Home Messages Index
[Date Prev][Date Next][Thread Prev][Thread Next]
Author IndexDate IndexThread Index

Re: [News] FLOSS Offers Short- and Long-term Savings

Roy Schestowitz wrote:

> (Proprietary software:) Pay a little now, pay a lot later
> 
> ,----[ Quote ]
> | It's often prohibitively expensive to alter a company's computing
> | infrastructure once it has been established, so choosing well from the
> | first day is critically important. That is one of the main reasons why I
> | could never recommend a proprietary system to a business owner.
> `----
> 
>                 http://www.freesoftwaremagazine.com/node/1666

That will be the main sticking point, even for a fairly average company you
could be heading towards the 1/4 million mark to entirely rewrite your main
system. Plus all the problems that go along with a change over.

Linux though can make that easier, when coming from UNIX I mean, it wont be
painless and still unlikely to be cheap. But it allows one crucial extra,
you do not have to migrate everything in one go. You can plan a staged
transfer.

Also as more of the majors in the Dbase world come onboard then we can take
on more of these systems.

You have to be even more carefull and detailed in the Planning stage though
than for a UNIX->UNIX move. Also think a few years ahead, will you actually
want the entire system on a single server, or would it be better to take
advantage of Linux's special talents and spread a major dbase application
across many servers, and other such questions. Thinking modular can help a
great deal, you have no choice but to examine every module (whether a job
function or a dbase function) anyway, so a good opertunity to question past
practices and concider whether that was the right way to do it.

I would say that a small to medium sized company requires probably a year of
planning and job function study. Then it could be up to a year of diagnosis
and rewriting to come up with the first modules.

The point being to make the move painless and to work in such a way that
they will be no down time caused by the change over.

So, is the target worth the effort? Yes, even looking at the surface costs,
the cost of Linux and the machines that Linux will sit on, you can estimate
for a much cheaper anual budget. I took this companies anual computer
budget from around Â89,000 down to Â15,000. Plus huge savings on insurance,
maintenance agreements (though you still need some)  and even then I am
often asked to spend Â8000-10,000 around the end of the budget period. It
can be hard to spend that much when you don't actually need anything (and
would you believe, I still don't have a decent printer in my own office).

We have zero down time. Used to have some, we got hit be lightening three
times in the same place, was an anual event. The first time it took out
everything, that cost us a fair bit. So we put in lightening buffers etc,
issolated everything network across the whole site. The second lightening
strike took out a couple of the network switches. Didn't cost much to
replace those, but it hilighted a weakness which our electricians sorted
out. So the third strike did nothing other than blowing some of the
buffers, which is what these buffers are for.



[Date Prev][Date Next][Thread Prev][Thread Next]
Author IndexDate IndexThread Index