Re: Patents [was Re: GPS implants are here... NOT...]

From: Charlie Stross (charlie@antipope.org)
Date: Wed Dec 22 1999 - 08:04:30 MST


On Wed, Dec 22, 1999 at 05:43:10AM -0800, Robert J. Bradbury wrote:
> > (Witness
> > attempts to patent large chunks of the human genome, for example.)
>
> This is interesting, because you would never have gotten the funding
> to accelerate progress in these areas (Human Genome Sciences or Celera
> for example) if there was not a perception that these companies *would*
> have patentable "products".

That's a rather American-centric viewpoint. Over here, we have the example
of the Welcome Trust -- which is pumping huge amounts of money into HGP
research because it's got a duty to provide money to beneficial medical
research in general. This money is going to academic research units
who are specifically _not_ patenting their sequences, but demonstrating
prior art, so that a biotechnology industry can be built on these publicly-
accessible foundations.

Maybe here we're seeing another fundamental conflict; patents were meant
to cover applications, not basic research.

> > Patents and copyright were originally designed as defenses. The problem
> > is that they've been extended and misapplied until they also constitute
> > very handy weapons.
>
> This is a good point. Perhaps it may be the *key* determinant of
> good use vs. bad use of patents. If you are defending your market
> its probably good. If you are trying to prevent the development
> of other markets or trying to extort royalties then they may be bad.
 
Your starter for $64Bn; how would _you_ overhaul the patent/copyright/IP
system so that:

a) It is difficult to use it to prevent market development outside your
   immediate turf

b) It doesn't stifle basic research

c) It rewards inventors (both large and small) by defending their rights

d) It takes into account the infinitely diminishing cost of replication
   of information (i.e. costs of selling an information product are
   front-loaded; the first copy of Windows NT 4.0 ever sold had cost
   Microsoft $200M to produce, but the second and subsequent were free)

> > And I don't know about you, but my idea of _good_ capitalism is a deal
> > where both sides walk away knowing they've benefited -- a positive sum
> > game. There's nothing very extropian against playing the market as a
> > zero-sum or negative-sum game for your own benefit.
> >
> It isn't so much that it is played as zero or negative-sum, but that
> broad application of patents increases the cost of doing business and
> that may slow down innovation or increases costs to the consumer. Its
> great however from the perspective of the ROI of investors holding stock
> in the companies who benefit from broad application of their patents.
 
On the small scale, maybe. But -- call me a socialist if you like --
I have this quaint notion that the market exists for the benefit of human
society as a whole. It's not *just* there to give day-traders a high
ROI; it's there to handle efficient resource allocation and facilitate
the production of goods and services. The patent regime may benefit a
few people at the cost of the many, without actively encouraging
the sort of innovation it was designed for. If that's the case, I
think it needs overhauling or replacing.

> It is probably worth considering that if patents exist and are known,
> they may *encourage* innovation because people just figure out ways
> around them. As the saying goes, necessity is the mother of invention.

Well, I have (tongue-in-cheek) considered filing for two patents! I'd
go for:

a) Patent the use of genetic algorithms to generate the set of all
   possible inventions that apply within a given problem domain (i.e.
   not just a better mousetrap, but the set of all possible better
   mousetraps -- and the set of all possible genetic algorithms that
   generate the set of all possible better inventions)

b) Patent the business practice of moving your headquarters and base
   of operations to a country with more restrictive definitions of
   intellectual property (that do not include business practices as
   patentable inventions) in order to evade liability for violating
   patents of business practices

Think these meta-patents would fly?

-- Charlie



This archive was generated by hypermail 2.1.5 : Fri Nov 01 2002 - 15:06:10 MST