how patents work (was Re: Microsoft prohibits GPLed work via licensing of CIFS standards)

Alex Hudson home at alexhudson.com
Wed Apr 10 14:18:12 UTC 2002


On Wed, 2002-04-10 at 15:02, Joerg Schilling wrote:
> From my understanding of patentable claims (may be this is different in USA), 
> you cannot get a patent for anything that has been around before.

No, sadly, not true, certainly not here in the UK anyway. (UK rules are
basically the same as the European Convention too). The only requirement
is that is hasn't been _published_ before - hence my comparison between
your example of the Sun network system, and some other Free Software
system. It would be a lot easier to argue that the Free Software was in
fact a published work than any implementation Sun had done.

In fact, even if a work _has_ been published before, that doesn't stop
you from patenting something. The fact that prior art exists doesn't
matter a huge deal - the patent office will not search hard. It means
that your patent is terribly weak if there is obvious prior art, and it
would be unlikely that you could enforce it. However, as always, the
balance is weighted in favour of those who can afford to go to court :(

> >No, you're not comparing like with like. Linux is Free Software. If you
> >accept that source code is speech (which is what makes this arguable),
> >having the code around is essentially equivilent to publishing. It's a
> 
> But patents and Copyright are not related. So why do you make this
> comparison?

Where am I comparing copyright with patents? I'm not saying that having
copyrighted code available invalidates any patented system... is that
what you're pointing at?

> My understanding of prior art is that the idea has been around before and
> used in products. The important thing in publishing ideas is to tell other
> people that the idea did exist.

As I said before to Alessandro, the patent isn't strictly on the idea,
it's on the solution. To invalidate a patent with prior art, you have to
show that the solution has been used before. Essentially, prior art
consists only of previous patents issued, and published works
(scientific/technical journals, etc.).

The fact that someone has solved the problem previously isn't itself
enough to invalidate a patent.

> >Don't turn this into a Sun vs. Linux argument again please, we've had
> >enough of that...
> 
> Well this was not my intention, but you started with Sun bashing.......

If that was how it sounded, I apologise. But I would stand by my
assertion that it is unlikely that proprietary software (Sun or
otherwise ;) could be used as prior art to weaken a patent.

Cheers,

Alex.

-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 232 bytes
Desc: This is a digitally signed message part
URL: <http://lists.fsfe.org/pipermail/discussion/attachments/20020410/bbb93084/attachment.sig>


More information about the Discussion mailing list