[Patents] Re: [Patents] wide realm of future patentability

[Bernard Lang Bernard.Lang@inria.fr]

On Thu, Oct 28, 1999 at 08:18:00PM +0200, PILCH Hartmut wrote:
> A current tendency of the patent system is to formulate the principles
> that
> 
> - anything "technical", e.g. anything that makes use of
>   natural laws to perform a certain function, is patentable
> - a Turing machine is a device that makes use of natural laws
> 
> ==> any process of executing a computer program is patentable, but
>     not the program itself

So we must obtain that this is clearly stated in the law ...
i.e. there is patent infringment only when the program is executing on
a general purpose computer, or when it is included in a specific
purpose device that is not functional without the program (so as to
allow for exemple companies to protect a specific hardware device
(e.g. a phone) independantly of whether it is hardware or software.

This is only a fall back issue for me, but he can be justified on
economic grounds, since bringing a specific purpose device on the
market entails production investments that are directly related to the
program concerned. Recouping industrialization investment is one
economic justification of patent.

  But no such investment exists, in significant proportion, for the
program alone to be run on a general purpose computer (I do not
consider advertising and legal fees as productive investment).

> ==> these conclusions are independent of the EPC.  Change of the EPC
>     is irrelevant to the legal situation.  The only reason to change
>     the EPC is that this will remove some confusion.

What is EPC ? ... anyway, I suppodse the reason is TRIPS

> This is what the specialist at the German Patent Office (DPA) told me.  I
> will receive a collection of his articles on the subject and know more
> details later.
> 
> There are of course quite a lot of things that can be executed via a
> Turing machine.  Including
> 
>       - postscript files
>       - musical compositions
>       - business methods
>       - the hypertext structure of articles typeset in TeX
>       - the semantic structure of texts written in Logical Language[1] 

So according to your specialist, a piece of music can be patented, if
it is known to make your cows sleep, when played.

This is essential, and natural, since there is no real difference
between program and data. But where does it lead us.

What is the difference between being asleep and being happy. Some
musics make people happy ... can that be patented ?

As for social process ... Jean-Paul uses the example of reverse auction...
which is a business method.

What about a safe voting system that would be used to run a community
(no business involved)... elections for an association, or a small town.

> The fact that one *can* make a business out of something is a sufficient
> reason to suppress any non-business use of that technique.  Anything
> non-business is potentially business, and business is the point of view,
> under which everything is seen by the patent system. As the patent
> specialist at the DPA/GPO told me, "Free Software has become a force in
> the market and can therefore not be placed under special protection".

That is new arguments they invented in the past 20 years... without
justification.

Originally, the purpose of patent is (a bit to reward people and) to
allow recoup the investment or potential loss when:
  a- inventing
  b- making public
  c- bringing to market

  The last justification c is more modern that the other two.

  a is the reason that requires SIGNIFICANT inventive step
  b is the reason that requires novelty
  c is the reason that that pretends to requires "industrial usefulness"
     and this also, I believe, is a modern phrasing
    (anyone has a history on this)

  The point is that (c) should justify not "anything that has
industrial use" but "anything that entails significant difficulty to
be brought to market".

  The logic of law is not to create gold mines for happy fews, but to
tune the system for maximum global benefit.
 
> The DPA itself has developped some large-scale non-commercial software
> which the authors wanted to make free.  But the Patent Office didn't allow
> this, because it "had been put under pressure by the industry not to
> engage in anything that could represent a competition to private
> business".

.. and did they make this pressure public ...
this is pure ideology... not economics

> This is also Mr Breese's (bresse.fr) line of argumentation: free software
> as an illicit competition by state-paid people to the proprietary software
> industry.

So what ... that has nothing to do with the economic justification of
patents. ... Breese runs on money and ideology

> -------------
> [1] Logical Language (http://www.lojban.org) makes the potential realm of
> patents especially wide.
> 
> There is a parser, developped on Linux of course, which turns Lojban texts
> into Prolog structures.  There is even a way of writing Prolog programs in
> Lojban.  Almost anything that can be expressed in English (except for
> wordplays that depend on English word forms) or other natural languages
> can be expressed in Lojban.  Complicated relationships can be expressed
> much more conveniently (e.g. 1 sentence rather than a paraphrase of 20).


> [2] I would bet that a good lawyer will prove this.

yes ... it is logical ... where does it stop, though ?

-- 
Bernard.Lang@inria.fr             ,_  /\o    \o/    Tel  +33 1 3963 5644
http://pauillac.inria.fr/~lang/  ^^^^^^^^^^^^^^^^^  Fax  +33 1 3963 5469
            INRIA / B.P. 105 / 78153 Le Chesnay CEDEX / France
         Je n'exprime que mon opinion - I express only my opinion
                 CAGED BEHIND WINDOWS or FREE WITH LINUX