I received many, many reactions to my decision to publish Craig James' article on reforming the patent system. One of the emails was from Richard Stallman of the Free Software Foundation, who now continues the conversation.
Richard Stallman On Software Idea Patents
The supposition that software idea patents are inevitable is a form of
defeatism that is already visibly mistaken. The movement against
software idea patents in Europe, led by FFII (fii.org [sic. Correct address is ffii.org]) and supported
by organizations as diverse as Deutsche Bank Research and the
Confederation of Associations of Small and Medium-size Enterprises,
has already persuaded the European Parliament once. The outcome will
be so close that it is absurd to think you can predict the winner.
Yet even now, thousands of people who oppose software idea patents
still shy away from imagining we can prevent them. At my speeches, I
constantly encounter people who look for methods to reform the
system's details or to game the system, partial solutions that could
at best palliate the problem, because they simply can't believe in the
possibility of a real solution.
These people don't realize how little practical difference even a
major reform would make. We recently learned that Linux, the kernel
of the GNU/Linux system, is covered by 283 US software idea patents.
(If Linux is .25% of the system, as I read recently, the whole
GNU/Linux system would be covered by something like 100,000 US
software patents.) This is an overkill situation.
Imagine a patent reform so effective that it would cut the number of
issued patents in half. If such a reform were enacted today in the
US, all 286 patents that already cover ideas used in Linux would
remain. If such a reform had been enacted 15 years ago, Linux would
now be covered by around 143 US software idea patents today.
Practically speaking, this would not change the situation much. If
there are 286 malaria-infected mosquitos buzzing around your house,
killing half of them won't make you safe from malaria. "Get rid of
half the patents" sounds like a big change, but the chance that your
large program is safe would remain small.
The idea of enacting substantial reforms of software idea patents is a
remote possibility. It would require a long sustained strong
campaign. For completely rejection of software idea patents, we have
already mounted the campaign, already build up strength--aided by the
fact that this would really solve the problem, not just reduce it
slightly. If you want to support the campaign that has a chance to
win, support our campaign now!