RE: GPL vs. sweatshops (was Re: Bias)

On Tue, 11 Jun 2002, Patrik Stridvall wrote:

> Do not underestimate the value of peer review and community
> debugging.

Agreed.  But the value of keeping strategic code secret does outweigh this.

> So instead of the non-strategic code, you are likely to get nothing.
> Is that better?

This is the Catch-22 of the GPL.  You'll get everything if people use the 
GPL, but you're much more likely to get nothing if you use the GPL.  Maybe 
that's why the truly successful projects _developed_ under an open-source 
model tend to be under truly free licenses? (Artistic/BSD/X11/Apache/etc.)

If you leave out code developed under the FSF (driven by Stallman himself), 
what significant project has been developed under the GPL from day one, and 
achieved significant success?  Linux is the only example I can think of -- 
can anyone offer others?

Now, what company has been successful commercially developing code from 
scratch, under the GPL, rather than exploiting code written by others?
I'm drawing a blank on this one -- any ideas?

> I'm also unsure what you mean by priviliged minority?
> The clients paid a company to write improvements for them
> so they HAVE paid for they get. You on the other hand haven't.
> So I wouldn't consider it a priviliges.

Isn't the GPL community itself a privileged minority?  Others can't use 
that proprietary GPL code.

> The GPL might or might not be a good "development model".

Collaboration generally improves development, and the GPL encourages 
collaboration -- but I'd like to see a viable economic model to fund 
development before I'd call it a good "development model" on the whole.

Note, I'm not saying that there ISN'T a good development model to create 
open-source software and be able to fund it.  But it's getting harder and 
harder to believe the claims of idealists, when they never seem to pan out 
well in practice.  The Sleepycat model has been successful, as have some 
other hybrid models -- but these all tend to rely on proprietary interests 
to subsidize the open-source development.  If the GPL ultimately drives 
these proprietary companies out of business, the hybrid models will fail.

I'd like to see a model allowing as many of the benefits of free software 
as possible, while being a self-sustaining economic model.  I've only been 
able to think of solutions that demand no "free beer" to make it work...

If anyone has any idea how to keep "free beer" and still fund development, 
without relying on funding from proprietary sources or wishful thinking, 
that would be great.  Does anyone have a realistic suggestion?

Deven