[Patents] Black and white

[phm@a2e.de (PILCH Hartmut)]

> However, you do not mention the most important problem of software
> patents: the vast majority of them are obvious patents designed to
> block competition. (Actually, some obvious patents are used for
> defensive purposes; see bellow).  Patent offices ignore the
> requirement of "non-obvious".

It would be more accurate to say that this requirement is defined
formalistically and has nothing to do with what the layman might think it
should mean.  It is not designed to eliminate trivial patents, see

        http://swpat.ffii.org/stidi/frili/

> I am not against patents used for the protection of research. However,
> very few software patents have a significant research work behind.

Risking to appear as a "black and white" thinker, I may remind you that
you said triviality is only the "most important problem of software
patents".  There are so many others which may not be as spectacular but
also quite important, that piggybacking the "protection of research work"
onto the software patent bandwaggon would be completely unacceptable.

If it turns out that in some cases current copyright offers less
"protection" than seems appropriate, there are several ways to solve the
problem:

(1) fiddle around with copyright:
    - whenever someone publishes a source first, put the burden of proof
      on the plagiator.  Require that he must do something really original
      and creative, based on understanding underlying problems but
      completely "reinvented", not just transscribed to another form
      if expression.
    - be more strict during the first few years of copyright than during
      later years.  be more strict when the copier is aiming at a
      competing product than when he is striving for compatibility
(2) formulate the above and some more principles into a specialised
    software copyright (sui generis approach)

Patents are not a heritage of human history, shaped by many arbitrary
factors. I can see no reason in the world why the patent system should
carry any merit with itself that might qualify it for being applicable to
software.  On the contrary, it is very dangerous for the patent system
itself if you extend it to the sphere of logics.  The patent system is a
system of property on material objects, delimited by certain immaterial
characteristics.  When you break that rule, you break everything that
makes the patent system work, however inadequately, in its home sphere.

It would be "black & white" thinking to say "Copyright has flaws, here
comes the patent bandwagon, so let's jump on."  Copyright is designed for
immaterial objects, patents are not.  Moreover patents cause lots of
trouble and redtape even in their home sphere, while copyright is
unproblematic and unbureaucratic. If copyright still has deficiencies, any
openminded (not "black & white") person would rather say "let's see where
copyright is deficient and try to address the problem."

By trying to avoid being "black & white", people risk falling into a trap
of opportunism, that is really black & white.

-phm