Re: [tlug] RE: Shuttleworth Interview at Red Herring (Roger Markus)

["Stephen J. Turnbull" <stephen@example.com>]

Fergal Daly writes:

 > On 06/01/07, Kenneth Burling <meyaden_emrys@example.com> wrote:

 > > I certainly hope this is not the case.  The current stream of judges seems
 > > to have little understanding of patent law,
 > 
 > I don't think you can say that the current set of judges have little
 > understanding of patent law. Their only job is to understand the law.

What you can say is that the Slashdaughters have *no* clue about
patent law, and almost none about copyright law.  Most of what you
read outside of the legal journals is plain FUD (about half of what
professional economists produce is, AFAICT), and free software
advocates tend to stoop as low as, say, SCO's legal team.

In the U.S. the problem is not the judges, but the USPTO which is
woefully understaffed for its task, and the legislation itself which
basically is biased toward granting patents, and provides almost no
protection against patent-based extortion.  I don't know much about
Japan, but what little I do know suggests that the IP law here is
pretty hozed.  Eg, the way they've deliberately shot themselves in the
foot vis-a-vis search engines (cf. this morning's Asahi).

 > However it's fair to say they have little understanding of the
 > consequences of applying patent law to software the same way they
 > apply it to widget making.

*sigh* Same mistake you just tried to point out yourself!  Lawyers and
judges (especially the latter) have fairly good understanding of the
consequences of applying law.  But that's completely irrelevant.  *In
common-law (Anglo-American) systems, they simply care more about the
consistency of the law.*  And guess what?  They're right to do so.  A
consistent legal system is a much bigger deal than a bad patent law.
As the Japanese started to find out ca. 1995 ....

But in fact, *we simply have no clue about the actual effects* of
thorough-going application of patent law to software.  For 50 years,
people have been talking about reusable software and object
orientation[1], but the only time software developers actually reuse
software is when it's proprietary.  One has to wonder if a system
which actually enforced proprietary restrictions might actually have
the effect of inducing reuse.  ("Actually enforce" is in contrast to
the current system of screwing the unlucky and enabling extortion.)

Note that Stallman has been badmouthing my project for more than a
decade, and has *never* reused anything we've developed since 1994.
In fact, he bitches like crazy that we reuse stuff developed by GNU
Emacs, and has recently decided that he's unwilling to relicense GFDL
documentation under the XEmacs documentation license (of which he and
Moglen are the authors, of course), forcing us to write the manual
from scratch even when we can use the code.  The point being that the
structure of free software allows people to be totally undisciplined
about reuse, and the most prominent opponent of software patents is
one of the worst offenders.[2]  Coincidence?  Not likely!

The BSDs are much better, of course, but even there there's a certain
amount of not-invented-here disease.  And does the world really need
97 scripting languages for the Web starting with the letter "P"?

Of course, under current circumstances where the cost of simply
determining whether a relevant patent exists for *one* technique is in
the hundreds of dollars, and determining whether your program actually
infringes is an order of magnitude bigger, where you may have
thousands of such components in your product---software patents are a
bad idea.

But if the bar for granting patents were raised, and the fees for
application and grant increased, and those additional fees were
applied to better training for examiners and implementation of a high
quality, no-fee, public access search facility for existing patents,
and an adversarial system for patent examination (ie, after the patent
examiner grants the patent, there is a 6 month period where the public
is invited to submit potential prior art etc, and any royalties go
into escrow), and maybe even a "fair use" exemption (I like the
"International Characters" patent covenant) ....  It's hard to be sure
that that would be a bad thing for software users and most developers,
although licentious software advocates would certainly suffer.

 > For the lower courts though, they have no choice but apply it like
 > that. it's their job to apply existing law and precedent. It's only
 > when you get up to the supreme court etc that the judges are
 > supposed to start reversing precedents or radically reinterpreting
 > things due to new circumstances and subtleties,

That may be true in Britain or Ireland.  It is most definitely not
true in the U.S.  The role of U.S. judges, and especially the Supreme
Court, is to preserve the consistency of the law.  It's true that
justices are characterized as "activist" or "original intention"
according to whether they resolve conflicting precedents in favor of
addressing current circumstances or in favor of the closest possible,
literal-minded analogy to the problem the legislation was intended to
address.  But the degree of freedom they have is typically quite
small, and most decisions by SCOTUS are unanimous---the differing
opinions of justices will typically differ only in the rationale,
which may affect future precedents created by other judges.


Footnotes: 
[1]  Yeah, I know, "object" is a 90s kind of word.  Nevertheless.

[2]  Yes, I have a personal interest, but the point is the social
costs of his behavior.