(Changing subject as we are no longer talking about licensing issues)

On Fri, 16 Jan 2004, Alexander Terekhov wrote:

> Russell McOrmond wrote:
> [...]
> > deal with some of the worst cases we are currently dealing with.
> Care to provide some SPECIFIC example(s) involving IBM?

  I suspect this may be off-topic for this forum as it does not relate to
licensing.  I just want to make clear the problem which I think the OSI
should be in discussions with IBM about.  I am not aware of a different
forum that deals with OSI issues that are not licensing issues.

  The problem is not licensing from IBM of already existing software
patents that IBM happens to hold the patent on, but IBM lobbying for an
increase in the number of software patents that will exist and the number
of countries that will honor these highly controversial monopolies.

  The Open Source movement needs to try to render as many software patents
harmless as possible through Open Source patent licensing.  This is only
part of the solution, and the other part is to deal with the validity of
this form of monopoly in the first place.  It is in this latter context
that I launched my complaint about IBM.

  There are people who believe that software patents are a valid form of
government granted monopoly, but I do not.  Even those people who agree
with the concept of software patents will suggest that approximately 60%
of those granted by the USPTO would be found invalid if adequate novelty
(prior art) testing was done.  Others suggest this is as high as 95% when
appropriately rigorous testing for useful (Industrial application) and
novelty are done.  I happen to believe that 100% are invalid and should
simply state this in statutory limits.

Note: This is software itself, not software that just happens to be part 
of an otherwise patentable manufacturing process.  Software should neither 
add to, not subtract from, the patentability of a manufacturing process.  
If all you have is a general purpose computer and some specific software 
then this should not be considered patentable.  If you have a 
manufacturing process that uses software to control the process then that 
process should still be patentable even though it involves software.

  I believe that patents should only be on industrial/manufacturing
processes which involve manipulations of nature to create tangible goods.  
When dealing with "information process" patents we are no longer dealing
with manufacturing processes to create tangible goods, but situations
where the processes are themselves the intangible good.  The monopoly is
not on the process used to create the good, but on the good itself.

  IBM has been lobbying for unlimited patentability, pushing the rhetoric
of "technology neutrality" that is the most common political phrase used
against Open Source software.  The problem is, software is not a
'technology' any more than laws, acts of parliament or Roberts Rules are a
'technology'.  When used the way CompTIA uses the phrase it is also
nonsense:  the choice of Open Source or not is not a technology choice,
but more like the decision to mandate certain public safety laws be
followed when building a building.

  Regardless of how IBM licenses individual patents the fact remains that
IBM is lobbying worldwide to increase the number of invalid "information
process" patents.  If IBM takes 10 steps backwards and 1 step forward, are
we really expected to congratulate IBM on the forward motion?

  The promotion of the exclusive monopoly called the "software patent" is
in and of itself an opposition of Open Source.  You should not focus on
individual cases where IBM, Microsoft, or some other company puts an Open
Source license on specific software, but ask why IBM, Microsoft or others
aggressively lobby against Open Source.

 Russell McOrmond, Internet Consultant: <http://www.flora.ca/> 
 Governance software that controls ICT, automates government policy, or
 electronically counts votes, shouldn't be bought any more than 
 politicians should be bought.  -- http://www.flora.ca/russell/

license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3

Reply via email to