[ale] Re: [ale-unemployed] Mission Statement / Business Model

[Greg]

I have just read a book from a consultant that does just that  - rips out MS
to put in *nix based diskless systems w/ nice flat screen monitors for
everyone  running Star Office on a server.  I also like the easy factor
here - ALL info is now able to be backed up and the users have less chance
to put the "virus of the week" on their pc's.  The chance of disgruntled
employees formatting their box and wiping out 6 months of work (like what
happened to a friend of mine) is non existent.

The problem is that IT managers don't want to see their domains diminished
and subsequently their status.  Many would prefer to be manager A than B.
It boggles the mind, but business rewards the one with the biggest fiefdom
and penalizes the more efficient IT manager.  Unless the senior management
is pushing this and an uncooperative IT manager is taken out of the equation
(they often undo the consultant's work after he leaves or sabotage it), the
concept is doomed to failure.  Of course, with smaller companies this might
work, but I don't know.

[manager A] Is seen to be constantly in motion as his dept is putting out
MS - caused fires, constantly upgrades his system to run the latest and
greatest - spending mucho dinero for bloatware, extra machines to back up
the MS servers, and has a multitude of lackeys to service the MS server farm
he needs to support a business of 50 employees.  Also spends big bucks on
system mgmt software for licensing etc etc... and gets a 95% uptime for his
efforts.  Needless to say, no one is calculating the costs to the business
of the down time and the Total Cost of Ownership of software (i.e., needing
bigger PC's to run it).  Lackeys are often menu - focused and don't know how
the underlying technology works because MS "does it all for them".

[manager B] Is not seen because the IT infrastructure never breaks, uses old
machines to run all the needed services, needs only one guy working a few
hours a week and has a budget roughly equivalent to the amount spent on
office supplies.  Only upgrades to increase robustness, uses common
standards as opposed to proprietary technology and does not worry about the
license police. Can adapt to mixed environments (windows & Macs - no
problem).  UNIX technology has been in place for awhile and so manager B
doesn't have to worry about constant training.  Operators are command line
based and know the 0's and 1's of how IT works.  The mass of computing power
spent on fancy GUI's is now spent supporting the business.  Operators can
code at some level.

Despite this, if anyone wants my help in building a "proof of concept"
system to show how this works I would love to help, as I use this stuff
(manager B) at home.  Nothing would please me more than taking a room full
of MS servers and putting them all on one box. </end dream sequence>

Greg Canter


-----Original Message-----
From:	jeff hubbs [mailto:hbbs at mediaone.net]
Sent:	Saturday, February 09, 2002 11:15 AM
To:	Irv Mullins
Cc:	ale at ale.org
Subject:	Re: [ale] Re: [ale-unemployed] Mission Statement / Business Model


> FWIW: some of my more computer-savvy clients have been asking about
> converting to Linux, without me mentioning the subject. Some people DO
read
> the news.
>
> Regards,
> Irv

I have been wondering if this has been happening in reality.  It sure
seems like a question that SHOULD be being asked by anyone who wonders
if they really need to be spending all that money.

Sorry I didn't make the meeting; I will try to make the next one.

A capability that I personally want to develop involves the creation of
business-type IT resources utilizing older PC systems (for the sake of
discussion, I'll call "old" anything that predates the PIII).  Several
things drive me in this direction.

For one, even though I am not a rabid tree-hugger, it bothers me that MS
OSses and software seem to drive a cycle of diesktop hardware upgrading
that is resource- and money-wasteful.  I've almost gotten to the point
where I feel like a PIII on every desk in most office situations is
egregiously wasteful.

A little principle I "invented" (I put that in quotes because I don't
think I could possibly be the only person who came up with it) when I
worked for teh US Department of Energy is that of "returned value."  A
desktop PC costs a certain amount of money X.  To get X in returned
value, tha machine needs to be 1) on all the time 2) busying its CPU(s)
at 100% doing gainful work 3) have its drive be about 75-80% full for a
good reason with a lot of I/O going on 4) have plenty of network traffic
going on.  Oh, and the machine must run until it breaks.  To the extent
that those characteristics aren't being met, you've got some
semi-arbitrary reduction down from X in returned value.

It shouldn't be too hard to imagine that in most cases, given that MOST
computer hardware is desktop hardware, companies get ludricously low
amounts of returned value for their financial outlay for IT.
Furthermore, what I'm saying runs counter to how bean-counters handle
computing resources (w.r.t. depreciation, etc.); in fact, it says that
what bean-counters do is flat-out wrong-headed.

Anyway, my point - and I do have one - is that when developing IT
resources for an organization, something needs to be done to preserve
and/or maximize returned value.  This means that if a company presents
an office full of perfectly functional desktop machines that happen to
be P90s, then a solution that calls for the replacement of the P90s with
1.5GHz P4s or what have you is no solution at all, even if you could get
the cost of the new machines very low.

Acknlowledging that StarOffice on a P90 is probably not pleasant,
thoughts must turn to remote execution and treating the desktop machine
like an X terminal, perhaps going so far as to eliminate disk drives at
the desktop altogether.  If you can achieve this, you wind up
trivializing the desktop hardware, making it highly interchangeable and
disposable.    Thoughts turn to making very cost-effective and robust
file servers and application servers.

So, if this "cooperative" wants to offer something that would really
impress the typical business owner, come up with a pre-engineered
solution that primarily utilizes existing hardware of most any age and
covers 90% of the company's IT activity right out of the box.  I think
that among all of us, we can probably agree on what the final result
should look like and make it so that it can be ported to all kinds of
hardware.  Companies that utilize this shouldn't expect to get away with
paying nothing for hardware - if there isn't a decent server with
hot-swappable disk drives, then one has to be bought or built - but, my
gosh, compared to the money that companies are paying now for the pretty
boxes from Dell, etc. and the MS tax, how bad can that be?

- Jeff


---
This message has been sent through the ALE general discussion list.
See http://www.ale.org/mailing-lists.shtml for more info. Problems should be
sent to listmaster at ale dot org.




---
This message has been sent through the ALE general discussion list.
See http://www.ale.org/mailing-lists.shtml for more info. Problems should be 
sent to listmaster at ale dot org.