Obama Calls for Patent Reform to Topple Trolls
eff.org
eff.org
Having companies that licence out technology is a perfectly reasonable system, if the company has done significant research or purchased patents from people who have. A secondary market for significant inventions is probably a good thing; a secondary market for borderline obvious inventions is not.
It's clear that the patent office can't tell an obvious invention apart from a non-obvious one, so I'm not sure if there is much to be gained in that direction either. I think there are a lot of collateral ways to attack the problem, though:
1) Reforms on how damages are calculated. If you neither sell a product nor license IP to third parties, there is no way you're losing any revenue from an infringement. You should not be awarded even "reasonable royalty" damages if you wouldn't have received any royalties in any case.
2) Reforms involving intent and copying. The economic heart of the patent system is unfair competition: I invent something and you copy it and undercut me in the market because you have no R&D costs. As you get further from this core justification, the less justifiable the regime becomes. A lot of troll litigation could be shut down quickly if patent infringement functioned similarly to copyright infringement in requiring direct copying. This, unlike the above reform, would be a fundamental change in the system, from "monopoly" to a more "unfair competition" grounds, but the fact of the matter is that independent invention is strongly probative of something that probably doesn't deserve patent protection, either for being obvious generally or for being obvious in the context of other advancements in technology.
Why? While this line of reasoning might carry some weight if Northwestern were a public university (though I'd still require some convincing), Northwestern is a private non-profit institution, with a highly ranked medical school and clinic.
Supplementing their income with drug sales instead of higher tuition seem like a perfectly logical thing to do.
If Northwestern did the research with its own money (which does sometimes happen, I think), then it is good that Pfizer had to pay Northwestern for the research, because this funds more research. The patent is just the legal mechanism that makes Pfizer have to pay as opposed to getting the research for free.
In some cases, the government funding issue is relevant. Not in all cases.
So threaten to sue a small company that doesn't want to pay for lawyers, and settle for a royalty agreement. Now you've licensed your IP to a third party, and you can sue someone else for greater damages.
Okay, here's my wild idea of the hour:
AIUI (IANAL, and I've never been involved in a patent application), a patent contains claims and specification. I'm not sure how much this is like C declaration vs definition, but in all the patents I've seen, the claims contain no implementation details. What if, as part of a patent application, the USPTO published the claims in a public forum, and allowed the public to put forward suggested implementations. If any are filed, obviously the patent is obvious. Interested amateurs could amuse themselves by contending, but also competing businesses could protect themselves from trivial patents from their competitors, by allocating paid employee time to monitor patent claims in areas of interest. Admittedly, future market entrants wouldn't yet know they should monitor the area, so it's not perfect.
As for your specific suggestions, rayiner, I think these are similarly not-quite-the-point as the focus on NPEs. I agree that #1 is a "collateral way to attack the problem", but then again so is focussing on NPEs. (After all, how many useful NPEs are there? If the cost of eliminating the parasitic NPEs is the loss of the legitimate drug patent holding company that you describe, I could live with that.)
And as for #2, it's not at all clear to me that it'd be an improvement. Companies would double down on requiring engineers to not read patents, as some companies already do to avoid triple damages. In my ideal world, we continue to stand on the shoulders of giants.
Interesting idea.
> After all, how many useful NPEs are there?
For all intents and purposes, the investment vehicle that funded Spaceship One is an NPE: http://en.wikipedia.org/wiki/Mojave_Aerospace_Ventures
> Companies would double down on requiring engineers to not read patents, as some companies already do to avoid triple damages. In my ideal world, we continue to stand on the shoulders of giants.
I think a lot of frustration with the patent system comes from independent inventions that nonetheless infringe on a patent. The obvious solution is to make independent invention a defense of some sort.
There is a trade-off, as you recognize. The less you can defend a patent against someone else developing the same technology, the less incentive you have to disclose anything that would help them do so. As an inventor, I don't want other people to "stand on my shoulders" unless they are required to compensate me for my contribution. You see this heavily in areas where patents are for whatever reason hard to use. E.g. Intel doesn't use patents on its process technology, it uses closely guarded trade secrets.
Perhaps the discovery would have been made regardless and the laws are set up for a few to make a buck to the detriment of society at large (playing devil's advocate here, not convinced this is the case).
However, universities are a piece of a larger system that includes big corporations in capital-intensive industries, federal/state government, and scientists/engineers themselves. Would universities keep doing research if they couldn't directly make money from licenses? Probably. Would the system keep working if the pharma/petrochem/aerospace/etc industries couldn't protect inventions? Those post-docs doing research at NWU aren't doing it for their love of humanity. They're there so they can get a job at Pfizer, Exxon, Lockheed-Martin, etc. Pfizer, Exxon, etc, certainly aren't in the game for their love of humanity. Would Pfizer keep hiring post-docs if some Chinese manufacturer could flood the market with cheap clone drugs within months of Pfizer bringing a drug to market? The government would keep investing, of course, but would they have to invest more to get the same output if they couldn't dangle the patent licensing carrot in front of researchers?
[citation needed]
A focus on NPEs also doesn't get at the abuse perpetrated by practicing entities. For example, Intel uses its stable of patents (many of them quite questionable) to prevent startups from breaking into its markets: you can only go up against Intel if you're someone like IBM who has their own patent stable to form a credible MAD threat. The aggressive use of patents by practicing entities with strong market positions (in order to stifle competition) is imo at least as big a problem as NPEs are.
What stops them just publishing a sham product whose sole purpose is to be for sale while infringing the patent and then show a few licenses to a couple of fishing buddies?
The real trouble with how damages are calculated for software patents is that there is no good way to determine the degree to which the patent was integral to the success of the product, and juries have had a tendency to grossly overestimate their value. Witness Apple v. Samsung.
I imagine your typical smart phone infringes probably twenty or thirty thousand software patents. Even if you assume that every penny of the device's price is attributable solely to patents (and nothing for, say, manufacturing, shipping or advertising), and the device "only" infringes 10,000 patents, the average value of one of those patents against the sale of ten million $500 devices would be $500,000. Which would hardly be even worth litigating over, given that the lawyers cost more than that. But there is no real way to prove that any given patent subscribes to the averages, and plaintiffs will naturally argue that their patent is worth more than all the other patents, and the jury who finds for them on infringement is predisposed to agree on damages as well. So we get billion dollar awards that should be sub-million dollar awards and the trolls get fully funded to buy more radiological weapons and lobby hard for their own continued existence.
>A lot of troll litigation could be shut down quickly if patent infringement functioned similarly to copyright infringement in requiring direct copying.
What standard could you use to prove that? Correct me if I'm wrong, but if I recall copyright is "similarity and access" and for published works "access" is kind of a formality (because everybody has access), which is why the degree of similarity requires a significant amount of specificity. I don't imagine you could find anyone to argue that Chrome infringes the copyright on Firefox as a result of their similarity.
But that's not going to cut it with patents. The intent of drafting a patent claim is to make it as broad as possible while still getting it through the patent office. Any non-mandatory specificity is the enemy. The only way you could give defendants any hope of actually proving that they didn't copy a patent under a standard like that would be to crank the level of specificity required in patent claims up to the point where you'll just have reinvented copyright and called it a patent.
Which is kind of the issue. Copyright works really well for software. The implementation is the hard part, and that is the part that copyright covers. Patents just don't work at all -- because there are so many ways to implement something in software that the only way a third party would ever actually infringe a software patent is to allow it to be so broad that it covers the problem itself rather than any specific solution to it (e.g. shopping cart patents), or covers any implementation that applies the mathematical formula that describes the optimized result for a specific problem dictated by information theory (e.g. encoding or encryption patents).
Whether design or implementation is the hard part depends entirely what kind of software you're building. Think of something like Google's self-driving car. I can't imagine that the hard part of that is just sitting down and writing the code.
That's kind of what I'm talking about. You can't get a patent claim on "a self-driving car" -- and if you do the patent office has failed miserably. Software patents don't cover the overall product, they cover the building blocks you use to make one. And each of the individual building blocks is not the hard part, the hard part is the overall implementation -- assembling them all into a cohesive whole that accomplishes the major goal you set out to achieve.
You want someone who starts with the Google self-driving car and sets out to copy it to have a hard time of it legally, so that they decide instead to do their own independent implementation and have their own costs (and maybe advance the field themselves) so that they can't just undercut Google having done no R&D, but at the same time still make it so that they can do an independent implementation without infringing. Those are the two important criteria.
Copyright pretty well does the first one. If you can't just copy Google's code out of their car and put it in yours, you have to write some tens of millions or more lines of code yourself. And then you have to test it and stamp out the bugs, which is by far the most expensive part of producing software in practice. (I expect this goes double when the thing you're making is a car.) Patents don't really add very much on top of that -- if someone is already doing an independent implementation, having to work around a patent is a small incremental cost, especially if the issued patent is not so broad that it covers the problem itself rather than a specific solution.
Which brings us back to the trouble: In order for the patent system to do anything for software R&D, you can't just have one patent, you need thousands. Which thoroughly violates the second criteria -- now no one can do an independent implementation without accidentally stepping on some fraction of the outlandish number of issued software patents. To clear yourself you have to clear tens or hundreds of thousands of issued patents, but to sue you the plaintiffs need only find a single patent for which you failed to do so. And to shake you down they don't even need to do that, as long as the cost of proving they're full of crap in court is more than the amount they're demanding for a license. These costs vastly exceed any possible incremental benefit over copyright of having software patents at all.
The courts are adept at investigating those issues. For example, Enron's sham electricity sales were discovered once the files were opened.
Well, maybe you're shopping for licensees.
A better solution would be to oblige you to license your inventions at a public auction if you can't find a licensee after a reasonable period of time (six months to a year?)
Licensee collusion at auction should be prosecuted under anti-competitive behavior acts, but probably wouldn't need to be if the auctions were fairly open.
The owner could participate with a shell company and put in an incredibly high bid to block the process... you could discourage this by putting a significant portion of the bid in escrow until the product is actually monetized. If it's never monetized, let other bidders sue for a new auction.
In fact, I would argue that trying to set in place a system to decide between trivial and nontrivial inventions is more or less a waste of time because there will always be inventions that sit on the fence.
What patent law should achieve is purposefully vague - it's the ultimate subversion of the legal system for capitalistic purposes. The state is supposed to give exclusive rights over "ideas", in effect granting the monopoly of anything to anyone. Then the patent office is handed over the hair ball of defining what are "ideas", which ideas are worth a patent (aka "trivial inventions", as if such thing exists), or even which ideas are original (a joke in itself).
In my mind, we're asking the patent office to solve a philosophical problem, in the hopes they will sort out a logical conundrum caused by a vague law.
In practice, however, that's not important. What important is the economical function the patent office has. Which is, to use granted monopolies on order to incentivice inventions. It's fine if they give us a faulty answer the the philosophical conundrum we've charged them with, as long as the faulty answer furthers this purpose. That's a more measurable stick we can use. And I think by that stick, they are currently doing a much worse job than they could.
http://archive.mises.org/18812/andrew-torrance-patents-and-t...
"""
Patent systems are often justified by an assumption that innovation will be spurred by the prospect of patent protection, leading to the accrual of greater societal benefits than would be possible under non-patent systems. However, little empirical evidence exists to support this assumption. One way to test the hypothesis that a patent system promotes innovation is experimentally to simulate the behavior of inventors and competitors under conditions approximating patent and non-patent systems. Employing a multi-user interactive simulation of patent and non-patent (commons and open source) systems (“The Patent Game”), this study compares rates of innovation, productivity, and societal utility. The Patent Game uses an abstracted and cumulative model of potential innovations, a database of potential innovations, an interactive interface that allows users to invent, make, and sell these innovations, and a network over which users may interact with one another to license, assign, infringe, and enforce patents. Initial data generated using The Patent Game suggest that a system combining patent and open source protection for inventions (that is, similar to modern patent systems) generates significantly lower rates of innovation (p<0.05), productivity (p<0.001), and societal utility (p<0.002) than does a commons system. These data also indicate that there is no statistical difference in innovation, productivity, or societal utility between a pure patent system and a system combining patent and open source protection.
"""
Look at the criminal justice system. You can't always prove guilt or innocence in many trials, many times there are cases that sit on the fence but our system is set up with the assumption of innocence.
PEs are encumbered by a number of other restrictions - their "cold war" type standoff against competitors, their need to maintain a mostly positive brand in the public eye, their need to minimize the risk of a catastrophic loss in a lawsuit. So they utilize only a fraction of the latent power of their patents.
NPEs have no such restrictions. They have no competitors, no brand to maintain, no economic risk beyond what it costs to register a new shell company.
So what we see from NPEs is what the patent system actually means. It's "pure patent system". By implication, the fix is to change the patent system. Preferably by abolishing software patents, at least. (I can't speak for other fields.)
In the history of computing, I think Babbage's difference engine, von Neuman and some of the stuff from Xerox Parc in the 1970s probably deserved a patent or two. Apart from those, no.
[1] Non-practicing entities, for those who don't know. (I had to look it up.)
But on the other hand, if a patented invention is something that several other folks in the industry would have invented within the next 5 years regardless then patents start to become a drag on progress. It's telling when you see a burst of progress happen after a patent runs out, and that sort of thing happens all the time. The question is how do we strike the right balance. Unfortunately right now there are some very strong forces intent on keeping things very much off balance.
The pharmaceutical industry also provides us with plenty of opportunities to pick on them. Take a drug that's about to go off-patent, make a trivial change to their process, get a fast-track approval from FDA (because it's essentially the same), and a new patent (because it's new and improved!). Competitors play this game too, making copycat drugs which are very similar enough for marketing, but just different enough to deserve their own patent. I'm having a hard time sympathizing with any pharmaceutical company. Disclaimer, I'm almost done reading Goldacre's new book.
What does this mean?
If A is patented, A+B is covered by A's patent AND B's patent.
The only way to avoid A is to use a non-patented yet critical invention. The inventor of A should have included any such thing as one of their patent claims.
This could create a self-funding process for patent regulation and monitoring, and a check against unhealthy patent proliferation. It would discourage patent trolls from sitting on patents, large companies from accumulating a "war chest" of patents, and encourage FRAND licensing.
There needs to be a recognition that in software entire segments are born and die within 20 years (even within 10 years). 20 years protection on a simple idea is way too long, even if it is truly novel. This is not like drugs where it may take 10 years of trials before you can even get it on the market.
I've been involved in quite a bit of hardcore algorithm R&D. In no case where material advancements were made was it possible to really bring it to market in five years from inception. Computer science has the same inertia for how things are done as any other industry. Altering the way things are done and getting uptake requires a surprising amount of time no matter how good the advancement. Which roughly matches most other patentable subject matters.
I'm not saying algorithm patents need to be 20 years but most software engineers have no idea how long it actually takes to commercialize material algorithm advancements. It looks a lot more like industrial patents than most programmers seem to assume. And many algorithm patents have millions of dollars in R&D invested in them; I have been involved in several such examples.
I think #4 (no liability for independent invention) would effectively eliminate software patents by itself. I like what's there even if it seems unlikely to ever be implemented.
As long as it starts with a presumption of independent invention until proven otherwise by the patent holder, yes.
Perhaps that should be true of all trivial inventions. Drug companies, for example, tweak best sellers coming off patent protection by adding another compatible compound and getting another 20 years, although the original compound looses protection.
Copyright covers everything important.
If I am going to build a system and try to sell it, I sell it as a service with add-on support and make it open source. If I am buying a software system I usually expect the same thing. Or hope that I can sell it for a small fee or close source it and add copyright protection for a totally unique work.
But generally I feel that if someone invents a software concept or set of concepts that works well enough, then everyone should adopt it. Patents are just going to make that impractical. The idea of everyone licensing whatever useful system gets invented isn't practical and doesn't work with open source.
Got a patent? Someone infringing? Sue them! Cost? Astronomical.
Selling something? Patent Trolls trying to make a grab? Defend yourself in court! Cost? Astronomical.
Those scenarios merely address outright abuse; but, the way patents are granted guarantees that those situations will also arise between parties who aren't trolls.
And it's not just a reform. All the issued patents must be revised and a lot of them must be revoked.
The solution seems to be simple, isn't it: don't patent trivial things, and only patent the actual methods, not just ideas of them.