost over $1000/year to
support their machine, the common response is "I could buy a new computer
for that much!"

Or on the other side, we support a computer for Daniel McFadden, the
Nobel Prize-winning economist. That computer is over 7 years old, and
he's still buying new replacement hardware for it, because he doesn't
see the value in purchasing a new machine.

> > Furthermore, it is only within the past couple of years that
> > departments have even been told that they have a fiscal responsibility
> > for the support of computers under their control. In 1994, when a
> > faculty member asked for a new computer, no one except IT managers was
> > pointing out the ancillary costs of computer ownership, and most
> > departments didn't have IT managers. Certainly the campus wasn't
> > providing any guidance in this area. Decisions about adding computers
> > were distributed to whomever held the funds--in many cases, faculty
> > members, further insulated from any sort of centralized IT planning.
> > And again, no one was telling faculty members that they shouldn't be
> > doing that, except in those few departments which had creatively found
> > funding to support a centralized IT operation strong enough to fight
> > battles with the faculty.
>
> I can believe that...I still have a hard time accepting it fully....maybe
> I'm just in denial. Don't people own cars? Bikes? Pets? Things that
> you expect to be around for a while have recurring costs.

Certainly there are recurring costs, but not on the same order of
magnitude as with computers. And if you buy a car or a bike or a dog,
you don't have to pay for professional management of it; you just
bring it in when something is wrong with it.

> > So the network grew, not so much in defiance of good practices but in
> > ignorance of them.
> >
> > So we're now at the point where consciousness is being raised about
> > the true cost of ownership of computers, at a time when department
> > budgets are being cut for the third year in a row. Even departments
> > who fully support the idea and have the best intentions will have
> > great difficulty coming up with funding to deal with any mandated
> > expenses.
> >
> > Departments shouldn't be blamed for the situation they're in; it's
> > only very recently that they've been given any guidance on IT
> > planning. And even if they had been given guidance years ago, they
> > still would have needed an infusion of new funds to pay for new
> > expenses.
>
> You have inspired my sympathy :), and I agree with the point raised about the
> amount of valuable work being done on deprecated systems. But the
> internet has been a scary place for more than the past 6 months, and SNS
> is giving people a lot of time IMO to comply. I think compulsion is a
> much bigger factor than cost in fixing this situation.

I think it's not too hard to compel departments to get their primary
machines modernized; in L&S, at least, most are in pretty good shape
already. But if you want to get all the machines used by GSIs, and in
labs, and by work-study students and such, there just isn't
departmental money to upgrade many of them.

-- 
Tom Holub (tom_holub@LS.Berkeley.EDU, 510-642-9069)
College of Letters & Science
249 Campbell Hall
------------------------------------------------------------------------
The following was automatically added to this message by the list server:
For information about MAGNet, its meetings and events, and its
mailing list, including information on subscribing and unsubscribing,
see the MAGNet Web site at <http://magnet.berkeley.edu/>.