Subject: Re: EY invests in online patent exchange
From: "Karsten M. Self" <>
Date: Thu, 23 Sep 1999 08:28:42 +0000

"Stephen J. Turnbull" wrote:

> Any value in a patent system must be due to encouraging _future_
> innovation.  It's pretty clear that patents encourage a lot of
> innovation in pharmaceuticals.  The numbers show that the typical
> drug, with a share of total R&D allocated to it as an overhead cost,
> takes about 10 years to repay investment at monopoly prices.
> Curiously enough, the effective length of patent (FDA approval to
> termination) is about 10 years.  Development of a generic (a) reduces
> price of the branded drug by 40-60% in typical cases, and (b) takes
> about two years.  QED: current levels of R&D in drugs would be
> dramatically unprofitable without patents.
> What's _different_ about software?

We might ask:  what's different about drugs?

Once basic safety and efficacy have been established (granted, no mean
feat), much of the remaining R&D required to bring a drug to market is
in satisfaction of government approval requirements.  For an
international drug company, this means satisfying submission
requirements of _several_ countries, though this can be approximated to
the US, EU, and Japan for most purposes.  I don't have hard numbers, but
I could probably come up with some rough estimates on request.  

Drug development becomes a major crap shoot with thousands of basic
science leads thinning out to a handful of products -- a dozen or so if
you're lucky, none if you're not....  Failure to market a drug can occur
because it's unsafe, ineffective, too expensive, worse than
alternatives, didn't get approval, or was clobbered by a competing new
drug once it did hit market.

While the typical SW patent debate runs along the lines of "patents are
bad for software but necessary for drugs so you can't do away with
patents", I'd like to see the debate turned on its head for once:  we
should do away with patents and give drugs the sui generis protections
they require which are peculiar to the scientific and regulatory process
they must pass through.

Software doesn't (currently) face a similar hurdle.  Why should it be
chained to a system which works very well, but for an extreme instance
of technology development.

Karsten M. Self (
    What part of "Gestalt" don't you understand?

SAS for Linux:
Mailing list:  "subscribe sas-linux" to    
  1:17am  up  4:18,  0 users,  load average: 0.00, 0.00, 0.00