I don’t have a great deal to add to coverage of last week’s big patent story, which concerned the filing of a complaint by Microsoft co-founder Paul Allen against major technology companies including Apple, Google, Facebook and Yahoo. Diane Searcey of The Wall Street Journal, Tom Krazit at CNET News.com, and Mike Masnick on Techdirt pretty much lay out as much as is known so far.
But given the notoriety of the case and the scope of its claims (the Journal, or at least its headline writer, has declared an all-out “patent war”), it seems like a good opportunity to dispel some common myths about the patent system and its discontents.
And then I want to offer one completely unfounded theory about what is really going on that no one yet has suggested. Which is: Paul Allen is out to become the greatest champion that patent reform will ever know.
The Interval Patents
First, a few facts. In the mid 1990’s, Allen co-founded Interval Research, an effort to replicate Xerox’s Palo Alto Research Center (PARC) without the company politics that made it impossible for any of the brilliant inventions developed there to achieve commercial success—at least not by Xerox, in any case. (The other co-founder, David Liddle, was a former PARC scientist.)
Paul Allen intended Interval to do research the “right” way, and was quoted shortly before Interval was closed down as saying his goal was to produce world class research and development (R&D), with an emphasis on “less R and more D.” The company was founded in 1992, and shut down in 2000.
Interval, it turned out, produced a good deal of R and very little D. And now the R is generating an alternative and increasingly popular kind of D, taking the form of new patent litigation. Here, the litigation involves some pretty basic aspects of digital life.
One of the patents at issue, for example, is U.S. Pat. No. 6,263,507, which describes the claimed invention as a “Browser for use in Navigating a Body of Information.” Originally filed in 1996, the application was granted in 2001.
As the title suggests, the claimed invention is broadly defined, and the details implicate many aspects of core technologies of search, browsing, graphical user interface—pretty much everything. Below is a diagram showing a news browser from the application, which speaks for itself.
The Patent System, De Jure and De Facto
As I’ve been at pains to point out in prior posts on patents, the system in its current state of dysfunction encourages broad—perhaps ridiculously broad—applications. Worse, the lack of relevant expertise and the pressures of productivity on patent examiners results in many bad patents (perhaps many many) being granted.
The result has been the creation of a shadow patent examination process through litigation. The grant of a patent is no longer the final step, in other words. The de facto examination really takes place when the holder tries to enforce the patent against an alleged infringer, and the defendant claims invalidity of the patent as a defense. When such cases go to trial, which they rarely do, a jury of laymen are then tasked with doing the work avoided by the patent examiner.
In effect, the patent office has outsourced its job to the judiciary and in particular to a jury of non-experts. If nothing else, that is a feature of the modern system that absolutely no one is happy with, or in any event that no one can justify.
It also needs to be emphasized that patent infringement (as opposed to copyright infringement), need not and indeed rarely does include any suggestion of “theft” or other hint of immoral conduct. Most patent infringers do not copy the work of another inventor—they create their own innovation independently, often completely unaware of the existence of the relevant patents or pending applications. The broader the patents that are granted, of course, the more likely coincidental or seemingly “innocent” infringements are to occur. From a legal standpoint, however, ignorance of existing patents is no defense.
In that important respect, the difference between patent and copyright is significant. A copyright is a monopoly only on the particular expression of an idea. Margaret Mitchell gets a copyright for the uniquely creative elements of “Gone with the Wind,” not for the general idea of telling the story of 19th century Southern life through the device of a particular woman and her plantation.
And if by some miracle a second author wrote a nearly identical work (the theoretical room full of monkeys and typewriters, e.g.) and could prove no awareness with Mitchell’s novel, there would be no infringement.
A patent, on the other hand, is a monopoly on an idea, and is protected regardless of how someone else arrives at the same idea.
That difference explains why the requirements for a patent are so much more burdensome than for copyright, and why a patent lasts a relatively short amount of time (unlike copyright, Congress has generally resisted industry calls for greater and longer protection). Because a patent protects an idea however arrived at, its potential constraint on future invention is great. So there are important limits baked into system to keep all but the most novel inventions out of its protection.
First, patents last only 20 years from the date of filing (copyright today lasts roughly 100 years). So in this case Patent No. 6,263,507 will expire in 2016 (though later-claimed enhancements will last longer). After 2016, anyone can forever after duplicate the invention without paying any royalty to Interval.
Indeed one of the requirements of receiving a patent is that the claimant must provide sufficient detail in the application to provide a usable specification for the public to make free use of once the patent expires. So not only does the idea become part of the public domain, the inventor must a priori assist the public in making full use of it.
Other important limits require that valid patents only apply for inventions that are truly novel and not mere enhancements to existing inventions already in the public domain. The invention must be “non-obvious,” and must be clearly distinguished from earlier inventions (“prior art”).
What are Patents Good For, and For Whom?
Unfortunately, as noted above, the application of these and other limits are increasingly being left to the byzantine maneuverings of strategic litigation, negotiations, patent pools, defensive patents, patent trolls and, in extreme cases, judge and jury.
That was surely not the intent of the creators of the system. But given how wasteful and dangerous patents now seem, it’s worth remembering why we have them in the first place. In principle, the goal of the patent system is to encourage investment in new innovations that will contribute (perhaps greatly) to the overall social and economic good of a nation.
How does society benefit from giving a monopoly to an individual inventor? The answer has to do with perceived if not real incentives. Inventors, individually or in large corporations, may toil for years without generating any revenue from their efforts—it could be all R and no D, in other words, and the D may only come if at all at the end of a long and circuitous route.
Once—if–a marketable product or service results from all that research, however, inventors may find themselves unable to recover their investment. A potential competitor need not repeat the research, but may simply buy the resulting product, reverse engineer it, and start producing its own equivalent offering, perhaps immediately.
Having made no investment, the new competitors have no research investment to recover, and so can offer identical products at a much lower price. The inventor must pro rate the cost of research into the price of the product or service; the competitor does not. Without patent, the inventor could be quickly forced out of the market she invented.
Beyond questions of “fairness” in such a result, the nation as a whole would lose out, even though for the particular product or service we would get a cheaper price, or get a cheaper price sooner. Because, in the long term, without some reasonable time period to recover research costs without fear of undercutting competitors, only the most idealistic inventors would continue to invent.
Fewer inventions mean slower progress for civilization overall. Governments might have to fund research directly, with no hope of having the expertise to evaluate good from bad proposals.
Hence, the powerful but limited monopoly protection of patents—a necessary evil, or so it has been understood.
Nitpicking the Definition of “Patent Troll”
That, at least, is the theory. But the practice has mutated into something very different. The generosity of the Patent Office and the high cost of litigation have created new opportunities in the last quarter-century or so to game the system.
The most obnoxious variation is companies who do no research or development of any kind, but who simply buy up large blocks of patents from desperate or bankrupt patent holders. They then threaten to assert these patents against companies actually producing products, hoping not to stop their efforts but to extract a royalty from then in the form of a license for a potentially infringed patent.
These companies are referred to as patent trolls, perhaps in reference to an old Norwegian folk tale of the “Three Billy Goats Gruff.” In the story, three goats must outsmart a troll who lives under a bridge they must cross to find food. The troll lies in wait and eats all who try to use the bridge. So the image of the “patent” troll suggests an entity that tries to extract a toll for the use of something they didn’t build but have simply staked a claim to on the basis of superior strength and fortunate location.
So I take some issue with Mike Masnick and Mark Lemley’s characterizations of Paul Allen as a newly-revealed patent troll. It is true that Interval never developed any products from its patented inventions and, one presumes, won’t in the future.
But unlike a “classic” patent troll (to quote Mark Lemley), Allen did invest substantial amounts of money in research–the kind of research the patent system is intended to protect and, if necessary, make enforceable through litigation.
That is not to say that I approve of the lawsuit, or even that I understand why Allen has brought it. I’ve made only a cursory review of the claimed inventions, and while certainly broad and seemingly the subject of earlier invention, I can’t say with any confidence that they ought not to have been issued and therefore would ultimately be found invalid at a hypothetical trial.
(I should also say I’m not even sure we need a patent system, or rather that we need the one we now have, whose costs have long ago greatly outpaced its social benefits. That’s a different discussion and, given Congress’s wilting response to patent reform for the last decade, mostly an academic one.)
Paul Allen’s Misdirection?
But if the patents are valid and if the defendants have infringed them (very big and complicated ifs, mind you), there’s nothing “disgusting” (to quote Mike Masnick) about a decision by Allen to enforce them. Paul Allen is presumably not in dire need of recovering the estimated $100 million he invested in Interval, but if the system is to work, the viability of enforcement must remain. And given the high cost today of litigating these claims, it may only be people with deep pockets who are still able to keep up the necessary threat of infringement claims.
A real determination of the merits of the patents and the claimed infringements would require the development of a voluminous record. Which is to say that neither Allen nor the defendants ought to feel especially confident of their position at this opening stage of litigation. Indeed, the complaint itself is entirely pro forma, making only the most basic allegations necessary to initiate a patent infringement suit. No suggestion of intentional or willful infringement is alleged.
Without knowing for certain, it’s safe to say that the purpose of the litigation is not to stop any of the defendants from offering the products and services claimed to infringe on Interval patents but, for better or worse or most likely both, to extract royalties and other licensing terms from them on behalf of Interval.
So the mystery isn’t so much whether the patents are valid or whether the defendants, knowingly or otherwise, infringed them. The mystery is why Allen is suing, and why he has waited so long to do so.
There’s no penalty to waiting, of course, so long as the term of the patents hasn’t expired. Rather, the longer a patent holder waits the more time the infringer can establish its products and, therefore, the more royalties available for the patent holder to extract. On the other hand, if you wait too long, the infringer’s products may fail and the infringer may go broke.
That explanation for why Interval has dropped this bomb when it has may be that simple. But something tells me there’s something else going on. Patent litigation is an elaborate chess game, and it feels like this is a move deep inside a very long-running duel. There’s a great deal that’s broken about the patent system. I’m just not sure yet whether this lawsuit is Exhibit A.
What other answer is possible? Here’s at least one crazy possibility (there are crazier ones, but this one at least is plausible). Maybe Allen is not the world’s most famous patent troll. Maybe he’s out to become the world’s most famous patent reformer. Maybe he doesn’t want so much to win as to publicize how dangerous his patents are.
Perhaps in asserting these patents, with their potential to unsettle so much of what is taken as settled business practices in the digital economy, he hopes to force leading tech companies and Congress to acknowledge that the system is broken and fix it. If he wins, or even if he just wears down the other side, perhaps he’ll demand not financial tribute but actual reform of a system that gives patent holders like him the power to disrupt digital life.
If so, it’s a dangerous gambit. On the other hand, it’s hard to see how the patent system could get much worse than it already is.
There’s a great deal that’s broken about the patent system. I’m just not sure yet whether this lawsuit is Exhibit A.