United States Patent Reform – Give Up and Do It Right?

Lee Hollaar is a professor of computer science at the University of Utah, where he teaches engineering and intellectual property law as well as networking and operating systems.

On June 8, 2005, the Patent Reform Act of 2005 was introduced in the United States House of Representatives, to whatever fanfare accompanies patent legislation. Its primary sponsor, the Chair of the House Judiciary subcommittee for intellectual property, called it “the most comprehensive change to U.S. patent law since Congress passed the 1952 Patent Act.” Because of the extensive discussions with stakeholders including revisions to circulated draft language, and a number of hearings before it was introduced, he predicted “In a couple of weeks, there should be a general agreement on a piece of legislation” and schedule the start of a fast track to passage.

That didn’t happen, and after the Patent Reform Act of 2006, the Patent Reform Act of 2007, and the Patent Reform Act of 2008, it appears that Congress may be no closer in passing the Patent Reform Act of 2009. And if it does, it certainly won’t be the big change that was promised.

So, what went wrong?

First, the reason of the big change was never clear. Was going to “first inventor to file” a goal, or just one part of “harmonization.” But harmonization requires more than just changing patent priority from first-to-invent to first-inventor-to-file. The original House bill also included harmonizing patent applications by getting rid of the United States’ special “best mode” requirement, harmonizing what is prior art by not having the changes effective until Japan and Europe join the United States (and Canada) in having a one-year grace period for applicants, and creating a post-grant opposition proceeding to handle those cases where national interpretations of patent law differ.

Those provisions were scattered throughout the bill, and over time Congress seems to have forgotten that they were part of harmonization. The latest Senate bill does not drop “best mode,” but strangely makes it a requirement that can’t be brought up in a post-grant review, but could be used to invalidate a patent in litigation. It does not couple the change in priority to first-inventor-to-file with Europe and Japan adopting an inventor grace period, so, even if it becomes law, there won’t be harmonization and its benefits.

The proponents of the bill also included a wish-list of other provisions, since you might as well get some special favors for your support of the legislation. These included making it more difficult to raise the issue of applicant conduct to invalidate a patent, limits on continuing applications, how damages should be handled when the patented invention is just a part of the infringing thing, and when injunctions should be granted.

The latter caused the first break between the proponents, killing any hope for fast passage of the 2005 bill. It was dropped from the bill after the Supreme Court moderated the Federal Circuit’s rule on injunctions in its 2006 eBay v. MercExchange decision, even though that case just reiterated a general rule in line with the original 2005 bill, rather than the specific language each side wanted. The current split over the damages provision is a replay of the injunctions fight, with groups of proponents trying to get their dream language as a replacement for the general and sensible language in the 2005 bill.

And along the way, other wish-list provisions were added to try to gain supporters, and then dropped, further obscuring the big goals of the legislation.

Right now, it appears that goal has shifted from harmonization to whatever it takes to pass a “patent reform” bill to end this four-year exercise. It’s unclear whether the “reform” will actually improve things, is change for the sake of change, or will make things worse. Certainly dropping the requirement that Japan and Europe adopt a one-year grace period as the quid pro quo for the United States going to first-inventor-to-file will likely mean that any future harmonization won’t include a grace period, since the United States will have already given up its bargaining chip.

In the process, language is being changed overnight to get senators to vote for the bill, with seemingly little thought of how the changes affect the original goals of the bill or even the existing patent system. And that caused one of the bill’s original sponsors, Orrin Hatch (R-Utah) to vote against it in the Judiciary Committee, saying “My primary purpose for doing this bill was to improve patent quality and limit unnecessary and counterproductive litigation costs. I do not believe the bill, in its current form, accomplishes these goals.”

The great architect Daniel Burnham said: “Make no little plans. They have no magic to stir men’s blood and probably will not themselves be realized.” He could have been talking about “patent reform” in the United States.

So, what should a patent reform bill do that will improve the patent system for the United States and the world? In a follow-up to this, I’ll tell you my ideas, but right now, it’s time to hear from you …

Professor Hollaar is a patentee and registered patent agent, and has been an expert consultant in a number of prominent cases, include antitrust cases against Microsoft. While on sabbatical during the 1996-1997 academic year, he was a Committee Fellow with the United States Senate Committee on the Judiciary, working on the last major patent reform legislation as well as what became the Digital Millennium Copyright Act.  He is the author of Legal Protection of Digital Information, online at digital-law-online.info.

  1. I think one of the problems plaguing the 2009 versions of the Act is that there are so many other problems going on within the United States right now that everyone is just trying to pass the Act just to get it off of their minds. I haven’t heard President Obama speak about patent reform in quite awhile (understandably so), but perhaps it is time he takes a moment to think about it before an Act is passed that creates more problems than it solves.

    It also seems that Americans have not looked towards other patent systems around the globe for some ‘inspiration’ as to how to improve their patent system. Because the United States is often regarded as the leader in innovation, I feel that it is important the United States consider and adopts some of the good aspects of European and Canadian patent law in order to complement their position as a global leader.

  2. While having all stakeholders agreeing on a comprehensive patent reform bill has proved difficult, the courts have made progress in resolving issues that have created controversy in proposals to reform the patent system. Over the past few years, courts in the U.S. have rendered decisions on what obviousness means for patents, the extent to which business methods can be patented and how injunctions are used.

    Perhaps a more favourable course of action might involve giving time to the USPTO to interpret the recent decisions, rather than complicating issues further by hastily passing legislation. In the meantime, Congress could focus on making the reforms that all stakeholders agree are necessary.

  3. Dan Burk and Mark Lemley have just published a book, “The Patent Crisis and How the Courts Can Solve It,” that is well worth reading.
    But I’m not as confident as they are that the courts are the solution to the problem, at least not without some legislative help. I think that Bilski shows how the “case or controversy” and stare decisis limitations make it difficult for the Federal Circuit to play the role Burk and Lemley suggest, especially for software-related patents. That decision tries to resolve the business method patentability line while also trying to follow confusing and inconsistent Supreme Court precedent. It doesn’t do either well.
    Regarding software methods, it simply kicks the can down the street by opening up a new debate over what’s a “particular machine or apparatus.” And as much as it tries to appear to follow Benson, it doesn’t, since under the test it gives, at least claim 8 of Benson should be patentable because it requires a particular machine — one with a reentrant shift register. (As Judge Rich notes in his CCPA opinion, the PTO Solicitor called this limitation a “most embarrassing phrase.”) It’s not a method running on a general-purpose computer, but a way of describing a particular piece of electronic circuitry not by how it is wired, but by how it does what it does.
    Because Bilski was really about a non-computerized business method, the Federal Circuit was hamstrung to address the broader issue. Sure, they said they wanted amicus briefs on the broader issues, but in the end all they could do is include dicta once they said that not anything expressible as a series of steps is patentable.
    Another example of the limitations on the courts comes from Beauregard. That was a case brought because the important infringer of a software patent will be the company producing and distributing the software, but often it is only a marginal infringer (for method claims, and they only run the software for testing) or an inducer or contributory infringer. Finding liability as a contributor can be a problem, since most computer programs do a lot of other things besides what was patented — “substantial noninfringing use.”
    Rather than go to Congress for a simple change to the infringement statute, which could be tailored to address any problems, IBM decided to do some tricky claiming and hope that the Federal Circuit would allow that as a backdoor way to make the right party liable. Unfortunately, the PTO “threw in the towel” so the Federal Circuit never got a chance to bless such claims. Now, we have Judge Patel in Cybersource extending Bilski so that claims that are clearly for an article of manufacture and not a method are also unpatentable.
    In fact, that case makes clear that if you want to change the law through the courts, you either have to get your patent in light of the change you want and hope that it stands up during an infringement suit, or you have to make sure that you lose at the BPAI but have a strong enough case to win at the Federal Circuit. The worst thing that could have happened for Bilski is if the BPAI had found in his favor, and then years later when he tried to litigate the patent, the Federal Circuit found that it really wasn’t statutory subject matter. (Bilski has made thousands of software method patents questionable, with little the patent holder can do if they are outside of the reissue window and problems if they do go that route.)
    The other problem with using the courts that Beauregard illustrates is that it may not be possible to draw a desirable line because there is no statutory anchor. For example, there is no physical difference between a CD containing software and a CD containing music, but one is a statutory “article of manufacture” and the other isn’t. Or maybe it’s that one has “functional” material (but then, a PostScript file is really a computer program in a special programming language) and one isn’t. But ignoring “non-functional” material doesn’t square with the 103 requirement to consider the claim “as a whole.” In contrast, Congress, if it wanted to do something worthwhile for patents, could draw such a line because it isn’t constrained by statutory language or past decisions.

Comments are closed.