Last week, there was a fascinating full day conference on ways to fix the patent system. There was a lot to digest, and I've been working on a much longer post about it, which I may not finish until next week. Someone from the USPTO was in attendance, and at one point stated that director David Kappos fully "heard" and understood the problems that people were complaining about, and that he fully intended to fix the system. Apparently someone forgot to tell Kappos. In a talk today, he completely blasted "complainers" telling them to "give it a rest already." His argument is that the America Invents Act -- the completely toothless patent reform that passed last year after being stripped of almost everything useful -- was just implemented, so we should just sit around and twiddle our thumbs while hundreds of billions of dollars are wasted due to bogus patents and disputes.

Except to Kappos, all that dead weight loss is a sign that the system is working. Why? I have no idea.

Indeed, Kappos suggested that the volume of patent litigation in the smartphone industry was a sign that the patent system was working as intended. "The explosion of litigation we are seeing is a reflection of how the patent system wires us for innovation," Kappos said. "It's natural and reasonable that innovators would seek to protect their breakthroughs using the patent system."

Note the giant and very questionable assumptions in the middle of that one: that it's "innovators" seeking to "protect" "breakthroughs." I'd argue that none of the three things in quotes is accurate. Quite frequently it's lawyers who haven't actually innovated at all looking to shakedown actual innovators for broadly worded patents that never should have been granted, and which are being interpreted to cover things they don't really have anything to do with. That's not innovation. It's extortion... backed up by the US government. It's a travesty.

Even worse, Kappos is still relying on the absolutely ridiculous "study" that the USPTO put out earlier this year, despite the fact that its methodology has been widely debunked for including grocery store baggers as "IP innovators." Sorry. And, if you look at what their actual report shows, it suggests that patent-intensive businesses aren't doing so well. Somehow he ignores that. Of course, perhaps that's why his office rejected a promised interview with me earlier this year, and could only defend the patent claims by arguing the most bizarre correlation argument in the world, that because Steve Jobs was innovative and had patents, therefore, patents worked.

Rather than address any of the real and well documented concerns with the patent system, Kappos apparently just decided to spin a fairy tale. He insists, as he's done in the past, that stronger patents automatically create more incentives, even as the evidence suggests that's not even close to true.

More recent research has estimated that litigation by patent trolls costs the economy at least $29 billion per year, and that figure may be as high as $83 billion.

Rather than engage in this empirical debate, or even acknowledge its existence, Kappos acted as though it was self-evident that stronger patents always create a larger incentive for innovation.

"To those commenting on the smartphone patent war with categorical statements that blame the so-called broken system on bad software patents, what I say is: get the facts. The facts don't support your position."

With all due respect, Mr. Kappos, we do have the facts. And they support the position of software patent critics.

It's sad that Kappos sends his lackeys to Silicon Valley to claim that he's "listening" and then spews such pure crap. The system is broken and either Kappos is lying or clueless. Neither reflects well on him.

Perhaps one of the most famous patent lawsuits -- which really highlighted the whole patent troll problem -- was the case of NTP vs. RIM, which ended with RIM paying over $600 million to settle the case, even as the USPTO was rejecting NTP's patents. Since then, RIM has been involved in a number of other patent lawsuits as well, including one that it lost last week against "Mformation Technologies." RIM has been told to pay up another $147.2 million in this case.

Of course, what some people forget is that RIM brought much of this on itself. Before NTP even came on the scene, it was RIM who started suing a bunch of other companies for patent infringement, based on its broad portfolio of patents around wireless email and mobile devices. On top of all that, RIM's business is collapsing. The company is fighting for relevancy as its latest operating system has been delayed -- and there's growing evidence that even once it comes out, no one's going to care about it.

RIM can try to put a nice spin on things, but it seems clear that the company is in serious trouble. Perhaps, next time, it will focus on improving its products more than getting caught up in the patent game. Yes, many of these more recent lawsuits came from it getting sued, but there's no doubt that RIM drew a lot of attention to itself early on with its own patent lawsuits against others.

We're told, repeatedly, by some patent system supporters that the whole point of the patent system is really about "disclosure." Of course, this is a myth for a variety of reasons. The biggest, of course, is that in many industries patents are both completely useless to learn anything from and are never used to learn how things are done. This is especially true in software, where you will never hear about anyone learning how to do anything from a patent. A few years back, the US Patent and Trademark Office (USPTO) started publishing nearly all patent applications 18 months after the application came in. This is a pretty common practice around the globe. If a patent system is about disclosure, this makes sense -- and in theory, allows for people to point out prior art or protest certain patents before they're issued. There is a current exception for patents deemed in the interest of national security to keep secret (such as patents on nuclear energy).

However, as Francisco George points us to the news that the USPTO is now considering also keeping "economically significant" patents secret too (pdf and embedded below). They're exploring this after being directed by Congress -- and it's not hard to see the lobbyists' fingerprints on the specific request:

By statute, patent applications are published no earlier than 18 months after the filing date, but it takes an average of about three years for a patent application to be processed. This period of time between publication and patent award provides worldwide access to the information included in those applications. In some circumstances, this information allows competitors to design around U.S. technologies and seize markets before the U.S. inventor is able to raise financing and secure a market.

As far as I can tell, this directive seems to be saying that because patent applications might do what the patent system is supposed to do -- help disclose ideas -- we should keep the applications secret. The problem, of course, is that pretty much every patent applicant is going to think their patent is economically significant. The reality is that pretty much no one knows if their patents are economically meaningful until years later when a product is actually on the market. This proposal seems to serve no purpose other than helping companies not to disclose ideas while still letting them get patents.

When the numbers came out showing that 2011 represented yet another record year for patents granted, it was such a non-surprise that I didn't even bother mentioning it. The number of patents granted just keeps going up. And yes, there were two small dips during the past decade, but they corresponded with the rare situations in which the Supreme Court finally took an interest in some element of patent law and pushed back on the Federal Circuit (the appeals court that handles all patent issues) and the USPTO. We've discussed at length in the past, the problems of having a single appeals court that solely focuses on patent issues, because you lose the diversity of opinions (made worse at times when some of the judges on the panel have been former patent attorneys -- or, most famously, when a judge at the court was the same former patent attorney who wrote the last major update to patent law...).

However, Steve sends over a fascinating Yale Law Journal review article by Jonathan Masur that notes this problem of the Federal Circuit can be explained structurally, in that the relationship between the PTO and the Federal Circuit combined with the fact that there's no "adversarial" party contesting a patent grant, means that the patent system is effectively rigged to only expand, even if that goes against the best interests of society:

Because of the manner in which patent cases make their way from the PTO to the Federal Circuit, the PTO has a decided institutional interest in granting more patents than it should. And because of this same interaction, the Federal Circuit is engaged in an unwitting expansion of the patentability rules. The key lies with the asymmetric nature of appeals from the PTO to the Federal Circuit. When the PTO denies a patent application, the aggrieved applicant may appeal to the Federal Circuit. When the PTO grants a patent, however, there is no losing party to appeal--the victorious applicant merely walks away with its patent. That patent is unlikely ever to see the inside of a courtroom, given how few infringement lawsuits are litigated. Like most administrative agencies, the PTO wishes to avoid appeals and especially reversals. In order for the Agency to accomplish this, it need only err on the side of granting excessive numbers of patents--even invalid patents--for which there is no appeal. This desire to avoid litigation is a source of the invalid patents now being issued by the PTO in vast numbers--the patent system’s first problem.

But, as he notes, that's not the only problem. There's also the fact that since the PTO will now only reject patents that are clearly outside the law, the Federal Circuit only has opportunities to expand the granting of patents, not to reign it in:

The second problem, the ongoing expansion of the rules governing what types of inventions may be patented, stems from the PTO’s proclivity to grant any plausible patent. Because of the PTO’s efforts, the patent applications that the Agency denies will predominantly concern inventions that are unpatentable under current law. When a disappointed patent applicant appeals such an application to the Federal Circuit, that court has two options. It can reject the patent under existing law, preserving the law as it stands, or it can grant the patent under a new, more expansive understanding of what is patentable. The circuit denies most of these applications. But when the Federal Circuit eventually decides a case in favor of an applicant, it creates a new precedent that enlarges the scope of what may be patented. The process then repeats itself, with the PTO denying more boundary-pushing patent applications and the Federal Circuit being presented with further opportunities to expand the limits of patentability. The result is a natural inflationary pressure on the law, generated entirely by the types of cases that the PTO sends to the Federal Circuit.

Is there a way to fix this? Masur suggests that awareness alone might create some pressure on the Federal Circuit to "create new, more constraining precedent." He also suggests that they might "vote strategically against self-interest in certain cases" such as by voting against a patent they think is a good patent "simply in order to forestall the law's outward momentum." That seems like significant wishful thinking -- which he admits with a bit of an understatement: "This would require a focus on issues with which circuit judges do not typically concern themselves, not to mention a surprising level of tactical shrewdness from a circuit that has not previously displayed any such inclination."

He also notes how Congress and the Supreme Court have the power to curb excesses, which the Supreme Court has done just a little bit (which, as noted above, explains the brief dips in patents). Separately, the executive branch could get involved and pressure the PTO to limit its approvals.

But, in the end, the key point that set off this problem is the lack of anyone to push back on a granted patent (until such time as they're sued, in which case a different series of issues are being dealt with). So of his key suggestions, one is to open up the ability for third parties to challenge patents -- even before they're issued. And to weigh against the inflationary pressure of the current Federal Circuit, he suggests that these challenges show up in the Federal Circuit too -- so that that court now also has incentive to curb excesses, rather than just inflate them. Definitely an interesting idea.

It's pretty difficult to find software engineers who take the patent system seriously. There are a few, but it's still pretty difficult. For the most part, they recognize that code is just a tool: you can make it do all sorts of things, given enough time and resources, but that doesn't mean that doing any particular thing in code is an "invention" that no one else should be able to do. And then, sometimes, they discover that something pretty basic and old has suddenly been given a patent. Brad Feld discusses his discovery that doubly linked lists were apparently patented in 2006 (patent number 7,028,023):

The prior art was extremely thin, only went back to 1995, and didn't mention that entire computer languages have been created around the list as a core data structure.  One of my first Pascal programming exercises in high school (in 1981 -- on an Apple II using USDC Pascal) was to write a series of operations on lists, including both linked and doubly-linked lists (I always thought it was funny they were called "doubly-linked" instead of "double-linked" lists.)  Anyone who ever graduated from MIT and took 6.001 learned to love all varieties of the linked list, including the doubly-linked one.  That was 1984 for me by the way.

Ironically, Wikipedia had great entries -- with source code no less -- about both linked lists and doubly-linked lists.  The linked list history goes back to 2001, well before the patent was filed.

Another day, another reason to question why software is patentable at all -- and to question who approves these kinds of patents.