Most conversations about the cost of free software deal with its effects
on the software industry. Microsoft people often talk about how much
money the proprietary software industry can add to a developing
country's economy. At the same time, proprietary software vendors tell
us the total cost of ownership (TCO) for their products is often less
than cost of running competing open source products, even though in
developing countries the cost of labor is almost always low enough that
license fees for proprietary software are huge by comparison. All these
conflicting numbers get wearisome. Perhaps we need to look beyond the
software industry -- and beyond software pricing -- to see what effects
free and open source software have on a country's economy.Let's take a $1,000 software budget and spend it two ways:
- As license fees that go to a foreign corporation
- As local spending to modify a free software package
Money sent to a foreign corporation does nothing to bolster the local
economy. Money used to pay local developers gets spent locally on food,
rents or mortgages, services, and in many other ways. The businesses a
local software developer patronizes spend money locally themselves. The
total multiplier effect of that $1,000 may be anywhere between three and
six, depending on the economic model you use, but in any case we're
saying that $1,000 put into the pockets of local software developers is
worth between $3,000 and $6,000 in value to the country where it is spent.
In the case of free software modified where it is being used, the
country doing the spending reaps that benefit. If that same money goes
out of the country as licensing fees, it will most likely benefit a
North American or European country, since North America and Europe are
home to the overwhelming majority of proprietary software companies.
Co-operative tech support
I've gotten all the free software tech support I've needed in the past
seven or eight years from email lists and IRC channels. Several of the
email lists I'm on are primarily tech support for and by people who use
free software in their jobs. And yet, over and over I hear that support
is one of the great advantages of proprietary software. "Ah," says the
proprietary software proponent, "when you buy software you have someone
to blame when it messes up. You have a company that stands behind it."
Has anyone ever read a proprietary software licensing agreement? (Here's
download of Microsoft's license agreement for Windows XP Pro, in
case you haven't.) Have you noticed that you never get any kind of
warranty that matters, or that support beyond basic installation usually
costs beaucoup bucks?
The "companies can't accept free tech support, they need someone to
blame," argument has been done to death, and I have trouble believing
it. I know plenty of corporate IT people who rely on free IRC and email
group support for both free and proprietary software. And in countries
where most proprietary software is obtained illegally, it's obvious that
most proprietary software tech support isn't coming through formal
None of this casual support registers as economic activity, but it
obviously has some sort of value. How do we measure it? Can we measure
it? Should we even try?
Companies that provide software support are as threatened by free
software support as proprietary software companies are by free software,
but we don't have IT support company executives making thunderous
speeches about how helping your neighbor with a computer problem is
going to ruin the economy, do we?
Knowledge is power
I've said the following words (or similar ones) while speaking in
several developing countries:
One reason I support free and open source software everywhere is
self-interest. FOSS is the greatest training tool for programmers ever
developed, and for all I know the next world-changing genius programmer
is here, in your country, not able to afford college. If we put a
computer running GNU/Linux and open source applications in that person's
hands, or give that person access to a public computing center that runs
GNU/Linux, everyone in the world -- including me -- may someday benefit
from that person's work.
There is no reason to believe that the next great genius programmer will
be white, male, and European or American. For all we know, a shoeless
slum child in Brazil, Egypt, or Mexico has more programming talent than
Linus Torvalds. We must make sure she has access to computers, and
specifically to computers filled with software she can take apart to see
how it works, not software whose source code is hidden from her.
Learning how to push buttons in Windows is nice, but it won't teach you
how the software behind the buttons works. We realize
that most users will never look at source code, but it is an invaluable
training tool for anyone who may need to write or modify code one day,
and any school or other public or semi-public computer facility that
doesn't make plenty of open source software available to youngsters who
are interested in computers is short-changing them.
Some of the kids whose eyes are opened by open source will become
programmers. Many -- possibly most -- won't, but will probably end up in
jobs where computer knowledge is useful in one way or another.
Does it matter whether these jobs are in "the software industry" or in
auto parts warehouses, marine biology research labs, and other
non-software businesses? Does it matter whether someone who writes (or
helps write) a program that makes work easier at his place of employment
is called a "software developer" or carries another title?
Either way, the software gets written and used -- and hopefully saves
labor or has other economic or non-economic benefits. And as far as I'm
concerned, this is the point of free and open source software, no
matter what effect its existence has on the software industry as it