Designing a Patent System

A while back, I posted a note about the fact that the federal District Court in the Western District of Pennsylvania (a.k.a., mostly, Pittsburgh) has adopted a set of Local Rules specifically to govern patent cases. Pittsburgh is the third court to do this; San Francisco (Northern District of California) and Atlanta (Northern District of Georgia) also have their own rules.

The Pittsburgh rules have been in effect for several months now, and while it’s far too soon to judge how well they work, it’s not too soon to reflect on some related topics.

What’s the point? I was on the committee that drafted and proposed the rules, and there were and are a number of things going on. Foremost in the minds of the practicing lawyers involved was the hope that a “favorable” set of rules would attract more patent litigation to Pittsburgh, and bring more business to local law firms. (More on what “favorable” means, below.) If the proposed patent reform legislation goes through and includes a new venue provision designed to stop filings in Marshall, Texas, we’ll have to rethink that part of the strategy! There were (and are) other important considerations. We wanted to encourage the bench to focus on the nuances of patent cases, so that Pittsburgh could begin to develop a reputation as a good place to litigate and try a patent case regardless of the specifics of the Local Rules. (This depends on judges’ denying motions to change venue, and by involving the bench in the Rules process, we hope that judges will do just that.) We wanted to send a signal to the local technology community that the legal community is a team-player when it comes to supporting local investment. The existence of a friendly (or at least knowledgeable) local court might give some assurance to entrepreneurs and investors that they’ll be treated fairly if they get hauled into court — or have to haul someone else in. And we wanted to put together a set of rules that differed in some important respects from the existing model, supplied by the Northern District of California. Uniformity is a big deal in the patent world, but sometimes choice is good, too.

How did we do this? With the blessing of the court, the local IP bar group appointed a committee, consisting mostly of practicing lawyers (each major firm in town was invited to appoint a delegate), along with a couple of in-house patent practitioners, an experienced arbitrator/mediator, and a token academic, to keep everyone honest (!). We met for about a year, breaking down the litigation process along functional lines, creating subcommittees, and having a lot of heated debate as a group. The rules were supplied to the court, which adopted them, with some minor changes, after following the appropriate notice-and-comment procedure.

What we didn’t do, though, was offer any open-ended mechanism for input from the bar or from the public. Given the breadth of perspective represented on the committee, I think that it’s unlikely that we would have heard anything that wasn’t aired fully in any case. And both the members of the committee and the bench (which has been an enthusiastic partner in this process) take pains to explain that the rules will be revised if they prove to be awkward or even unworkable in some way. Many of the Rules reflect compromises among members of the committee. Nonetheless, we’re now in the education phase of the project, explaining the “what” and the “why” and the “how” of the rules to local lawyers — and to the national patent community. And while lots of people seem to like the Rules in general, not everyone likes all of them.

How are the rules different? I highlight three salient features:

First is that a default Stipulated Protective Order is entered automatically when the case is filed. The model Order is part of the Rules. It’s a two-tier protective order. Its most controversial provision, from what I can tell, is the default requirement that source code be disclosed (on an Attorneys’ Eyes-Only basis) in appropriate cases. The thinking here is that if the Order needs to be modified, it can be modified, but the burden should lie on the side of the party alleging prejudice from disclosure. If the burden were reversed, then all too often the parties would find themselves back in the position that we usually see in patent cases, wasting time and money arguing over a protective order before disclosing anything at all. The point of the default Order is to cut off that dynamic in most cases.

Second is that claim construction occurs somewhat earlier in the case compared to the timing of claim construction under the Northern District of California rules. Also, the mechanics of claim construction (the time and manner of identifying claim limitations in dispute; identifying relevant evidence; briefing and argument) are relatively compact. This was especially important to the in-house counsel on the committee, as well as to lawyers who primarily represent plaintiffs. (This is the sense in which the rules are thoughtful to be somewhat “favorable,” at least to prospective plaintiffs. There are some in town who refer to the Rules as “balanced,” when compared to the Northern District of California rules. The latter are thought by some to favor defendants.) The first group looked at this as a risk management issue; the second group looked at it from a strategic perspective.

Third is that not all of the rules are framed in terms of a dispute that assumes that the patent holder goes first. Going first can be a big advantage in litigation, for a host of reasons. Going first is one reason why prospective accused infringers sometimes like to file preemptive Declaratory Judgment (DJ) actions against patent holders. If Local Rules specify that the “party alleging infringement” retains the right to go first (say, in arguing claim construction), then the DJ plaintiff loses some of its advantage. In Pittsburgh, the claim construction rules refer to “plaintiff” and “defendant.” A DJ plaintiff gets to keep that advantage.

On the problems of designing a legal system: From a practical standpoint, there’s little empirical evidence to support the premise that Local Rules can influence economic development more broadly; until recently, there was relatively little empirical evidence to support the (narrower) theory of jurisdictional competition. Time will tell. Unless you’re part of a couple of relatively narrow technology communities, Pittsburgh isn’t generally known for cutting-edge innovation. A little out-of-the-box thinking can’t hurt.

Somewhat more abstractly, the development of the Rules is instructive because — particularly when it comes to claim construction, which is the only dimension of the Rules that isn’t derivative of the Federal Rules of Civil Procedure — it illustrates two things. Taken together, they manifest the Federal Circuit’s “black box” approach to claim construction. First, in some sense, regardless of what we do with Rules at the trial level, litigants and trial courts are still stuck with de novo review. That black box has four solid walls, and there’s no escaping it. Second, it illustrates how little guidance the Federal Circuit (and the Supreme Court) has really given us in terms of how claim construction in the trenches is supposed to happen. For all of the academic talk about theories of language and meaning as applied to patent claims, there’s nothing quite so humbling as designing a little legal system from stratch — and arguing over how many days the defendant has to respond to the plaintiff, and whether a week is enough time to depose experts on claim construction issues. The contents of that black box, which still matter very much on a day-to-day and case-by-case basis, are entirely up to us.