Date: Tue, 9 Oct 2001 23:13:43 -0700 From: "Crist J. Clark" <cristjc@earthlink.net> To: Salvo Bartolotta <bartequi@neomedia.it> Cc: Ted Mittelstaedt <tedm@toybox.placo.com>, "P. U. (Uli) Kruppa" <root@pukruppa.de>, freebsd-chat@FreeBSD.ORG Subject: Re: Use of the UNIX Trademark Message-ID: <20011009231343.C387@blossom.cjclark.org> In-Reply-To: <1002663600.3bc36eb096ee5@webmail.neomedia.it>; from bartequi@neomedia.it on Tue, Oct 09, 2001 at 11:40:00PM %2B0200 References: <000601c15084$87edd360$1401a8c0@tedm.placo.com> <1002663600.3bc36eb096ee5@webmail.neomedia.it>
next in thread | previous in thread | raw e-mail | index | archive | help
[OK, way off of -questions. Redirected to -chat.]
On Tue, Oct 09, 2001 at 11:40:00PM +0200, Salvo Bartolotta wrote:
> Ted Mittelstaedt <tedm@toybox.placo.com> wrote:
>
> > There's currently a huge argument over software patents, ie: patentable
> > algorithims.
>
>
>
>
> <snip>
>
>
>
>
> I seem to undertstand that the law applying to software is different in
> Europe, ie copyright-oriented (Europe) rather than patent-oriented (USA).
>
> The patent question perplexes me, probably because I have a very limited
> understanding/knowledge of its issues and niceties.
The copyright question has always perplexed me. A computer program is
a set of instructions to make a machine do something. Copyrights have
historically meant to protect artistic works, literature, and the
like.
I can write a program to do something really l33t and copyright
it. Only I can authorize distributions of copies of that
software. Someone else can write a program to do the same thing and
they can distribute it however they want. It's not a copy of my code,
which is protected, but it does the same thing. Functionality is not
protected by copyrights. I can spend a big wad of cash to come up with
this nifty new idea and develop it, but someone else can come along
and write a program to do the same thing and there is nothing I can
do. What incentive do I have to spend the cash when Microsoft will
make a clone of my neat app and make it a part of their operating
system next month? None. Inovation stops.
To use another anology. I build a big mechanical box with lots of
wheels, wires, motors, and vacuum tubes. I takes punch cards in one
end and spits out some interesting output at the other end. I patent
the process. That's what patents are for, no one will argue I cannot
or should not. Now instead of building a big custom box to do this
process, I write a set of instructions that make some other generic
piece of machinery (a com-pu-ter) do this same kind of thing. Why can
I not patent the process?
> <exercise type="wild imagination">
>
> For one moment, suppose that the principle of algorithm patentability came
> true to the fullest extent. [the choice of "come true" is NOT coincidental
> ;-)]
>
> The next day, I would wake up and patent the algorithm solving 2nd degree
> algebraic equations.
You could try, but you couldn't patent the methods to do this that you
find in every basic algebra book. Prior art. You only can patent
something new that you discover.
> I chose a trivial example just for the sake of simplicity. You could
> substitute algorithms/theorems on [differential or algebraic] equations;
> numerical analysis/calculus algorithms (eg Runge-Kutta methods); etc. etc.
> etc. By the way, the discussion is not purely theoretical: think eg of CRC
> polynomials...
All of these algorithms already exist. You have no claim to patent
them. OTOH, if you devised a _new_ method to do some interesting math,
yes, you can patent.
(Not to say that some people haven't pulled some things off with the
Patent Office. One of the jokes about a former employer of mine was
that the company had a patent on least-squares regression. It really
did.)
> Next, I would write a program in BEEEE_sick solving 2nd degree algebraic
> equations. A month later, you would chance to write another such program,
> without any prior knowledge of my patent(s) or even my program(s).
>
> Finally, I would sue you for two patent infringements: the algorithm and the
> program. Rich lawsuits. :-))
Probably not. If I had acted in good faith (you admit I didn't know
anything), you probably wouldn't receive much of anything in a civil
court other than an assurance I would stop. I don't have any money
anyway.
> Alternatively, you would have to pay [$$$]$$$ each and every time you made use
> of the aforementioned algorithm. Hmm, that would sound like quick and steady
> progress for the whole field of studies and/or applications. :-)
Yep. People, corporations, and educational institutions spend
mindboggling amounts of effort and money trying to discover new
medicines, devices, technologies, numerical methods, etc. which they
can patent.
> I may be missing something, er, quite a lot of things, but such scenarios make
> little to no sense (to me) -- however
> subtle/clever/precise/interesting/rigorous/etc may be, computationally
> speaking, the chosen definition of algorithm complexity and/or ahem
> "originality".
>
> Incidentally -- it's just my impression, mind you -- I would say this kind of
> law, in the long run, might be very harmful to software industry itself.
It's not patents or copyrights that are a hazard, but the shrink-wrap
license agreements. If useful algorthims were properly _patented,_ then
software distribitors might not try to tag on anti-reverse-engineering
clauses and other bogus legalese in their license agreements. They can
even hand out source code without fear. Even if other people can see
the algorithms used in the software in front of their noses, they
cannot steal the distributors investment since making work-alike
(as opposed to not bit-image, copyright-violating) copies would
still violate the _patent._
I'm not saying that all software should be patented. But a lot of the
shrink-wrap license hell that we are in is because applying copyrights
to programs isn't a correct fit. Programs are instructions to make a
machine do something. They in effect make a generic piece of hardware
behave like some specialized piece of pseudo-hardware. You patent
devices and processes. If you want to protect what a program _does_
you should patent the process, the algorithm.
--
Crist J. Clark cjclark@alum.mit.edu
cjclark@jhu.edu
cjc@freebsd.org
To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-chat" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20011009231343.C387>