blog

At TACD's Brussels conference, we've had an interesting panel that recounted the campaing about software patents in Europe. The panel consisted of the liberal EU parliamentarian Sharon Bowles, who is also a patent lawyer; the (not very) covert Microsoft activist Jonathan Zuck; and Florian Müller nosoftwarepatents.com

While there were a lot of the old arguments tossed about, there were some interesting points. Sharon Bowles made a rather daring point when she said that "I never knew a single patent attorney who wanted [american style software patents]". To her mind, there's an area that shouldn't be covered by patents, but as she presumably didn't feel like it, she couldn't define that. Defining it would cause great "collateral damage" to the "electronics industry".

All speakers feigned to be blissfully unaware that software is protected by copyright, and that this does not seem to have harmed the industry so far. Some more far-out claims were that software patents would somehow be an answer to the problem of illicit copying of proprietary software. Mind you, this is not your local coffee-house chat. These people are experts in their field. It's probably best to see any such mistakes as intentional.

A more interesting argument came from Jonathan Zuck, who says that his organisation represents about 3000 small and medium enterprises (try joining, it's free and they don't vote). When I mentioned that software protection happens via copyright, he added that software is also protected by trade secrets. But trade secrets are becoming less feasible due to concerns about interoperability, state sovereignty.

Though minor, this is a point we might want to prepare ourselves to run into more frequently in the future. He didn't answer to my question about how large a percentage the five largest funders of his organisation contribute.

Sharon was kind enough to give a hint for those aspiring to have something patented. Since it's already impossible for patent lawyers to do a reasonably good clearance of prior art - i.e. to check that the invention doesn't already exist - she recommends that you should just apply for a patent and let the reviewers do the checking. According to her, that is the cheapest way to do it.

The ensuing discussion reached a charming conclusion when a German physicist got up and spoke rather passionately about how stupid the idea of patenting software is, since it amounts to patenting mathematic formulae. Though rather new to the debate, he hit the core point: It really *is* a stupid idea to patent mathematic formulae. Period.

For those who read German, Stefan Krempl of heise.de is reporting from this conference. (Thanks, Markus)

After Nobel prize winners Joseph Stiglitz and John Sulston last week, now the Wall Street Journal carries an article about the problem that the patent system has become. It's US-focused, but it pretty neatly outlines how the debate on a mild patent reform there sets the pharmaceutical industry against technology companies:

Yet the fault line over patent reform signals the
deeper problems. Many pharmaceutical companies lobbied against the
proposals, fearful of reduced value in their key intellectual property.
In contrast, most technology firms supported the reforms, worried more
about uncertainty in the law than about the value of their patents.

Both sides may be right. New empirical research by
Boston University law professors James Bessen and Michael Meurer,
reported in their book, "Patent Failure," found that the value of
pharmaceutical patents outweighed the costs of pharmaceutical-patent
litigation. But for all other industries combined, they estimate that
since the mid-1990s, the cost of U.S. patent litigation to alleged
infringers ($12 billion in legal and business costs in 1999) is greater
than the global profits that companies earn from patents (less than $4
billion in 1999). Since the 1980s, patent litigation has tripled and
the probability that a particular patent is litigated within four years
has more than doubled. Small inventors feel the brunt of the
uncertainty costs, since bigger companies only pay for rights they
think the system will protect.

Link