Subject: Re: Examples needed against Soft Patents
From: "Stephen J. Turnbull" <>
Date: Tue, 28 Dec 2004 16:06:37 +0900

>>>>> "Ben" == Ben Tilly <> writes:

    Ben> Can you offer me concrete examples where groups have actually
    Ben> [organized to protect themselves]?

The FSF, the LPF, the OSI.  The (core) GNU Project.  NATO, the Warsaw
Pact.  The (original) Mafia.  Friends of the Earth, Sierra Club,
Greenpeace.  The Enclosure Movement in Britain.  The gangs in LA.  The
Kyoto Protocol and CTBT signatories.  The WTO.  The UN.  The League of
Nations.  The EU.  The United States of [various examples].  APEC,
ASEAN.  OPEC.  Many labor unions.

Some successful in accomplishing their goals, others not.  All are
formal organizations, composed of (at the time of organization, at
least) mutually sovereign members.

    >> The fact that free riders benefit is of interest only to those
    >> who habitually build walls around their gardens just to spite
    >> their neighbors.  But the fact that you can look at my garden
    >> won't stop _me_ from cultivating it.

    Ben> This statement misses the critical point.

No, the whole point is that there _is_ a critical point (hypersurface,
actually).  _Some_ groups, on one side of the surface, can and do
form.  Some are successful, others not.  The factors determining
success only partially overlap those determining formation.  Groups on
the other side of the surface don't form, and almost by definition fail.

    >> There are four factors that are important to justify the
    >> (practically important) conclusion you reach:

    >> (1) network externalities in consumption (the small group can't
    >> form)

    Ben> This is only true if your definition of "consumption"
    Ben> includes the possibility of a consumption that consumes
    Ben> nothing.

As always in economics, we are at some level talking in "net" terms,
so that zero consumption is possible.

    Ben> That is, the analysis applies even with something like
    Ben> information orsoftware where my use does not diminish what is
    Ben> available for others to use.

My point here is that the network externalities may be _positive_,
requiring a large group to get the total benefits over the hump of the
initial investment.

    >> (2) transaction costs tend to dominate benefits of organizing
    >> large groups unless network externalities are strongly positive

    Ben> Not necessarily a factor.  Even if you assume that there are
    Ben> no transaction costs, the problem remains.

_None_ of the factors are necessary, but all apply to some situations.

    Ben> The difficulty is that every individual may be better off
    Ben> trying to be a free rider than they are contributing to a
    Ben> solution.

Sure.  However, the applicability of the analysis to reality breaks
down when the incentive to free ride is low and the social benefits
are high.  Cf the extreme difficulty experimenters face in reducing
the incidence of the use of the cooperative strategy in the Prisoners'
dilemma game, or even in getting people to demand 99% of the pie in an
ultimatum game (player A chooses a fraction between 0 and 1, the
fraction is revealed to player B, B says "yes" or "no", if "yes" A
gets the fraction of the pie he proposed and B the remainder, if "no"
both get zero).

    Ben> Somewhat trickier is when a small group can supply the good,
    Ben> but none can on their own.  Then they can negotiate out a
    Ben> solution, but will constantly be battling the fact that each
    Ben> is under pressure to "cheat".  Negotiations will therefore
    Ben> get complex.  Think OPEC for a classic example.

Sorry: I don't know how to treat "complex negotiations" other than as
a transactions cost.  That was implicit; let me make it explicit here.

    Ben> When no small group exists, then you've got the problem in
    Ben> spades and no general solution exists.  Sometimes it has been
    Ben> solved though.

As Russ phrased the issue, I took it that he believes that even a
fairly small group has the incentive to lobby against software
patents, and that the network externalities of such a lobby should
cause it to grow (although he didn't express it that way).  He was
wondering why that doesn't happen.

To explain why not, you have to go beyond saying "the logic of
collective action is an impossibility theorem".

    Ben> You can view both these solutions [government and AAA-like
    Ben> lobbies] as being fundamentally the same, just varying
    Ben> whether to create individual motivations through a carrot or
    Ben> a stick.

I see your point, of course, but on this list, I'd rather not draw
such an equivalence.  The voluntary lobby that coheres by having a
number of common interests is far more compatible with freedom as I
understand it.

    >> (3) competition (and thus shrinking the pie) from free riders

    Ben> It is not that free riders shrink the pie, it is that people
    Ben> would individually be better off trying to be free riders
    Ben> themselves.

In the case of the alleged logic for patents, it _is_ important that
the free riders shrink the pie.  Even for free software: Russ McOrmond
himself describes a business model based on achieving a satisfactory
revenue commitment before releasing any work.

    >> (4) the dynamics of political organization are such that if we
    >> as a subset cultivate our gardens together, we will typically
    >> lobby the government for protection from "rose-eat-tulip"
    >> competition, and probably for direct subsidy as well.

    Ben> And this is an endemic problem with no good solution.

That's not a problem, that's an explanation of why we rarely see
successful purely voluntary lobbying groups like the FSF or LPF.

Well, I guess it's a problem, too, but that's not why I brought it
up.  :-)

    Ben> For instance most countries in the first world (with Canada
    Ben> being an extreme example)

Canada's military policy is extreme only in being rather principled.
:-)  As for size, it's the US military that's extreme---I don't
really see Canada going to the lengths of (say) Japan.  I suppose it's
smaller than France, but that's because AFAICT Canada has no
geopolitical ambitions, except perhaps in British Columbia and Quebec.

    Ben> But then they are left with no recourse when they disagree
    Ben> with the USA on how the USA should use its army...

They never had any (constructive) recourse, anyway.  That's what
"sovereign" means.  This is why I dislike the FSF assignment policy
enforced in many GNU projects; it concedes a lot of sovereignty to a
capricious agent whose goals and perspectives differ perceptibly from
my own.

Institute of Policy and Planning Sciences
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.