Re: [fsl-discuss] Re: [Patents] The politics of software over-protection -- moving forward..

[Erik Josefsson <erikjosefsson@telia.com>]

On Sun, 27 Apr 2003 13:06:40 -0400 (EDT)
Gregory Aharonian <srctran@world.std.com> wrote:

> >Software can be written so that a single program can be used for
> >many different purposes.
> 
> Electronic circuits can be written (in Verilog)

And thus the verilog code is protected by copyright (for the
fiftyeleventh time, look at the LGPLed LEON processor).

> so that a single
> circuit can be used for many different purposes.

Which breaks the ideal "one product one protection" symmetry. This
happened in the chemical industries a long time ago and in Sweden we
ended up regulating that problem with "användningsbundet produktskydd" -
that is something like "protection of specific use only" - in chemtech.
The problem of broadness and/or unwanted broken protection symmetry is
not new, and well documented even from Nordic Patent Inquiry of 1963: Is
it not desireable to grant patents on chemicals with unknown multiple
future uses.

> That is to say,
> a single electronic circuit can be used to cover more applciations
> and industries than say a patent for lawn mower blades.

Which I guess was the reason for the su generis image protection
directive?

> By improving the quality of patent examination for both software
> and electronic circuits, the result will be the type of narrow
> application patents that the patent system should be issuing,
> covering just a new specific use.

Narrowing a software patent to solve the problem you describe ends up in
a machine translateble text which I guess by many computer scientsts
would be regarded as source code.

//Erik