Hi, my name is Ben Klemens, and I’m honored to say that I’ve been invited to write a bit here on Bureauphile. My background is mostly in Game Theory, Statistics, computational modeling, and other sundry methodological pursuits. But I have read far too much on patent law, so my first few posts will likely be on that subject.
Obvious patents almost seem to be the norm these days, because of all of the Bureauphile’s favorite problems, including regulatory capture and a budget-maximizing bureaucracy.
The parade of idiotic patents has been a common trope since Amazon’s one-click patent in the `00s. The slide-to-unlock feature on the iPhone, Google’s patent on doodles, or Friendster’s patent on circles of friends. I mean, if I stated to you this problem—
I have a smartphone with a screen and only one button. I need a way to keep it from turning on in users’ pockets.
—how long would it take before it occurs to you to require a gesture on the screen to unlock the telephone? As for making it work, here’s an implementation in about 450 lines of code—get it before the author gets sued.
As a starting point, 35 USC § 102 is pretty clear that you can’t patent an obvious invention:
A patent may not be obtained […] if the differences between the subject matter sought to be patented and the prior art are such that the subject matter as a whole would have been obvious at the time the invention was made to a person having ordinary skill in the art to which said subject matter pertains.
But the measure of “differences between” has no objective yardstick. Pressure to consider even the smallest difference to be patent-worthy comes from all directions.
▶ Patents are typically denied on first instance (I’ve been told that patent examiners are instructed to do so), and the applicant is expected to submit a revision. Revisions and replies can be made to any but a final denial. In the case of a final denial from an examiner, the denial can be appealed to the BPAI (Board of Patent Appeals and Interferences), an administrative court within the Patent office. If that’s denied, you can keep appealing the denial through the judiciary all the way up to the Supreme Court.
Conversely, if a patent is granted, that’s the end of the story. It is currently illegal for a third party to sue to have a patent invalidated (due to lack of standing), and the Patent Office and the grantee sure aren’t going to do it. [There is a recently-established post-grant review process, but it is, um, not yet a success.]
You can see how the asymmetry, where grants aren’t litigated and denials get fought all the way to the Supreme Court, will push against rulings of obviousness.
There are other ways to rule something unpatentable: if an invention is not an obvious variant of the prior art, but is actually in the prior art, then there is minimal judgment required, and no point trying to get a new judge to see differently. Meanwhile, if an examiner finds something to not be in the literature but too close to it, maybe the next judge will see more daylight between the prior lit and the new claims.
▶ In fact, I think that the ethos in the patent world is that any furthering of the literature, no matter how small, is a novel thing. After all, everything is obvious after the fact, but if it’s not in the literature then that means that it did not occur to all the people who came before. Geniuses from Plato to Kant to Einstein (a patent examiner!) did not invent the slide-to-unlock feature.
Patent lawyers are generally people who chose to enter the field because they think inventors are awesome, and as I understand it, if we totaled up all the billable hours of all the patent lawyers of the world, most of them would go toward advocating to the USPTO for the grant of a client’s patent. So the patent lawyers of the world have a concept of obviousness that is tuned to believe that even the tiniest advance by a client is worthy of a patent. This clearly means pressure to validate obvious patents from private lawyers, but it’ll turn into something else when some of those lawyers get appointed to judgeships a few paragraphs from now.
▶ The Patent Office is supposed to be an adversary to the applicants, but the model of the budget-maximizing bureaucracy applies pretty darn well, and budget-maximizing means using the smiley-face stamp on more patents. In 1997, 124,000 patents were granted; in 2011, 248,000 patents were granted. I have argued (in book form) that courts and the USPTO have worked tirelessly to expand the range of what is patentable and eliminate key barriers to patent granting.
The Obama administration appointed Dave Kappos to head the USPTO. Before he made day-to-day decisions about how examiners should accept or reject patents, he was the vice president for intellectual property at IBM (which brags about being the largest user of the patent system). So if you’re wondering why all these software patents suddenly started appearing, you can largely thank the Obama administration.
[A digression, mostly for humor value. About a month after Kappos was appointed, I corresponded with a Department of Commerce ethics officer. He wrote:
Government-wide Standards of Conduct bar a Federal employee from participating in a matter in his Government job if a party to that matter is an organization with whom he has a “covered relationship” if doing so would cause an appearance of a loss of impartiality. An employee has a covered relationship with an organization with which he has a business relationship, and with a former, non-Federal employer for one year after termination of an employment relationship.
I tried to get him to explain how Kappos could act as the head of the USPTO when his every decision would materially affect IBM, but the Commerce ethics officer chose to not discuss the matter further.]
From an amusing and/or depressing NYT article: ” ‘It’s called the patent office,’ [Kappos] said, noting that issuing patents is the agency’s job.”
▶ What about the courts? Here, we have seen a battle between specialized courts and generalists. We can blame Reagan for this one, because he set up the Court of Appeals for the Federal Circuit, intending it to be the highest court for patent law (among other duties). A subset of CAFC judges hear nothing but patent cases. Of course, it’s only natural that the person you’d appoint for such a position would be a patent attorney.
Naturally, the CAFC has made ruling after ruling expanding the scope of patentability and narrowing concepts like obviousness.
From the founding of the CAFC in 1982 to about 2005, the Supreme Court heard no patent law cases of any serious import, allowing the CAFC to be the highest court of the patent land, as per its intent. Then they seem to have gotten fed up. From 2006 to present, the court has heard cases in patent law at a steady clip. At first I thought it was just going to be three patent cases, but they keep coming year after year, and almost every one of them is either a unanimous or 8-1 overturning of the CAFC’s ruling.
It is a serious smackdown, and the CAFC is having trouble adjusting. It’s not just me who thinks so: one of the CAFC judges, Judge Prost, wrote a dissenting opinion pointing out the failure to adjust. I (heart) reading pissed off judges; here are my favorite parts:
The majority has failed to follow the Supreme Court’s instructions—not just in its holding, but more importantly in its approach. […] The majority does not even attempt to inquire whether the claims disclose anything inventive. […] The majority has resurrected the very approach to [patentable subject matter] that … the Supreme Court laid to rest. I cannot agree.
So where is the invention? […] The claim in effect presents an abstract idea and then says “apply it.” That is not enough.
What legal principle justifies responding to a unanimous Supreme Court decision against patentability with even a stricter subject matter standard? I do not know, and I cannot find the answer in the majority opinion.
I respectfully dissent.
Getting back to just the obviousness standard, the Supreme Court made an effort in KSR v Teleflex (2007) to reign in this obviousness thing. The one-sentence version of the ruling is that if it is “obvious to try” something, then it is too obvious to be patented. If an electronic brake pedal is in the patent literature, and an adjustable-height brake pedal is in the patent literature, then it is obvious to try an adjustable-height electronic brake pedal. After this ruling, there was a spate of patent overturnings in the lower courts using the “obvious to try” verbiage provided by the Supremes. Nonetheless, the USPTO and the courts continue to be OK with Amazon’s one-click patent, Apple’s slide-to-unlock patent, and Google’s doodle patent.
The amazing thing about the KSR ruling is that it had to happen: the CAFC thought that the electronic + adjustable pedal was a nonobvious extension of the already-extant electronic pedal and the adjustable pedal. That’s a real rift in worldview between the generalists and the specialists.
So there you have a few reasons pushing for the validity of obvious and otherwise lousy patents from the private sector and from the Executive and Judicial branches (the Legislative checked out of the discussion long ago). Maybe you can offer more in the comments. Next time, I’ll continue with the discussion of the specialists versus the generalists, in a post that will use the term “natural law”.