lynx-dev Between two Worlds (Slashdot posting)

[Brett Glass]

Between two worlds

(Score:1) by Brett Glass on 11:34 AM November 28th, 1999 MST (User Info) 
http://www.brettglass.com/mailbrett.html

Those who see the entire world in terms of open source tend to forget: 
there is a necessary synergism between open source and commercial software.

The relationship -- at least up to the time when the GPL was created -- has 
always been similar to that between pure research and engineering. 
Researchers do their work in a rarefied, somewhat artificial environment 
where the sharing of information -- via publishing -- brings the maximum 
reward. The scientists in this environment know full well that they are 
foregoing some monetary rewards for their work, but are generally quite 
happy to do so -- since in return, they are allowed to live and work in an 
incredibly rich academic "playground."

Once scientists publish their results, engineers who produce commercial 
(or, as Stallman calls them, "proprietary") products may use it to develop 
practical solutions for the real world. These engineers, being outside the 
research "bubble," do not reveal everything they build upon the scientists' 
work (though they may reveal some of it, in exchange for a temporary 
monopoly, via patents). Why? Because in their world, rewards arise from the 
unique value that they add to their products -- value that their 
competitors cannot duplicate.

Richard Stallman worked in a research facility: the MIT AI Lab. However, 
like many hackers who have trouble dealing with the "real world," he failed 
to understand the symbiotic relationship between the artificial, rarified 
environment inside the research "bubble" and the world at large. What's 
more, he was jealous of his peers who left the research environment to join 
companies which made use of what had been done in the AI Lab. (In 
Stallman's "GNU Manifesto," he advocates that high programmer salaries be 
"banned" to prevent programmers from leaving academia.)

Now, the truth is that not many of these people actually achieved great 
wealth. Symbolics -- the AI Lab spinoff whose creation enraged Stallman and 
led directly to the creation of the GPL -- failed miserably. But Stallman, 
in his rage, took it as a given that all of them were going to make tons of 
money. This is reflected in Stallman's writings, none of which recognize 
the fact that most commercial software companies fail. (Of the ones that 
survive, few are more than marginally profitable.)

Feeling that he himself could never leave the research environment (he was 
so inured to it that he literally lived on a couch in the lab), Stallman 
sought -- out of spite -- to "queer the deal," as it were, between the two 
worlds. This was the purpose of the GPL: to prohibit reuse of work done in 
a research or research-like environment -- where knowledge is shared openly 
and the rewards are largely non-monetary -- in the commercial sector. In 
addition, it was intended to undermine and ultimately destroy businesses of 
the type which so enraged Stallman.

What Stallman failed to recognize, however, is that the academic research 
environment is not self-sustaining. In fact, it is artificially created 
(and heavily funded) by government and by commercial interests in the hope 
that all of society can share and use the results. And, yes, some of us 
will use those results for profit. But that's a good thing -- it creates 
money which industry willingly feeds back into the research environment to 
generate more knowledge. And commercial spinoffs from publicly funded 
projects such as the Space Shuttle have improved all of our lives.

The same is true of projects which make use of freely reusable (that is, 
non-GPLed) open source software. We all can spend less time re-inventing 
the wheel, and more time solving tough, new problems, if we can stand on 
other programmers' shoulders -- rather than, as Brian Reid once said, on 
their feet. But, like many companies that build on publicly available 
knowledge, we would not be able to consider doing a highly specialized 
browser unless we can at least break even on our efforts. And this means 
making our product commercial. Given the limited market for our work, we do 
not expect to be able to use the "give away the source/profit from support" 
business model. The market simply isn't large enough. Without closed 
source, we couldn't contemplate the project at all.

In short, the greatest benefit to everyone -- in this case -- will arise 
from a product which is, at least initially, closed source.

Again, there is a symbiosis and a balance here. Giving some of one's work 
away is fine. But mandating that everyone give everything away destroys the 
relationship between the two worlds by attempting to impose the artificial 
rules of one upon the other. Instead, we must recognize, support, and 
sustain the very useful and positive relationship between the two worlds, 
and allow them to benefit from one another.

--Brett Glass