Re: [Patents] newsscan: Software patent limits 'go too far'

[James Heald <j.heald@ucl.ac.uk>]

Russell McOrmond wrote:

>  
> On Tue, 21 Oct 2003, Gregory Aharonian wrote:
>
>
> > To that extent, I view C computer programs no different than I view
> > digital hardware circuit diagrams - two different but logically
> > equivalent ways of representing a specific machine.  And therefore both
> > equally patentable.
>
>
>   You must eventually admit that that this is a personal perspective, and
> another valid perspective is that both should equally not be patentable.  
>
>   You can agree with your suggesting that they are equivalent without
> agreeing that either should be patentable.  The manufacturing process used
> to produce the FPGA should be patentable, but I see no reason or logic to
> considering the programming itself to be patentable.

I would like to second this.

The proposed EU parliament test for "technical" would allow an new 
algorithm for FGPA layout to be patentable if it had a physical effect 
-- perhaps a new way to reduce capacitative coupling for example.

But the program itself laid out on the FGPA should not be patentable. 
The specific implementation should be defended by copyright.  But there 
is no economic case for allowing the sequence of the abstract data 
processing to be monopolised by patent.

All best,

   James.