[dvd-discuss] Re: More On State Sovereignty and Senate Bill 2031, Leahy's "IP" ProtectionRestoration Act

[Seth Johnson <seth.johnson(at)realmeasures.dyndns.org>]

(Jonathan replies.  -- Seth)


-------- Original Message --------
Date: Sat, 13 Apr 2002 11:42:27 -0400
From: "Jonathan S. Shapiro" <shap@eros-os.org>


Seth, All:


It may have appeared that Seth and I are in disagreement. We
aren't, so I want to try to clear things up.

In my earlier note, I specifically tried *not* to give an
opinion on Bayh-Dole. I wanted the readers to form a
judgment based on an accurate understanding of what
Bayh-Dole does and does not say. I didn't say anything about
S-2031 because I didn't know enough about it to have an
informed opinion.

Saying that Seth was "flat wrong" was excessive. He
described some of the particulars of Bayh-Dole in a way that
was not fully accurate.  Specifically, he said that
Bayh-Dole *requires* patenting when it does not. This was
the main point of fact that I wanted to correct.

Since I have apparently muddied the waters rather than
calmed them, here is what I actually think:

Concerning Bayh-Dole:

From a standpoint of principle, Bayh-Dole was a mistake. You
and I (the taxpayers) pay for academic research, but we are
not getting compensated. This is theft, plain and simple.
Legally supported theft seems to me to be bad policy.

There is a strong case that the policy prior to Bayh-Dole
was ineffective.  Inventions that do not get into the world
don't help anybody. From a policy perspective this leads two
principles into conflict: the principle of ownership and the
principle of "greatest good for greatest number."

The theory behind Bayh-Dole boils down to: "Unless there is
some exclusivity, companies will not invest in these ideas.
We want these ideas to end up providing some concrete
benefit to the public, so let's introduce the exclusivity
that is needed by encouraging patents and allowing the
inventors to license them."

The problem with this story lies in the first sentence. You
may remember that drug companies used the same argument to
press for drug patents and biotechnology patents. The test
is to turn the question around: if the claim is true then
there should have been essentially no investment *before*
patents.

With Bayh-Dole it is pretty clear that there was (for the
most part) no investment before patents existed. The problem
is that there were other factors involved, and we cannot
know whether patents were in any way decisive. In 1980,
there was no world wide web. Universities had no means to
promote their ideas at low cost, and the result was that
industrial companies simply never learned about much of the
research that was done. Also, it was not common practice in
prior to Bayh-Dole to bring research projects along until
they could be practically demonstrated. Even today,
academics do not get credit for releasing things that people
can use directly. If I tell you "here is a new technology
for spraying aerosols" you will yawn pretty quickly. If I
tell you "here is a new nasal inhaler that makes it easier
to take your asthma medication" the value of the idea is
suddenly real and immediate.

But in other areas we can see that the argument is plainly
false. There were very profitable drug companies before
there were drug patents. There were very profitable software
companies before there were software patents. Since there
were no patents, people clearly invested in these companies
without them.

What patents are really about is maintaining the status quo.
As long as there are no patents, small companies can be
started and use the work of others. At the end of the day
the resulting innovation is good for the public. Patents
allow large companies to squash these small companies. In
theory, a patent allows an individual to demand royalties
from a large company. In practice, the legal fees required
to file suit are excessive. Patents have become a game of
trading cards and deep pockets.


But Bayh-Dole is tricky to judge. It is clear that today
there *is* a world wide web where things can be found, and
that more and more academics are building proofs of concept.
Academia in the United States has come slowly to think this
is okay. In some measure, proofs of concept are okay
*because* Bayh-Dole was put in place.

Having said that, I think that Bayh-Dole has served its
social policy purpose and should now be repealed. The
ability to build proofs of concept will be a lasting social
change within academia, and at the end of the day, academic
patents are not good for either the nation or the academics.

Regarding S-2031, I can only say that shoring up a law that
should be repealed seems like a bad idea. S-2031 is such a
screwy, complicated bill that in the long term S-2031 will
cause Bayh-Dole to be replaced if it is passed. The problem
is that the end result may be worse than the current state
of affairs.


There is one key point about Bayh-Dole that I think needs
emphasis: as currently formulated, it needs the support of
the inventing faculty to obtain patents. Today, those of us
in academia who *choose* to do research for the public good
have the *option* to do so. We need to be careful here. When
you reopen a social policy issue in law, it is very easy to
get an outcome much worse than the one you have. At the
moment, with corporate interests driving law more strongly
than ever, I think this may be an unwise time to reopen this
particular can of worms. I am particuarly concerned about
this in the context of the EROS work -- I'll say more about
EROS later.


> Jonathan does something of the same thing: focusing on
> issues of "patent management" and exclusive and
> non-exclusive licensing of discoveries that are presupposed
> as properly being allowed to be proprietary, rather than
> taking up the issues associated with making property out of
> general scientific knowledge.  These are what he contends
> with every day.

Jonathan (that's me) has mixed feelings on patents
altogether. In software, it is utterly clear to me that
there shouldn't be *any* patents, so the issue of whether
it's okay to patent some stuff and not others seems a day
late and a dollar short.

That said, there *are* software patents today. Given that
they exist, and that fixing this is a separate problem, I
would say "he who pays should control", and in this case
that means the taxpayers should get to use the inventions.

> Jonathan... seems to think that the fact that
> universities get their funding for patents from outside
> sources is just a foolish government oversight...

It is certainly not a foolish oversight! I think this was
specifically an objective of the policy makers. By requiring
universities to come up with their own funds to generate the
patents, they effectively required universities to market
their ideas. In a nutshell, this was the whole *point* of
Bayh-Dole.

Regards, and apologies for the confusion I have caused.


Jonathan S. Shapiro