Subject: Re: Examples needed against Soft Patents
From: "Stephen J. Turnbull" <stephen@xemacs.org>
Date: Wed, 22 Dec 2004 14:41:10 +0900

>>>>> "Chris" == Chris Maeda <chrismaeda@comcast.net> writes:

    Chris> The best argument against software patents I have seen goes
    Chris> like this --

Unfortunately, it's only "best" if you already agree with the thesis.
It will never convince an opponent, and it's too breezy to convince a
thoughtful but unconvinced person who understands the economics fairly
well.

    Chris> - The patent system was designed to promote innovation by
    Chris> encouraging inventors to disclose their inventions.  Patent
    Chris> advocates need to prove that software patents will lead to
    Chris> increased innovation.

No, they don't.  For two reasons.  First, since there is no right of
users to force the inventor to disclose, this is purely a cost-benefit
issue.  (Unlike the mass theft that we call the "Copyright Extension
Act".)  So they need to convince the legislature that the expected net
benefits are positive.  "Lead to increased innovation" is, admittedly,
a fairly good approximation to that.

But only if you define "innovation" appropriately.  Here, "innovation"
means "invent, develop, market, and deliver" a new product.  Anti-IP
advocates usually focus on "invent" to the exclusion of everything
else, but it's the "everything else" that matters more to society.

    Chris> - In the experience of most practitioners, software patents
    Chris> actually inhibit innovation:

I've seen no trustworthy evidence on that.  Yes, I know that all the
wannabe FSBs etc bitch like crazy, and Larry Ellison also published a
screed against.  However, there's no good way to guess how many of the
allegedly inhibited innovations would have made it to market, or how
many of them were genuine improvements rather than close substitutes.

    Chris> incremental nature of innovation in software leads to large
    Chris> numbers of small patents which create barriers to entry for
    Chris> new firms; it's nearly impossible to do anything
    Chris> interesting in software without infringing on somebody's
    Chris> patent.

That's the _good_ thing about patents!  Go do something innovative,
instead of just copying and/or reinventing wheels, for heaven's sake!
And don't forget to _sell_ that widget once you've got it, so that you
can pay the royalties! ;-)

Seriously, that is exactly what IP is all about: diverting effort from
redundant reinvention to innovation.  It's just as important to
encourage reuse as it is to reward the innovator for her investment,
if not more so.  As an industry we've been screaming for more reuse
for decades, we teach it in our classes, some companies have VPs in
charge of reuse: well, patents enforce that!

That's not a proof that patents (in any field) are good; but pointing
to reduction in R&D effort is arguably _support for_ patents.  You
do need a weatherman to know which way the wind blows.

    Chris> - Evidence that software patents have not led to higher R&D
    Chris> spending or higher innovation in the software industry:
    Chris> http://www.researchoninnovation.org/patents.pdf

Uh, I told Bernard _not_ to cite papers he doesn't understand.  That's
generally good advice.

You see, that's a theory paper, and therefore has nothing to say about
what actually happened.  Furthermore, it's not a paper on software
patents, it applies to all patents.  The senior author, Maskin, is on
the pink side, politically, and furthermore has no empirical skills
whatsoever; you'd need to cite a dozen such, with several authors who
have done objective, preferably emplirical, research in the field,
before I'd consider such papers evidence for anything, let alone
historical fact.  (Maskin is one of the economists whose opinion I
consider significant without reading the whole paper, despite my
opinion about his political bias[1]; good choice.  But by himself, not
enough.)

Maybe you meant swpat.pdf (sibling URL) from Maskin's co-author Bessen
and another economist, Hunt.  But that's not really applicable,
either; it falls into the "ignore everything else category," and says
so, right in the abstract: "sw patents substitute for software R&D",
which is exactly what is supposed to happen!  The freed funds are then
used to improve the non-software aspects of the product, market it.
According to pro-patent theory, the non-software-R&D benefits outweigh
the R&D decrease, and society wins.  Furthermore, you must show that
the lost R&D is innovative, rather than wheel-reinvention, further
tilting the balance toward patents.

    Chris> - So there is no evidence that software patents have
    Chris> promoted innovation.  Given lack of evidence in favor of
    Chris> patents, copyright law is perfectly adequate to promote
    Chris> innovation in the software industry.

Sorry.  Although I agree with your conclusion, if that's the best
argument we can muster, I'll have to consider returning to the Dark
Side.  :-)

Remember, the Dark Side has a lot more money and lawyers and lobbyists
than we do.  We need to reach the congregation, not to get the choir
to sing "Amen" in 7-part harmony.

As I wrote to Bernard, the killer economic argument is that (enforced)
reuse simply is not socially beneficial; most of the time, it's
cheaper to reinvent than to look up and use a standard part, even
under the best of circumstances.  ISTR figures that even companies
that are fascist about reuse achieve maybe 30% reuse in new products,
and there's always the issue of just how "new" those products are.
That means that the transactions costs (search and legal fees) are
pure waste.  Patents suck!

The killer social arguments are harder to pigeonhole, but to my mind
they are the enhancement of one kind of freedom is good for abstract
freedom, empowerment, "small (free software) (business) is beautiful",
and the social responsibility exhibited by most FSBpeople (whether
members of this list or not :).  FSB rules!


Footnotes: 
[1]  Note that the issue with political bias is not refusal to mention
unpleasant facts or results.  It's in choice of assumptions: if you
hang out with people who state certain "facts" about patents, you are
likely to build those facts, as false as they are, into your model as
assumptions.  Economics?  It's just a house of cards, man.  :-)

-- 
Institute of Policy and Planning Sciences     http://turnbull.sk.tsukuba.ac.jp
University of Tsukuba                    Tennodai 1-1-1 Tsukuba 305-8573 JAPAN
               Ask not how you can "do" free software business;
              ask what your business can "do for" free software.