Saturday, January 19, 2008

TechCrunch wrong on reform

Erick Schonfeld at TechCrunch writes of patent reform 2008:

A much bigger problem is the general quality of patents that are approved. Hire and train more patent examiners and raise the bar for which inventions deserve a patent, and you will fix a lot of the underlying problems with the broken patent system.

The Reform Act of 2007 does try to take a few steps in this direction by making provisions for post-grant challenges that could nullify patents that are too obvious or replicate prior inventions. It also tries to streamline the review process by switching to a first-to-file, instead of a first-to-invent, system. This latter provision seems to favor big companies with patent-filing bureaucracies rather than small-time inventors, but it is the way most other countries do it and makes it a lot simpler to determine who was first. Neither of these steps, however, adequately address the bigger problem of poor patents being granted in the first place.

The second sentence is a non sequitur. Post-grant review is NOT a step in the direction of quality patents. If patents are not being examined properly, the right approach is to fix the examination process, not to REVIEW results of a (presumptively bad) examination process.

Schonfeld also writes:

The sad truth is that patents have always been more valuable in the courthouse than in the marketplace. But focusing on the size of awards is treating the symptom instead of the disease. Too many patents are granted that should never see the light of day in the first place.

IPBiz suggests Schonfeld review "how" the Wright Brothers did at the courthouse, in terms of "value" for their patent. Further, Chester Carlson, of xerography fame, had to create his own marketplace (IBM, Kodak, et al. were NOT listening), and he extracted no value from the "courthouse." Most patents are not litigated at all, and there are plenty of patents that have had value outside of the courthouse.

Of a comment on an earlier IPBiz post:

1. It seems kind of insulting that you directly analogize a putative inventor's patent application to a standardized, interchangeable car part. I much doubt that Deming himself would have thought his methods directly applicable to "inspection" of patent applications.

2. Deming did not generally favor 100% inspection and preferred more limited methods of statistical inspection. Because of this, I doubt that Deming would have like ex partes exam, even if patent applications could be analogized to car parts.

IPBiz notes-->

As LBE has written, Deming explicitly used "quality" techniques to analyze government services, including that provided by the US Post Office. Further, Deming did NOT consider quality approaches limited to methods to make standardized, interchangeable parts. The poster was dead wrong in point 1, and obviously not familiar with the writings of Deming.

Of point 2, Deming did not favor inspection AT ALL to resolve problems with production. One inspected to identify the problem, and then one fixed production. Thus, the commenter's point 2 missed the whole issue. Deming would NEVER advocate using an inspection technique to resolve a production problem. Talking about "what kind of inspection" is not the issue.


Post a Comment

<< Home