*BSD News Article 7271


Return to BSD News archive

Xref: sserve comp.org.eff.talk:9696 misc.int-property:736 comp.unix.bsd:7320
Path: sserve!manuel.anu.edu.au!munnari.oz.au!spool.mu.edu!uwm.edu!linac!att!ucbvax!virtualnews.nyu.edu!brnstnd
From: brnstnd@nyu.edu (D. J. Bernstein)
Newsgroups: comp.org.eff.talk,misc.int-property,alt.suit.att-bsdi,comp.unix.bsd
Subject: Re: Patents:  What they are.  What they aren't.  Other factors.
Message-ID: <6581.Nov112.31.0292@virtualnews.nyu.edu>
Date: 1 Nov 92 12:31:02 GMT
References: <1992Oct28.112222.5170@netcom.com> <1992Oct28.153748.3758@murdoch.acc.Virginia.EDU> <1992Oct28.220811.27666@netcom.com>
Organization: IR
Lines: 46

So now we have two supposed examples of beneficial software patents: an
unspecified Dolby patent (which I haven't seen, hence can't talk about),
and the well-known RSA patent.

A few people pointed out that the RSA patent *did not* benefit society.
The inventors published RSA many months before applying for a patent, so
the patent system obviously served no use as a publication mechanism in
this case. The farfetched idea that ``the possibility of a patent serves
as incentive to perform research'' cannot have been true for RSA: at
least one of the inventors was under a government contract requiring him
to publish all research results *without restriction*. So the patent
didn't benefit anyone except its inventors. (All this was obvious at the
time.) Furthermore, history shows that the inventors' company, RSADSI,
has squashed at least three independent software packages, obviously to
the detriment of everyone except RSADSI.

In article <1992Oct28.220811.27666@netcom.com> mcgregor@netcom.com
(Scott Mcgregor) attempts to defend the assertion that the RSA patent
benefitted society. Does he dispute anything above? No. The crux of his
argument is beating around the bush: ``How much marketing and
promotional efforts were really necessary to make Apple and Microsoft
want to use RSA rather than DES or rolling their own? Would they have
wanted to do this if RSA hadn't been working to get the ITAR
regualations changed to allow them to export?''

If the public-key cryptography market weren't profitable then RSADSI
would not exist. So it is profitable. If RSA weren't patented then there
would be competition for RSADSI---meaning *more*, not less, marketing
and government lobbying. The patent changes the nature of the market but
not the existence of the market.

The same is true of any research idea. Either it's profitable to market
the idea, or it isn't. If you see a company marketing the idea, then
it's obviously profitable. The argument ``society doesn't benefit from
the idea unless it is marketed'' is true but completely orthogonal to
the question of whether society benefits from the *patent*.

In theory, society gains from a (marketable) patent (which is novel,
unobvious, etc.) by virtue of its publication; the inventor gains from a
(marketable) patent because it is a temporary monopoly. When the idea
isn't marketable, nobody gains and nobody loses. When the idea is
marketable, somebody's going to market it whether or not there's a
patent, and arguments about the benefits to society of marketing are
entirely irrelevant.

---Dan