Date: Sun, 11 Feb 2001 01:09:26 +0100 From: Roelof Osinga <roelof@nisser.com> To: "Jeroen C. van Gelderen" <jeroen@vangelderen.org> Cc: Kris Kennaway <kris@obsecurity.org>, freebsd-security@FreeBSD.ORG Subject: Re: Reminder notice about FreeBSD Security Advisories Message-ID: <3A85D836.F9D001BC@nisser.com> References: <20010209214354.2FBD637B4EC@hub.freebsd.org> <Pine.BSF.4.33.0102092253530.63359-100000@deneb.dbai.tuwien.ac.at> <20010209140614.A67010@mollari.cthul.hu> <3A853F1C.DED59C4B@nisser.com> <20010210053726.A45756@mollari.cthul.hu> <3A854C41.7BA1B2A1@nisser.com> <3A85799E.C6A3745B@vangelderen.org> <3A85AF50.1DF04F6F@nisser.com> <3A85B391.DB3ABC2A@vangelderen.org>
next in thread | previous in thread | raw e-mail | index | archive | help
"Jeroen C. van Gelderen" wrote:
>
> But IDEA is an algorithm, not software and therefore perfectly
> patentable. So please, don't argue to the contrary until you
> have consulted with a patent lawyer and can present us with
> some *hard* facts.
> ...
> The independent source here seems to be your incorrect reading
> of an article in Wired neither of which, as far as I know, have
> any sound expertise in field of patent law.
>
> [snip unfounded speculation]
Allrighty. Lucky for you I have got David Pressman's [a patent
lawyer b.t.w.] Patent It Yourself, 8th edition standing by. Let
me quote a snippet from it (from page 9/14):
"If your invention involves (or actually is) a computer program
or /algorithm/ [!, my stressing] - that is, a set of instructions
for a computer - you must claim to indicate some practical, useful,
concrete, and tangible result, and not just as a set of steps
for manipulating data or numbers"
which, of course, is supported by Merriam-Webster as in:
" a procedure for solving a mathematical problem (as
of finding the greatest common divisor) in a finite
number of steps that frequently involves repetition of
an operation; broadly : a step-by-step procedure for
solving a problem or accomplishing some end
especially by a computer
"
but then, you already knew that. Didn't you?
Granted, there is recent proof that I've got blind spots from here
to Bombay, still... My memory has an even greater notoriety of being,
well, spotty I guess. Even so I would conjecture that even in the
US, lenient though it might be, patenting a pure algorithm without
gimmickry as indicated by Pressman would be a neat trick indeed.
Still, since I am not a lawyer but merely a simple programmer turned
businessman 20 years or so ago, I will gladly bow to the hard facts
you did provide. After all, did not you imply to be either a (patent)
lawyer or to have recently consulted one?
Lest I leave you confused if not befuddled as to what it was I was
just trying to say, my point here is that clearly no real distinction
is being made betwixt algorithm and software. Hence if Wired *reports*
the EU decided not to allow software patents, commercial pressure to
the contrary notwithstanding, then it follows the same holds for
algorithmic patents (or rather, patents covering algorithms).
A very simple induction, but an induction nonetheless. Inductions
not carrying the weight of deductions I could still easily be proven
wrong. And often am. C'est la vie.
Mind you, is it really an induction? We've got:
WRD -> ~PATENT(SOFTWARE)
M-W -> EQUALS(SOFTWARE, ALGORITHMS)
from which it follows that
WRD -> ~PATENT(ALGORITHMS)
not even any need to drag a patent lawyer into it, now is there?
Still, it hinges on Wired reporting the truth. Then again, the same
'facts' got reported by a multitude of others.
The reason I do not gladly give up is because I much rather see
the EU as a bulwark 'gainst the commercialization of pure
mathematics than as bowing their collective head in shame and
confessing they were wrong all the time: mathematics is just
another invention.
So I'm defending an ideal. I'll gladly confess that as a mere human
I'm far from perfect. Hence I might be the next Don Quichotte. Or
just a plain idiot.
Still... I showed you mine (patent lawyer, that is) now show
me yours!
Roelof
PS I *do* believe it is important. For if algorithms are indeed
patentable right now... what's next? I would say prior art. Who
cares about prior art anyway. If nobody important knows about it,
then is it really prior art? So clearly prior art must be
established by important people or sources knowing about it!
Publishing in some 'never heard of' magazine/source just doesn't
cut it!
--
Home is where the (@) http://eboa.com/ is.
Nisser home -- http://www.Nisser.com/
To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-security" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?3A85D836.F9D001BC>