[Patents] Re: Hypothetical Example

[PILCH Hartmut phm@a2e.de]

> Ok, these are my final words on this subject,. 

I hope you're not leaving our campaign though.
 
> Patent Protection gives the owner of the patent:
>     1)    20 years of exclusive rights, except in exceptional circumstances
>            to use the invention in whatever commercial manner they see fit.
>     2)   The right to demand a royalty for usage of the invention and more
>          importantly the right to sue for willful infringement for lots
           of money.

I think you can cross out the "wilful", at least for most countries.
In some cases, you can only be sued if you don't withdraw your work after
having been notified by the patentee.  If however "software as such"
became patentable, publishers could be sued without notification.

> The problem comes I think from people assuming :
> 
>     1) There is always an "alternative algorithm". Ok, so I wasnt too clear 
> on this.
>         My apologies !. I still believe there may not be alternatives 
> available and that
>         the alternative algorithms may be inappropriate for your 
> requirements.
>         The point was that if we have software patents, as the U.S and 
> Japan already do, 
>         then you may not be able to write software without infringing on 
> SOMEONE'S patent..  

You may be mixing two problems:
(1) non-availability of alternative algorithms
(2) one program is a complex system that may infringe on thousands   
    of patents.  Whatever you write, you may touch a mine.
 
>      2) Open Source is a good idea. Yes it is, but there seems to be some 
> misconception
>      about how this applies to the real world when software patents are 
> available. If you
>      release the source code to the world, a competitor may patent it 
> before you do, or a
>      part of it - then sue you for infringement. 

That would be a case of fraud.  Only someone who invented something (not
necessarily first though) can get a patent on it.  But you may not be able
to prove it.

>      First to invent does not get the rights,

But does, if he can prove he was there before, get a free license.

>    For the example, I put my self in the position of a company who wanted 
> to defend
>    itself against legal attack - easiest solution ? Hide the source code.

right.
 
>     3) The belief that all you have to do is licence the technology and 
> re-write it. In the
>     real world, there are deadlines to meet, expenses to pay and mouths to 
> feed - (cats,
...

not only that: patent licenses are usually granted on an exclusive basis
to the party that can generate the greatest revenue from the patent and
will therefore pay the highest license sum.  So, if you want to solve your
simple database problem, be prepared to outbid Microsoft.
 
> Currently the only real legal defense against this is (possibly)
> Article 52(2) of the Muchen (Munich?) Convention. It has been

Muenchen = M\"unchen = München = Munich

> suggested (in the green paper no less) that this article could be
> "deleted" in order to allow the patenting of software. I believe this
> is more or less the approach the U.S. took in its efforts to harmonise
> and simplify there patent system, and to comply with the TRIP's
> article (I forget the number..) which says that patents must be
> allowed for all inventions - without discrimination. I think they

The Trips article leaves it open how to define "invention".
The Munich convention stipulates that it should be something with an
industrial application, not just ideas or software as such.   The US
patent law says that it should be "anything useful".

I don't see how harmonisation comes in here.  As far as I can see it is
nothing but a coverup maneuvre.  The patent organisations are trying to
enlarge the scope of their system without letting the public notice.  
They are already no longer talking about "industrial application", only
about "technical solution to a technical problem", and now they claim that
the Munich convention must be adapted to their illegal practise.  This
adaptation, they claim, will change nothing.  As the spokesman of the
German Federal Ministery of Justice (BMJu) put it (in my translation):  

 Patents for computer programs are already granted today under certain
 circumstances, namely when the program controls a technical process.
 This legal situation will in principle remain unchanged.  What is intended 
 however is a clarification of what is patentable and what is subject to
 authorship rights protection "only".

Nobody has so far come up with an explanation why clarification is needed.
Those that clamor for clarification are exactly those who create
obfuscation by using such ambiguous and law-deviant terms as "technical
process".

> For the benefit of generations of future software development and
> innovation. I submit that we should *NOT* delete Article 52(2) simply
> to allow harmonisation of the Patent System with the US and Japan. I
> believe this is harmful both socially and economically to everyone
> except those with extensive patent portfolios. We should not implement
> a bad idea everywhere just to have "harmonisation" for the few that
> can afford it.
 
Nobody is talking about harmonisation with Japan and the US, only about
harmonisation among the European patent systems.  But most of the
people from the patent organisations talk in an encrypted wooden language
(langue de bois), like members of the soviet politbureau.  A very special
meaning, known only to insiders, is hidden behind their apparently
pathetic and empty words.

-phm