Re: Why are software patents wrong?

[Paul Hovnanian P.E.]

Alexander Terekhov wrote:
> 
> "Paul Hovnanian P.E." wrote:
> [...]
> > Unlike hardware, there are numerous ways to implement the same function
> > (task) in software. Given a task description ...
> 
> You can't patent a task. You can patent a novel, useful, and non-
> obvious method to achieve it. In the EU, invention must have a
> technical effect (positively affect the ability of a "device" as
> a whole to carry out a technical function) to be patentable under
> the current EPO {case}law (and the upcoming EU patent directive).

OK. If I can't patent a task, but the application of 'software' to
accomplish that task is patentable, then how 'novel', 'useful' and
'non-obvious' does a second software application have to be to escape
coverage of the initial patent? Would an implementation on a different
hardware or OS platform be sufficient? What about writing the
application in a different language? 

This last point is particularly important because we are at a point with
software development tools where the description of the task itself is
sufficient 'software' to build an application. So now the description of
the task IS the software.
 
-- 
Paul Hovnanian     mailto:Paul@Hovnanian.com
------------------------------------------------------------------
I think you left the stove on.