[Squeakfoundation]Why grants and software go together

Paul Fernhout squeakfoundation@lists.squeakfoundation.org
Mon, 28 May 2001 01:27:11 -0400

Several people on this list imply there is money just begging to be
raised and spent on open source Squeak related things. They may well be
right, and such statements may be even more true as time passes. Here
are some ideas to consider using when persuading foundation program
officers to lobby their funding boards to donate funds to a
Squeak Community Foundation. It's a little political, but so are some of
the points that need to be made to make the argument that funding
free software will produce good things for society.

In general, there are many reasons "open source" or "free software"
projects might be funded as charitable causes. Here are some interesting
comments on this (written as a semi-open letter to Michael Phillips):
General statements of the public value of open or free results can be
found through on the pages of these two major organizations:
If anyone has any other pointers to links, I'd love to see them.

If you read any of Richard Stallman's philosophy at all [many people
criticize him without reading his works]
you will see he argues that the most important aspect of "free software"
is that you are free to adapt it to your needs -- not that you get it
for free or that it was made by volunteers (neither of which may be
true) or even that is under the viral GPL (which is not required for
software to be called "free" by his definition). 

The problem with the current funding approach for most software
development from Stallman's point of view is that as a side effect of
creating an artificial scarcity for their product (to encourage users to
pay for it), software vendors restrict users' ability to make and
redistribute derivative works. This limits the individual's
opportunities to improve the software to meet their own needs. It also
limits the community's ability to evolve software through cooperation.
It doesn't matter how efficient the current system may or may not be --
the deeper point is the desire for this "freedom" to adapt things to
local needs and to improve them and share improvements in a community. 

That limitation on collaboration imposed by proprietary software wasn't
as significant before the internet became as widespread and heavily
used as now. Before the internet, is was very difficult to do any sort
of fine-grained collaboration with anyone unless you were either in the
same building with them or were working at a large organization with a
private network infrastructure. Thus, groups collaborating as closely as
those in the Squeak community simply could not form or sustain
themselves on a practical basis (outside of large organizations or
universities, which is where much free software originated pre-web).

That limitation on derived works has a negative effect on service
businesses, because they may be unable to tailor software products to
meet the needs of their clients. Imagine for example if your lawyer
told you they could not modify their boilerplate contract for your
needs, or your doctor said they could not alter the dose of a drug to
your exact physiology. 

The issue also isn't just what a service person can do at the moment.
Service companies often make specific alliances with vendors of
proprietary solutions to sell their tools in a particular year. The
deeper issue is also what exactly the service provider has spent years
using and learning. Without widespread open solutions available for easy
study, it is less likely there will be knowledgeable and experienced
service people who know of a wide range of open solution that may be
tailored to your needs. It is also less likely they will be familiar
with them from personal experience. You probably won't find another
similar vendor around the block selling the same proprietary solution
modified in the same ways to give you a choice of service providers. The
point is -- proprietary solutions can have a negative effect on people
getting their needs met even when they are willing to pay money.

There are also larger overall societal costs to proprietary solutions.
Essentially we are otherwise heading towards Digital Rights Management
software on every computer. Charging for every use of every file is like
making every road a toll road. Lots of things can go wrong with such
systems. I discuss some of them in this post to gnu.misc.discuss where I
mention four things that went wrong during a recent drive I took down a
toll road:

Stallman would argue these are more subtle effects of proprietary
software that makes criminal the neighborly impulse to share
software. This is potentially corrupting morally. It can promote a
disrespect for the law. It is also slowly leading to draconian social
controls starting with the DMCA (unauthorized copying can now be a
felony instead of just incurring a civil penalty). It is also justifying
increased surveillance and decreased privacy. The Prohibition of alcohol
in the 1920s has a similar effect on our society, and eventually those
laws were repealed and alcohol became a legal, but somewhat regulated
and highly taxed drug. Unfortunately, the widening use of wiretaps and
other new police methods that weakened privacy in the United States
during Alcohol Prohibition remained after prohibition ended.

Both the medical and legal professions are highly paid, and both entail
professionals understanding a large body of generally public domain
knowledge and applying it to client's specific situations. Both doctors
and lawyers create new knowledge that is effectively put into the public
domain in the form of medical journal articles or court proceedings. To
help a lawyer to understand free or open source software, just ask her
or him to think about it in terms of the law itself -- from court
proceedings to legislative records. While lawyers may pay for a service
like Westlaw service for convenience or practical necessity, 
they are not paying to use the law itself, say when they make an
argument in court.

Surely nobody would suggest the world was better off in the days of
18th-century England when a medical student had to crawl on top of a
roof and look in from a skylight to find out the proprietary technique
used by one group of secretive obstetricians to have a lower rate of
infant and maternal mortality than their competitors:
Yet, people do die when software fails -- on airplanes, space stations,
cars, telephone networks, and medical equipment. Is it unreasonable to
hope someday the software "profession" might operate on a professional
basis more like doctors and lawyers do now?

While the the companies selling proprietary software may lose funds if
more software was developed on the basis of grants, the individuals and
companies who know how to work within the common software infrastructure
will be in demand. Right now, if you can believe the Linux trade press,
Linux skills are in widespread demand at high pay rates across the
country. Proprietary Unix skills (outside of a few specialized places)
were never in quite as widespread demand like this across all types of
products (embedded, desktop, server, mainframe). The same thing is true
for, say, proprietary VisualWorks skills. They are only in demand in a
few specialized places -- not everywhere in every industry. An open
Squeak, widely adopted, might create a widespread demand for Smalltalk
development experience and related consulting opportunities. While it is
a compromise, if you can spend half your time consulting and half your
time writing your own free software (and be twice as productive because
of access to other's work to build on), and earn from your half-time
consulting as much or more (because you and your client are more
productive) than when you tried just to sell proprietary software,
everyone might be better off. As someone who has tried to sell software,
I can definitely say consulting pays a lot better, and most of the other
people I have talked to who sold software have echoed this as well.
(Obviously a few people have succeeded wildly at selling proprietary
software, but most have not. Still, like with gambling, the allure

For software development processes funded directly by grants, where the
result was open and free, then even if you were not the one to get the
grant, you could at least benefit from the results and improve them to
meet your needs. So even if the granting process was unfair (and most
probably are for whatever reasons, typically because peer review rarely
funds the innovative proposal from the unknown inventor), non-grantees
at least would still benefit somewhat from the end product of the grant,
because they could build on that end product. (Currently much software
written from grant funds, say, at universities, is still allowed to be
considered proprietary by the developer, but that's another sad story.)

If grants for free software make sense, then grant requests to fund
services assisting developer communities, like a Squeak "Community"
Foundation, should be even more attractive to potential donors.

I feel things have come a long way from, say, 1997, when in response to
a grant pre-proposal to the NSF discussing creating a community of
developers around an open source version of our garden simulator, the
response was: "We don't understand why you don't want to sell the
program commercially". [We discussed exactly that issue in the

(By the way, that simulator project is still one I would like to move to
Squeak, and the Delphi source is available on our site if anyone is

-Paul Fernhout
Kurtz-Fernhout Software 
Developers of custom software and educational simulations
Creators of the Garden with Insight(TM) garden simulator