Verily I say unto thee, that Doug Mentohl spake thusly:
> How much did your computer cost? $500? $1,000? $2,000? How much has
> it cost you since you bought it? The price of a computer is not a
> one-time expense — it is rather, an ongoing one ..
The following list outlines additional ownership costs.
* Desktop-level Tech Support
* Hardware Upgrades
* Software Upgrades
* Hardware Failure
* Viruses, Spyware, and Other Malware
Which are generally all-inclusive with GNU/Linux distros, but need to be
sourced and purchased in /addition/ to the OS with Windows.
So what are those costs?
* Tech Support: Available for both Windows and GNU/Linux, although in
the former case, one pays regardless of whether or not one actually
receives any support (licensing), whereas in the latter case /paid/
support is fully optional.
* Hardware upgrades: Generally this is OS agnostic, although hardware
may need to be changed if it isn't supported by the OS. Linux
currently supports more hardware than any one version of Windows. The
current version of Windows (Vista) supports the least amount of
hardware out of all the major operating systems. This also impacts
storage and system stability, since old drivers must be uninstalled,
to be replaced with new drivers (which, under Windows, may require
hundreds of megabytes, including the support software). This may also
leave Windows in an unstable state, or indeed the drivers and support
software may not work at all, causing even more loss of productivity
as yet more hardware is sourced, and system issues are resolved (or
more likely, the HDD is simply wiped, and Windows is reinstalled, if
the issues are prohibitively difficult to resolve).
* Software upgrades: In the case of Windows, the OS and third-parties
provide interim upgrades at no cost, and major version upgrades at
cost. In the case of GNU/Linux, all upgrades, interim or otherwise,
are free (note that charges for commercial GNU/Linux distros are
/only/ for support, not software licensing).
* Hardware failures: Again, this is generally OS agnostic.
* Malware: Irrelevant to GNU/Linux, except in gateways on heterogeneous
networks (i.e. Linux box protecting Windows boxes). IOW cost is
dependent entirely on the presence of Windows.
* Training: Mostly OS agnostic, except where users are familiar with
other environments and applications, and need to be "untrained" first.
IOW training is specific to the task, and the tools used to undertake
that task, rather than the underlying OS, which should be transparent
to those users, as far as possible.
* Applications: The vast majority of applications under GNU/Linux are
free of cost, the reverse is true of Windows. Merely installing a
typical GNU/Linux distro is sufficient to provide most (if not all) of
the required applications. Installing Windows leaves the user with a
mostly functionless system, until third-party applications can be
found, purchased and installed. Windows applications come from a
variety of disparate sources, not all of which are automatically
updated, and those update mechanisms are not consistent. GNU/Linux
distros are updated and extended from a single repository, and the
entire system, including the OS and all available applications, can be
installed and upgraded from this single repository ... for free.
Given compatible hardware and competent staff, it's entirely possible to
deploy GNU/Linux at zero cost, other than the (company) time it takes to
actually deploy it. Automation can reduce the time and complexity of
that deployment considerably.
For Windows to attain the same zero-cost level, it would first need to
be licensed for free, and then come in a GNU/Linux-style distro which
included all available Windows applications ... also for free, and
provide upgrades to that software at no cost, in perpetuity. It would
also need to have the same level of hardware support as Linux, including
a wide range of hardware architectures currently unsupported by either
Windows or any of it's third-party applications. Windows' security model
would have to be completely redesigned to attain parity with Linux, thus
precluding the necessity to purchase additional security software.
It's a near certainty that none of this will ever happen.
The best case scenario (in each case) is that GNU/Linux deployment costs
could be near zero, and Windows deployment costs would be initially
moderate, with moderate to high long term ongoing costs in perpetuity.
The worst case scenario (in each case) is that the initial cost of
deploying GNU/Linux (in a previously Windows-only environment) could be
moderate to high, with diminishing costs over the mid to long term,
until operational costs reach near zero, and Windows deployment costs
could be high (infrastructure and hardware), with exponentially
increasing costs over the long term (hardware, training and format
There are other considerations however, such as application availability
and licensing flexibility. The two are closely connected, because Free
Software can be extended by those deploying it, to provide missing
functionality, so the lack of specific application availability can be
mitigated. The availability of certain applications may be desirable,
since the effort involved in utilising them is minimal, but then these
applications may be subject to high and recurring costs, whereas
extending Free Software to match that functionality may represent an
initially high cost, but zero costs going forward (community effort).
Advantage: GNU/Linux and Free Software.
| "Necessity is the plea for every infringement of human freedom. It
| is the argument of tyrants; it is the creed of slaves." ~ William
| Pitt the Younger
Fedora release 8 (Werewolf) on sky, running kernel 184.108.40.206-60.fc8
21:38:00 up 101 days, 5:20, 5 users, load average: 0.00, 0.02, 0.00