Helen Borrie wrote:

> Oh yes it does! Software is by very nature a commodity that knows no
> international boundaries. However, thanks to US patents legislation, which
> flies in the face of international copyright, one is required to file US
> patents, regardless of where one lives, in order to be protected from
> piracy from within the US. Otherwise, an American can get hold of your
> architecture (e.g. through reading a White Paper, or having your source
> code), alter enough of the code to escape American copyright constraints (a
> numbers game) and file a patent for your scheme. Translating the core code
> into a different source language would be more than sufficient to beat US
> copyright constraints.
>
> The cost of filing US patents from outside the US is an enormous, up-front
> cost and takes years; so, to us, it's a "might is right"
> situation. Equity and justice simply don't enter into it. IMO, it's the
> greatest possible disincentive to Open Source.

I think we've definitely drifted off topic, but here goes anyway.

I agree that software patents are a great wrong, and I certainly believe
that they shouldn't exist. However, piracy and patents are unrelated
issues; If someone pinches your source code, it is copyright law which
can catch up with them. If someone reimplements your algorithm, there
isn't much you can do about it, unless you are protected by a patent,
and the jurisdictions are appropriate.

With a patent, you have the concept of "prior art", where you can't
patent something which is obvious to someone skilled in the field, or
which has been previously published. If someone gets the algorithm they
are going to patent from a published paper or source code, there is
clearly prior art, and the patent should not stand up to a court
challenge.

Of course, the USPTO would probably let you patent standing on one leg,
so that a court challenge would probably be necessary to establish that
there was prior art, but that is another issue.

Jan Mikkelsen