« RIM Down | Main | Microsoft Bows to Eolas »

December 3, 2005

Uncertainty

The prospect of passage of the Patent Reform Act of 2005 will soon expire with the year-end congressional recess, but the impetus for patent reform won't. One driving force is the myth that the current patent system retards innovation by granting bad patents. The issue to address is not an imbalance between promoting innovation and protecting the rights of patent holders, it's in strengthening patent validity by reducing uncertainty.

A lot of flimsy patents were issued in the early 1990s. As a professional prior art searcher, I particularly enjoy an assignment to invalidate an early '90s patent, as the invalidity odds peak; past bad examinations make me look good to my clients. Many of those patent alumni are now seeing action in enforcement, with a corresponding outcry that patents are bad. Some of the problem of skimpy examination particularly existed in a bias towards companies which patented heavily in a particular area, with a concomitant examiner presumption that the patentee was well informed of the prior art, thus taking the load off the examiner that a prior art search was requisite.

Patents are presumed valid; legally, rightfully so. The patent office supposedly performed a thorough examination before granting the patent. If there is a problem with the patent office examination process, the solution lies in improving the examination process. Applying new band-aids, such as post-grant opposition, may simply further complicate the system without offsetting benefit.

There are complaints of pendency: that the patent office takes too long to grant a patent. But only applicants make that complaint, specifically applicants that want a patent as an investment inducement - a way to fill out the business plan.

Certainly, delay is undesirable. But better to get it right after awhile than get it wrong in a hurry.

The real problem with the patent office has been a lack of resources - examinations in the past have not been what they should have been, and given the rising backlog, the pressure to reduce the backlog may be having warp effects.

The patent office is the only government agency that actually makes money from its fees. Alas, Congress and successive administrations have seen fit to treat the patent office as a piggy bank to be pilfered. That, as much as anything, has weakened the patent system.

The patent office has taken note of its bad publicity in recent years, and has responded, but not always for the better. For an examiner to clear his desk quickly of backlog, it appears now that hasty rejections are more common. While seemingly showing quicker processing, it only adds to the cumulative backlog, as applicants have to fight longer, including more resource-sapping appeals, for a fair vetting.

It would be wrong to think that the patent office lacks prior art resources. The patent office even has a program to identify to examiners databases of particular note to the different art units. Most examiners know where to search for prior art. The problem is the time it takes to do a good prior art search, and examiners are pushed to crank the backlog.

There is already a post-grant review system: re-examination. The unspoken problem with re-examination is that ex parte attorneys, who very much like to feel as if they are driving a process, don't have any control in the current re-exam process - they only get one shot of input. Fact is, though, if touting re-exam, either you have the prior art and arguments to invalidate a patent or you don't. One ex parte shot ought to be sufficient - the patent office should be able to take it from there. And so it seems, with some rather high-profile re-examinations in recent years. As to a more extensive, and expensive, pre- or post-grant opposition process, it's unfair to an applicant to be held hostage to some moneyed interest that would like to kill a patent application because it's standing in the way of lucratively infringing. That's one of the real motives behind some high tech companies promoting patent "reform".

Litigation defendants regularly use re-examination as a last resort. It's a last resort precisely because of re-exam's one-shot no-control, and the effect a re-exam may have. In NTP v. RIM, for example, RIM has lost the case, and is grasping at straws with the NTP patents re-exam. Why did RIM wait so long to initiate the re-exam? Because if a patent passes re-exam, it's presumption of validity is practically invincible. Better to roll the dice in court, with re-exam as last resort. In RIM's case, the re-exam may come too late; bad luck with the timing.

Further raising the cost of patenting with a more drawn-out examination process would limit access of individual inventors to the patent system, resulting in less innovation. That may be exactly what high-tech companies might like, not to face so many patents by individual inventors, but it would be a poor trade-off for this country's competitiveness.

Another aspect of uncertainty is knowing what a patent's priority date is. America is the only country with a first-to-invent system, and that raises the costs of litigation in some cases by making invalidation more uncertain, as an applicant can swear back, if necessary, and so the priority date can remain unknown until litigation is well underway.

The one-year grace period for invalidation is also a formula for having bad patents, where the same invention arose contemporaneous among several parties, and so, from a social benefit standpoint, no patent would exist, but, under the current system, the first to file/invent gets a monopoly. A first-to-file system with no prior art grace period would make for a more competitive patent system - competitive from an invention standpoint, with fewer patent monopolies when they shouldn't exist, and it would eliminate the current uncertainty of priority date, which would reduce litigation cost and uncertainty with regard to in/validity.

There are interesting hypothetical counter-arguments to first-to-file, including that it would result in rushed, sloppy applications. If so, then those applications could have enablement problems, and thus not be granted, but provide prior art preventing later patenting of the same invention; that seems a decent outcome. Considering that the rest of the world has a first-to-file system, it must be at least tolerable.

Reducing uncertainty in the patent system is the best remedy to having a fair system. The nut of the problem with patent litigation has been that it's too much a roll of the dice. Since it's big money at stake, take the gamble. If the writing on the wall of patent validity was clearer, were the presumption of validity really a reasonable presumption, the patent game would be less of a crap shoot.

Posted by Patent Hawk at December 3, 2005 12:22 AM | The Patent System

Comments

Please review the further arguments against first to file at www.piausa.org. Thanks!

Posted by: Joe Eckstein at December 9, 2005 9:28 PM