That anti-patent pamphlet I mentioned

Marcus Brinkmann Marcus.Brinkmann at ruhr-uni-bochum.de
Tue Dec 17 02:04:27 UTC 2002


On Tue, Dec 17, 2002 at 01:30:46AM +0100, Simo Sorce wrote:
> > > How exactly can you tell you're not violating someone's patent while
> > > implementing your "invention"?
> > 
> > You study earlier patents that issued for your particular field.
> 
> Do you have any idea of how much that cost? Who is economically able to
> do that? Are you saying invention should be possible only for very rich
> people and nobody else?
> 
> > I know, today that's almost impossible due to the bad classification,
> > but that's the idea.
> 
> No classification can classify all the human knowledge and make it
> easily accessible ...

Also, it is impossible to make sure you are not violating a patent.  Even if
you go through each and every patent (and you have to because a database
patent might not even mention the word database), you still have to understand them,
which is difficult because of the obfuscated patent language.  And even if
you understand them, you can't really make sure you are not violating it at
all times.  Only a judge can make a final decision.

This is a general reason against patents, but the effect is much worse with
software than with hardware.  One reason for this is that a software program
contains dozens or hundreds of ideas, and so could violate easily a dozen or
more patents.

> > Or you make a reservation and you wait for
> > someone to come after you.
> 
> What business can you build on such basis? It would be too much a risk
> to afford, only real big players with hundreds of patents will be able
> to research once the system is running, as they will be the only one
> that can afford to cross license or afford the cost of the patent
> system.

In fact, even the really large companies probably just wait for you to
contact them, and then use their own patent portfolio to cross license.

Cross licensing means that the two participiants sign a contract that grants
each of them an unlimited license for all patents of the other participiant.
The large companies (like IBM) said this use of patents is 10 times more
worth than to use the patent as protection for your invention.

So, patents are ten times more worth as a protection against patents rather
than as protection of the process they claim.  The only people who can
really be happy about such a screwed up system are the patent attorneys and
the staff of the patent offices, because for them it doesn't matter, as long
as the patents keep flowing, they will have power and wealth.

I think this is a good place to repeat that virtually nobody except patent
attorneys, the patent offices and two large companies (Microsoft and IBM)
are in favor of a patent system that allows to patent software and
algorithms.  These two companies already have a large patent folio, and are
using it for a lot of interesting tricks like cross licensing, threatening
competition (to drag them into a costly law suit), or otherwise reinforcing
their monopoly (look at Microsofts use of patents to prevent free software
implementations of some of their standards, for example).

The big companies (second only to the biggest) don't care much either way. 
They have the money to pay a horde of patent attorneys and play the game as
the biggest companies, but they could also use the money for something else. 
The CEO of Oracle for example has said that software patents are a nuisance,
but they can cope with them.  They don't have any real advantage or
disadvantage.  Everything else is against them.

Surely there are still some individuals who are pro software patents for
whatever reason, and there are also alliances which have spoken in favor of
them.  The BSA has done so on pressure by IBM and Microsoft, and against the
opinion of the other members.  The German Bitkom has done so because their
advisors are mainly patent attorneys, not software economics.

Thanks,
Marcus


-- 
`Rhubarb is no Egyptian god.' GNU      http://www.gnu.org    marcus at gnu.org
Marcus Brinkmann              The Hurd http://www.gnu.org/software/hurd/
Marcus.Brinkmann at ruhr-uni-bochum.de
http://www.marcus-brinkmann.de/



More information about the Discussion mailing list