Roy Schestowitz wrote:
> So we all know that *nix environments are far more stable than many
> alternatives (I dare wintrolls to argue against it) and we also know about
> further advantages due to the diverse amount of software, which can be
> installed without monetary constraints. That given, in *unix-oriented
Such a novel approach, concede that Linux and UNIX do have far better
"Real World" TCO, then claim that this is a problem.
> * There is less need for code to be written and
> customised in the enterprise. Everything is in
> fm.net or sf.net already.
This is true. There is much LESS code that needs to be written for the
> * There is less of a need for IT staff to ever
> get involved in operation. Servers just don't
> require critical maintenance (updates) and
> remain up for months, if not years (reboots).
This is also true, most of the routine functions can be fully
automated, scripted, and run using crontab jobs or event driven
triggers. This does significantly reduce the cost of routine
> So, Linux sucks, right? It makes people redundant as it simply has the
> ability to cope on its own without outside intervention. I know many
> organisations that need to demonstrate 'competence' by proving that they
> throw some money (budget) at something, as well as provide their employees
> with chores, or else they would become redundant and bitter.
Actually, it just means that you have about 4-5 times more IT budget
available for HIGHLY LEVERAGED projects. Since you have to spend less
time doing routine stuff, you have more money available for the
projects that provide REAL VALUE to the corporation. Money that used
to be spent on "Windows Box Booters" can be spent doing Enterprise
Integration, Supply Chain B2B Integration, Customer Relationship
Management, Business Intelligence, and other high value initiatives
that directly increase the profits and profit margins.
If I spend $5 million on "Box Booters" each year, and all they do is
keep the boxes from crashing by rebooting them regularly and running
routine maintenance, then there is no net value. Worse, reducing the
number of "Box Booters" in such an unstable environment could reduce
margins. Cutting $1 million in "box booters" in a Windows server farm
could result in outages costing the company $5 million.
On the other hand, if I switch to the more stable and automated *nix
environment, I can not only free up $3-4 million in "box booter" budget
(assuming that we will still need some box-booters for remaining
Windows servers that have painted themselves into the corner.
If I can invest that $3-4 million into strategic projects, such as
those listed above, I could increase the profits of the organization by
$20-40 million each year through reduced costs, reduced delays, and
better ability to predict and accomodate surges in customer demand.
Of course, this means that some of that $20-40 million is likely to
come back into the IT department to fund more strategic projects. Show
me a CIO who doesn't have a laundry list of strategic projects he'd
like to fund, and I show you a CIO who will soon be unemployed. The
job of the CIO and CTO is to anticipate the needs of the business and
to know how best to accomodate those needs before their competitors do.
> Corollary: Windows makes everyone happy [sarcasm]. Budgets get burned and
> people do not get bored. I sure know I get bored when the portmaster stays
> up for over 100 days. That's why I'm here all the time chatting...
Well Roy, you might be ready for that climb up the corporate ladder if
you can spend some of that "chat time" focusing on what your corporate
users, customers, and other computer users needs and wants. That can
still leave you with enough time for a few posts a day.
Take some time to look over the shoulders of the "Line" IT workers, the
ones who work for the managers, the managers themselves, and especially
the people who deal with the customers over the phone or in person.
With just a little observation you will find huge opportunities for
*nix on the Desktop as well as highly integrated architectures among