Subject: Re: US Software Patents Hit Record High (fwd)
From: Rob Cameron <rob@riviere-cameron.net>
Date: Thu, 28 Sep 2006 06:46:48 -0700

Simo, I think this gives me a chance to say a bit about my view.

On September 27, 2006 09:43 pm, simo wrote:
> 
> We agree completely, let's not deviate from the point. The point is:
> copyrights or patents? I say copyrights as it works better for many
> reasons, the fact that Free Software in some cases is more difficult I
> think is clear to all by years now. And yet some Free Software projects
> are extremely successful and incompatible with patents as their success
> depend from being easily exploitable by anyone.
> 
> The problem of _who_ is reward is a different one. From the point of
> view of the public interest, it doesn't really matter who profits, what
> matters is that there is creation of new stuff and possibly also of new
> markets jobs, and value in general.
> 
> Patents in the software industry are not an enabler, they are just a
> barrier that closes markets. That's because while in other industries
> the incentive given by the monopoly is necessary, in the software
> industry the cost to develop new programs is so low that what prevails
> is the chilling effect of the monopoly over the incentive.
> 

Copyrights have 95 year terms (or life+70), are enforceable by DRM 
technology, and are automatically granted for every scrap of software.   
I also think that some FOSS projects may have a reason to fear the evolution
of copyright law around nonliteral copying.    Patents have much shorter
terms, are not mechanically enforceable and require an application
and approval process (which needs drastic improvement through
reforms such as the patent peer review project.)   That software
patents exist may also help forestall the intrusion of copyright
law from literal copying to increasingly abstract layers of nonliteral
copying.   

Patents enable IC-like business models:  small technology companies
that aim to develop novel software technology through a collaborative 
open source process, while preserving revenue opportunities from 
proprietary applications and hardware-software combinations.

Substantial investment will be required to develop IC technology
to be commercially valuable.   SIMD programming is hard and 
platform-specific.   Our particular type of SIMD programming is
very hard.   We would value skilled SIMD programmers very highly
and would like to pay them well.  

Without patents, going open source would simply allow big
proprietary software companies to invest resources to beat
us to the punch.    Without patents, IC's only hope would be to
go the trade secret route and develop and market proprietary
applications first.    

If open source business models can sustain a software
group with a few developers paid reasonable salaries,
then the business ought to be able to afford the occasional
$10K for patenting a truly novel technology.   

Software patents are a well-established and growing business
factor in  the US-based IT sector.   Patent reform is desparately
needed, but Microsoft, IBM and other big IT companies will be the 
main drivers.   Anti-patent lobbying may help add grass roots
pressure.    But the effect will be limited if abolition is the
only alternative presented.

We don't wish to debate different hypothetical worlds that
may arise through legislative changes such as term-limited
patents.    Our approach is reform-by-example.    We will
publish our applications and submit to a rigorous peer
review process.    We intend to support progress in science
and useful arts by making any granted patents free for 
research, teaching and experimentation as well as for
open source software development and distribution.   We also
are interested in assisting others who would follow this model.