Subject: Re: "a better world" (was Re: GIF/LZW patent)
From: <>
Date: Fri, 29 Sep 2006 18:12:44 +0900

Craig Brozefsky writes:

 > <> writes:

 > > The marginal cost of producing software is high, and increases rapidly
 > > with increasing functionality and complexity.
 > Really?  I'm not sure what "high" means here, relative to what?

Relative to serving an HTTP request.

 > So much friction, so little benefit all constrained to a very small
 > population.

Here, I've got an envelope.  Let's do a simple, plausible calculation.
If a hundred million people get a penny a day extra enjoyment from
some innovation that gets to market a year earlier than it would have
without the incentive and/or funding that a patent generated for it,
that's a $365 million ($366 million in a leap year) benefit.  That
will pay for a lot of frictional patent search![1]  And if it's *three*
cents, well, as the Senator said, "now, we're talking *real* money!"

Not to mention that IBM *alone* has several million stockholders, more
than all the programmers (let alone free software hackers) in the

And to repeat what Rob Cameron said, in the meantime, the U.S. and
Japan at least have software patents.  The system desperately needs
reform, and it will get it at the behest of IBM and the rest of that
herd.  Things *will* get somewhat better for the majority of software
producers.  For that reason alone, abolition is not in the cards for
many years, IMO.

[1]  In case you haven't noticed yet, let me point out that the
chilling effect gets the same leverage from low cost of reproduction
that the positive incentive effect does.  But having dropped an easy
$365 million on the floor the way you just did, are you really sure
you can estimate the relative magnitudes of the positive and negative
incentive effects so easily?  And suppose there was an effective
patent search industry (remember, there's easily $billions to pay for
it, but I doubt it will cost 1/100th of that)---how big is the
remaining chilling effect?