Patents (was Re:Re:Microsoft/Novell)

[Dave Mason]

As the author of two preliminary software patents, let me state that I
am extremely glad that the company that was making the application
didn't have the resources to pursue the final patent.  I was a little
queasy about the idea of software patents beforehand, but as we went
forward with the process and I looked more at existing software patents,
I became increasingly uncomfortable.

Software patents are either:

1) patenting ideas (which is explicitly outside the domain of patents), or

2) patenting the use of the idea in one particular context (which should
be useless, because nobody else will have that context, but in practice
it gets overinterpreted, used for bullying, and verges heavily into
patenting ideas).

While my exploration of software patents has perhaps been biased, I have
looked at dozens of software patents, from IBM, Microsoft, and others.
*All* of them were clearly patents of ideas *and* obvious to a
practitioner, but tried to wrap them up in some application in order to
be patentable.  From an examination of recent USPTO patents, they
clearly needn't have bothered wrapping them up.  In the last 5 years
patents have been issued for (1) the wheel, and (2) swinging.

Business practice patents are in the same category.

Hardware patents *may* make sense because of the capital-plant
investment, although my belief in even that has been seriously
challenged as I've thought about this for the last several years.

Lifeform/DNA/drug patents have similar properties to hardware patents
(huge up-front costs) but are a huge impediment to scientific progress.
As an example, there are currently 34 patents relating to malaria, so
anyone seriously attempting to do research for a cure needs to get
licenses for most or all of those patents before they can start.

Of the 2 patent applications in which I was involved, one was between a
software and a business-practice patent of dubious non-obviousness.  The
second was a fore-shadower of some of AJAX web technology, and was
rather clever so I might accept as being non-obvious, but subsequently
we uncovered related prior-art (which the USPTO would never have found).
I'm confident that both patents would have been granted.

On the other hand, I'm a big fan of copyright, particularly when paired
with a Creative Commons license (of which there are several and which
more-or-less subsume GNU and BSD type software licenses).

../Dave
--
The Toronto Linux Users Group.      Meetings: http://gtalug.org/
TLUG requests: Linux topics, No HTML, wrap text below 80 columns
How to UNSUBSCRIBE: http://gtalug.org/wiki/Mailing_lists