[Patents] economic disruptions by swpat

[phm@a2e.de (PILCH Hartmut)]

> >This domination must be justified for every field, and the balance tilts
> >very badly in disfavor of the patent system in the case of software. I
> >am not saying that it is very good elsewhere.
>
> An interesting economic question Hartmut raises.  All areas of patenting
> have problems (I agree) that need to be justified (I agree it is useful
> to reexamine the merits of patenting).  The question is, are the
> economic disruptions of software patents, when balanced, very badly
> in disfavor of software patents.
>
> I.E., as a percentage of industry incomes, are software patents leading
> to altered allocations much greater than in other industries?  And/or
> are software patents leading to altered corporate activities much
> greater than in other industries?  And/or are software patents leading
> to altered corporate R&D investments much greater than in other
> industries?

There are a few factors that make software patents more disruptive than
chemistry patents, e.g.

(a) difficulty of acquiring new empirical vs new logical knowledge:
    one requires costly experimentation plus thinking, the other only
    thinking
(b) sequentiality of logical knowledge: everything breaks neatly into
    steps, whereas in empirical knowledge you have constraints imposed
    by the perversity of matter (or, even worse, the perversity of
    society, when you talk about social sciences)
(c) relative transaction costs:  chemical products
    have a fixed cost due to replication costs while information
    rapidly approaches a cost of zero and is socioeconomically
    most effective when it has reached zero, leading to a division
    by zero in the relative transaction costs.
(d) to chemistry patents are what copyright is to software:  there
    is no anti-copying protection in chemistry apart from patents.

> Show me statistically reliable data confirming these three domination
> economic domiantion effects, and you will have a case for treating
> software patenting differently.  And to date, there is no such data
> (and Bessen's models didn't consider cross-industry comparisons).

The fact that we have such a lot of people complaining about bad
patents in the software area while they seem to be rather quiet in
the chemistry area also is evidence.  From my experience I know that
some of the chemical patents are rather obvious as well.  But I also
know that this obviousness is especially bad in those cases where
the innovation was based on mere logical combinations, which again
makes these bad patents come close to software patents.

> So if you are serious about this issue, why not start lobbying the
> EC to fund a study to measure the effects of patents in a variety
> of industries to see if indeed software patents are being orders
> more disruptive?

We are in fact lobbying to have more studies conducted.

Of course we are in favor of more intensive studies on this matter.
However such studies are not heeded by those who are at the helm in
Brussels and London at the moment.  They cut&paste EPO doctrines as if no
study had ever been conducted, maybe because studying EPO caselaw is what
they were trained to do.

-phm