Drew Curtis vs. A Patent Troll
April 17, 2012 1:06 PM Subscribe
TED Talk:"Drew Curtis: How I Beat A Patent Troll" "Patent trolls have done more damage to the United States economy than any domestic or foreign terrorist organization in history every year."
That talk wasn't too bad. But, my job would be a lot easier if the patent office granted patents as readily as so many people think it does. Sigh.
posted by exogenous at 1:19 PM on April 17, 2012 [1 favorite]
posted by exogenous at 1:19 PM on April 17, 2012 [1 favorite]
Previously: Drew Curtis' FARK.com has settled a lawsuit with a patent troll. (2011, not a dupe)
posted by filthy light thief at 1:24 PM on April 17, 2012 [2 favorites]
posted by filthy light thief at 1:24 PM on April 17, 2012 [2 favorites]
I'm so glad that people are fighting back against patent trolls (and winning!). I recall watching an episode of Dragon's Den a year or two ago and one of the dragon's chewing out an so-called entrepreneur because: "you're not an entrepreneur, you're a patent troll." It would have been fine if he'd left it at that, but then he continued: "You're a patent troll, but I can work with that business model. Let's make a deal."
posted by asnider at 1:42 PM on April 17, 2012 [1 favorite]
posted by asnider at 1:42 PM on April 17, 2012 [1 favorite]
I'm so glad that people are fighting back against patent trolls (and winning!)
With regard to summary judgment, non-practicing entities win about 24% of the time, and practicing entities win about 34% (source [pdf]). Further, about 60% of all cases brought by a non-practicing entity are concluded by summary judgment. So in about 45% of NPE cases the result is a loss for the NPE at summary judgment. That's a big savings over going to trial, which happens in less than 3% of patent cases.
With regard to Fark's industry (software), the overall NPE success rate is particularly bad, a mere 16%.
Obviously many cases settle before summary judgment, so there's a lot of selection bias in those statistics, but the overall point is that patent trolls are routinely—and successfully—fought against, particularly in the software industry.
The policy takeaway from those statistics, by the way, is that courts need to make heavier use of fee-shifting in NPE cases. If NPEs are frequently bringing questionable cases (and a high loss rate at summary judgment tends to indicate that), then we should push back on that. The most direct way is to make an NPE think twice about bringing a questionable case by substantially increasing the downside of a loss in the form of paying the winner's attorney's fees. Best of all, this is something that the courts are already empowered to do in "exceptional cases," and fee awards are reviewed for abuse of discretion, so they are not commonly overturned on appeal.
posted by jedicus at 2:04 PM on April 17, 2012 [9 favorites]
With regard to summary judgment, non-practicing entities win about 24% of the time, and practicing entities win about 34% (source [pdf]). Further, about 60% of all cases brought by a non-practicing entity are concluded by summary judgment. So in about 45% of NPE cases the result is a loss for the NPE at summary judgment. That's a big savings over going to trial, which happens in less than 3% of patent cases.
With regard to Fark's industry (software), the overall NPE success rate is particularly bad, a mere 16%.
Obviously many cases settle before summary judgment, so there's a lot of selection bias in those statistics, but the overall point is that patent trolls are routinely—and successfully—fought against, particularly in the software industry.
The policy takeaway from those statistics, by the way, is that courts need to make heavier use of fee-shifting in NPE cases. If NPEs are frequently bringing questionable cases (and a high loss rate at summary judgment tends to indicate that), then we should push back on that. The most direct way is to make an NPE think twice about bringing a questionable case by substantially increasing the downside of a loss in the form of paying the winner's attorney's fees. Best of all, this is something that the courts are already empowered to do in "exceptional cases," and fee awards are reviewed for abuse of discretion, so they are not commonly overturned on appeal.
posted by jedicus at 2:04 PM on April 17, 2012 [9 favorites]
@exogenous
I bet you're trying to create something genuinely new aren't you? And I bet you spend more than half your money on development rather than lawyers, am I right?
You are going to have to decide. Are you in the business of creativity, or are you in the business of patents? If you are a small or medium business then you can't be in both.
posted by EnterTheStory at 2:17 PM on April 17, 2012
I bet you're trying to create something genuinely new aren't you? And I bet you spend more than half your money on development rather than lawyers, am I right?
You are going to have to decide. Are you in the business of creativity, or are you in the business of patents? If you are a small or medium business then you can't be in both.
posted by EnterTheStory at 2:17 PM on April 17, 2012
Yeah, I reckon the relatively favorable settlement for Fark was motivated by the possibility of the plaintiff perhaps having to pay Fark's attorney fees (the fee-shifting that jedicus mentions) and/or possible Rule 11 sanctions against the attorney who filed the suit.
posted by exogenous at 2:22 PM on April 17, 2012 [1 favorite]
posted by exogenous at 2:22 PM on April 17, 2012 [1 favorite]
Holy crap, asnider! Here's the clip you were talking about (in the US, it's called Shark Tank, and has many of the same investors). In this case, it appears to be more trademark-trolling, but it is still pretty despicable. This guy even looks like some sort of comic book sleazebag!
posted by mysterpigg at 2:23 PM on April 17, 2012 [1 favorite]
posted by mysterpigg at 2:23 PM on April 17, 2012 [1 favorite]
In related news: Twitter announces the "Innovator's Patent Agreement," a contract under which patents filed by engineers working there can only be used "defensively" rather than "offensively." Some coverage in the NYTimes.
posted by whir at 2:46 PM on April 17, 2012 [1 favorite]
posted by whir at 2:46 PM on April 17, 2012 [1 favorite]
Kneejerk reaction to the above: “Defensively” means “we sue other people” and “offensively” means “other people sue us.”
And upon reading it, that's pretty much right. The exceptions to "we will never file an IP lawsuit" are big enough to drive Apple's portfolio through.
posted by Holy Zarquon's Singing Fish at 3:00 PM on April 17, 2012
And upon reading it, that's pretty much right. The exceptions to "we will never file an IP lawsuit" are big enough to drive Apple's portfolio through.
posted by Holy Zarquon's Singing Fish at 3:00 PM on April 17, 2012
Responding to some of Curtis's specific points:
He mentions patenting an existing idea as applied to an emerging industry. The courts have pretty well caught on to this and are coming down consistently against that kind of patent, as I've mentioned on MeFi a couple of times.
He mentions the burden of proof. I'm in favor of eliminating or reducing the presumption of validity. At this point it's kind of a joke to pretend that the Patent Office does an effective of examining every patent that comes through.
He says the best case scenario for the defendant is spending $2 million and 18 months. That depends on the defense approach. If the defense is invalidity due to certain kinds of prior art, then ex parte reexamination is much cheaper, requires little input from the defendant, and typically results in a stay of any co-pending litigation. However, in this case the defense seemed to be noninfringement, which does mean going to court.
He talks about fighting the infringement rather than the patent (i.e. alleging invalidity or unenforceability). It's very strange to hear someone recommend a blanket approach to patent infringement like that. It's a commonly observed truism in patent law that "a weak patent is strong and a strong patent is weak." That is, a patent that is very broad—and thus strong on offense—is easy to find prior art against—and thus weak on defense, and the reverse is also true.
He confuses the firm with its client when he discusses contingent fees, and he seems to assume that all patent trolls hire attorneys on a contingent fee basis. Or maybe it's that inventors license their patents out to patent troll companies on a contingent fee basis. In either case, that's far from universally true, and anyway many contingent fee contracts take into account case complexity (e.g. x% for a pre-trial settlement, y% if it goes to trial, z% if there's an appeal, and certain costs like expert fees, filing fees, etc are all taken off the top prior to the percentage calculation). Threatening to pull out all the stops in a case may scare off poorly funded clients or inexperienced firms, but it's hardly the universally effective strategy he makes it out to be.
The bit about terrorists is actually wrong on its own terms. The $500 billion figure is from a study by Bessen, Meurer, and Ford, and it calculated the cost from 1990-2010. The annual figure for the last four years was about $80 billion, significantly less than the cost Curtis gives for 9/11 ($123 billion).
And of course, that's presuming you actually buy Bessen et al's event study methodology in the first place. See Glynn S. Lunney, Jr., On the Continuing Misuse of Event Studies: The Example of Bessen and Meurer, 16 J. Intell. Prop. L. 35 (2008) (responding to similar event studies done by Bessen and Meurer when they wrote their book Patent Failure).
posted by jedicus at 3:16 PM on April 17, 2012 [1 favorite]
He mentions patenting an existing idea as applied to an emerging industry. The courts have pretty well caught on to this and are coming down consistently against that kind of patent, as I've mentioned on MeFi a couple of times.
He mentions the burden of proof. I'm in favor of eliminating or reducing the presumption of validity. At this point it's kind of a joke to pretend that the Patent Office does an effective of examining every patent that comes through.
He says the best case scenario for the defendant is spending $2 million and 18 months. That depends on the defense approach. If the defense is invalidity due to certain kinds of prior art, then ex parte reexamination is much cheaper, requires little input from the defendant, and typically results in a stay of any co-pending litigation. However, in this case the defense seemed to be noninfringement, which does mean going to court.
He talks about fighting the infringement rather than the patent (i.e. alleging invalidity or unenforceability). It's very strange to hear someone recommend a blanket approach to patent infringement like that. It's a commonly observed truism in patent law that "a weak patent is strong and a strong patent is weak." That is, a patent that is very broad—and thus strong on offense—is easy to find prior art against—and thus weak on defense, and the reverse is also true.
He confuses the firm with its client when he discusses contingent fees, and he seems to assume that all patent trolls hire attorneys on a contingent fee basis. Or maybe it's that inventors license their patents out to patent troll companies on a contingent fee basis. In either case, that's far from universally true, and anyway many contingent fee contracts take into account case complexity (e.g. x% for a pre-trial settlement, y% if it goes to trial, z% if there's an appeal, and certain costs like expert fees, filing fees, etc are all taken off the top prior to the percentage calculation). Threatening to pull out all the stops in a case may scare off poorly funded clients or inexperienced firms, but it's hardly the universally effective strategy he makes it out to be.
The bit about terrorists is actually wrong on its own terms. The $500 billion figure is from a study by Bessen, Meurer, and Ford, and it calculated the cost from 1990-2010. The annual figure for the last four years was about $80 billion, significantly less than the cost Curtis gives for 9/11 ($123 billion).
And of course, that's presuming you actually buy Bessen et al's event study methodology in the first place. See Glynn S. Lunney, Jr., On the Continuing Misuse of Event Studies: The Example of Bessen and Meurer, 16 J. Intell. Prop. L. 35 (2008) (responding to similar event studies done by Bessen and Meurer when they wrote their book Patent Failure).
posted by jedicus at 3:16 PM on April 17, 2012 [1 favorite]
In related news: Twitter announces the "Innovator's Patent Agreement," a contract under which patents filed by engineers working there can only be used "defensively" rather than "offensively."
This has worked out poorly in the past.
Good on for them for trying, but really without an extensive overhaul of the current set-up such efforts are doomed.
posted by Artw at 3:46 PM on April 17, 2012 [1 favorite]
This has worked out poorly in the past.
Good on for them for trying, but really without an extensive overhaul of the current set-up such efforts are doomed.
posted by Artw at 3:46 PM on April 17, 2012 [1 favorite]
This has worked out poorly in the past.
To be fair the Innovator's Patent Agreement gives the inventor a remedy in cases in which the patent is ultimately used offensively, whereas the Yahoo! employee in that story had only vague, nonbinding assurances. Specifically, the remedy is that the inventor has the right to license the invention to anyone that the company sues offensively in violation of the agreement, thus short-circuiting any attempt to do so.
The real problem is that the exceptions almost completely swallow the rule. The patent can be freely used
a) against any entity that sues or threatens to sue the company (fair enough, that's basically the definition of defensive)
b) against any entity that has sued or threatened to sue anyone else offensively in the past 10 years (so sue anyone offensively and it's open season on you for 10 years)
c) otherwise to deter the possibility of a suit against the company or just about anybody affiliated with it (preemptive strikes are okay as long as they're for deterrence)
At least the agreement runs with the patents and so would presumably survive a merger or sale.
posted by jedicus at 4:03 PM on April 17, 2012
To be fair the Innovator's Patent Agreement gives the inventor a remedy in cases in which the patent is ultimately used offensively, whereas the Yahoo! employee in that story had only vague, nonbinding assurances. Specifically, the remedy is that the inventor has the right to license the invention to anyone that the company sues offensively in violation of the agreement, thus short-circuiting any attempt to do so.
The real problem is that the exceptions almost completely swallow the rule. The patent can be freely used
a) against any entity that sues or threatens to sue the company (fair enough, that's basically the definition of defensive)
b) against any entity that has sued or threatened to sue anyone else offensively in the past 10 years (so sue anyone offensively and it's open season on you for 10 years)
c) otherwise to deter the possibility of a suit against the company or just about anybody affiliated with it (preemptive strikes are okay as long as they're for deterrence)
At least the agreement runs with the patents and so would presumably survive a merger or sale.
posted by jedicus at 4:03 PM on April 17, 2012
The Twitter IPA is specifically designed to prevent cases like the Yahoo one.
posted by nev at 4:04 PM on April 17, 2012
posted by nev at 4:04 PM on April 17, 2012
But in the Yahoo case the exact same assurances were given. I wouldn't trust it in the slightest.
posted by Artw at 4:13 PM on April 17, 2012
posted by Artw at 4:13 PM on April 17, 2012
But in the Yahoo case the exact same assurances were given. I wouldn't trust it in the slightest.
Nonbinding assurances, and the Yahoo! employee had no ability to license the patents to defendants under any circumstances. The IPA is both a binding contract and grants a (potentially) very effective remedy to the inventor.
posted by jedicus at 4:19 PM on April 17, 2012
Nonbinding assurances, and the Yahoo! employee had no ability to license the patents to defendants under any circumstances. The IPA is both a binding contract and grants a (potentially) very effective remedy to the inventor.
posted by jedicus at 4:19 PM on April 17, 2012
More lawyers would be involved in screwing you over, but you'd end up screwed over nonetheless. It's a nice gesture, but it's not a solution. The system needs revamping or scrapping.
posted by Artw at 4:22 PM on April 17, 2012
posted by Artw at 4:22 PM on April 17, 2012
To judge by the reciprocal language used in the IPA (e.g. the (b) exception applies to entities that sue any other entity other than for a defensive purpose) and the 'open source' approach to the agreement, I think Twitter hopes that it gets adopted by a lot of tech companies, eventually leading to a kind of detente, at least among the major players.
posted by jedicus at 4:35 PM on April 17, 2012
posted by jedicus at 4:35 PM on April 17, 2012
b) against any entity that has sued or threatened to sue anyone else offensively in the past 10 years (so sue anyone offensively and it's open season on you for 10 years)
This is actually kind of clever, as if it becomes widely adopted, then you have kind of a Warsaw/Nato situation where you have all these IPA companies who can sue in defense of each other.
posted by empath at 4:38 PM on April 17, 2012
This is actually kind of clever, as if it becomes widely adopted, then you have kind of a Warsaw/Nato situation where you have all these IPA companies who can sue in defense of each other.
posted by empath at 4:38 PM on April 17, 2012
I think that a bunch of tech companies agreeing not to sue each other over patents might be an anti trust violation, though, no?
posted by empath at 4:40 PM on April 17, 2012
posted by empath at 4:40 PM on April 17, 2012
Only about 180 software patent suits are filed in the United States each year.
I've worked at smaller companies that started to make good money. Suddenly boom, patent trolls left and right. No lawsuits, but a lot of money was wasted negotiating with lawyers and eventually paying the most likely to go to court, because going to court is a huge expense for a small company. So no, there's not that many court cases because the whole system is so fucked up it's too damn expensive to fight back against bullshit extortion.
posted by aspo at 7:02 PM on April 17, 2012 [1 favorite]
I've worked at smaller companies that started to make good money. Suddenly boom, patent trolls left and right. No lawsuits, but a lot of money was wasted negotiating with lawyers and eventually paying the most likely to go to court, because going to court is a huge expense for a small company. So no, there's not that many court cases because the whole system is so fucked up it's too damn expensive to fight back against bullshit extortion.
posted by aspo at 7:02 PM on April 17, 2012 [1 favorite]
Mod note: Don't call people out who aren't in the thread please. Do over? Thanks.
posted by jessamyn (staff) at 7:03 PM on April 17, 2012
posted by jessamyn (staff) at 7:03 PM on April 17, 2012
I think that a bunch of tech companies agreeing not to sue each other over patents might be an anti trust violation, though, no?
The interesting thing about this is there's no agreement or even communication between competitors. And according to the IPA, if the inventor agrees to an offensive suit (without further compensation or threat, so no bribing or cajoling them into it), then the suit can proceed. So offensive suits are still theoretically on the table, and the agreement is with the inventor, not the company's competitors.
posted by jedicus at 7:52 PM on April 17, 2012
The interesting thing about this is there's no agreement or even communication between competitors. And according to the IPA, if the inventor agrees to an offensive suit (without further compensation or threat, so no bribing or cajoling them into it), then the suit can proceed. So offensive suits are still theoretically on the table, and the agreement is with the inventor, not the company's competitors.
posted by jedicus at 7:52 PM on April 17, 2012
I've worked at smaller companies that started to make good money. Suddenly boom, patent trolls left and right. No lawsuits, but a lot of money was wasted negotiating with lawyers and eventually paying the most likely to go to court, because going to court is a huge expense for a small company. So no, there's not that many court cases because the whole system is so fucked up it's too damn expensive to fight back against bullshit extortion.This is the crux of the issue. All the arguments about how you can beat a software patent in court still require you to pay a shitload of money to patent lawyers. So it's completely useless in the most common case. The implication is that if software doesn't make enough money to hire patent lawyers to defend you, then the software shouldn't be written. Or at least there's zero economic value in you doing so from the perspective of a patent lawyer.
posted by delmoi at 8:49 PM on April 17, 2012
"I think that a bunch of tech companies agreeing not to sue each other over patents might be an anti trust violation, though, no?"
It has a long history though, and was once even mandated
posted by Blasdelb at 11:24 PM on April 17, 2012
It has a long history though, and was once even mandated
posted by Blasdelb at 11:24 PM on April 17, 2012
Incidently, one recent Supreme Court judgment might just pull the rug from under patent trolls.
posted by MartinWisse at 12:14 AM on April 18, 2012
posted by MartinWisse at 12:14 AM on April 18, 2012
That Twitter IPA isn't perfect, but it doesn't look too bad. Its main weakness is that since it's just an agreement between Twitter and the inventor there's nothing to stop Twitter just paying the inventor off to set the IPA aside (it kind of pretends that this isn't true, in the bit in paragraph 2 starting "If Assignee needs to assert any of the Patent claims...", but unless US contract law is very different to the kind I'm familiar with this wouldn't be enforceable by any third party).
It's not exactly a patent version of the GPL (which has its own problems of enforceability), but at least it's a start. Next step is to make the agreement enforceable by someone with a reliable interest in getting rid of software patents altogether.
posted by A Thousand Baited Hooks at 4:00 AM on April 18, 2012
It's not exactly a patent version of the GPL (which has its own problems of enforceability), but at least it's a start. Next step is to make the agreement enforceable by someone with a reliable interest in getting rid of software patents altogether.
posted by A Thousand Baited Hooks at 4:00 AM on April 18, 2012
It's not exactly a patent version of the GPL (which has its own problems of enforceability), but at least it's a start. Next step is to make the agreement enforceable by someone with a reliable interest in getting rid of software patents altogether.I posted in another thread that if enough software developers were willing to file patents on any ideas they came up pool them, and then go after any company that sues a member, it could prevent a lot of patent activity. But it wouldn't take care of "Non-practicing entities".
To do that, though you could take it one step further: Go after any company that pays off other patent trolls through settlements, or else open source their software. It would do a lot of damage to the closed source software industry in the short term, but in the long term, if it had a strong enough patent library it would put patent trolls out of business.
However, funding for such an organization would be a challenge. You would have to engage in patent trolling yourself to pay for all the lawyers.
posted by delmoi at 4:38 AM on April 18, 2012
All the arguments about how you can beat a software patent in court still require you to pay a shitload of money to patent lawyers. So it's completely useless in the most common case. The implication is that if software doesn't make enough money to hire patent lawyers to defend you, then the software shouldn't be written. Or at least there's zero economic value in you doing so from the perspective of a patent lawyer.
Hence my emphasis on greater use of fee-shifting. That substantially addresses the problem of litigation cost and greatly upsets the patent troll business model, which often relies on "cost of suit" nuisance settlements to bankroll other cases. What it doesn't do is throw the baby out with the bathwater.
I also happen to think that attorneys should generally make far, far less money, more on par with what they did 30 or 40 years ago. In 1967 the starting salary at a large firm in New York was $65k in 2010 dollars. Compare that with the current rate of $160k. As bad as the starting salary inflation has been, it's been even worse for partners. This despite the fact that the supply of lawyers has only increased and many of the costs have gone down due to computerization and automation (e.g. electronic research, case management, and discovery tools). Fixing that insanity would be beneficial in all areas of society, not just those directly affected by patents.
I really do not care about how patent policy affects patent attorneys. In an ideal world we would have well-defined, easily-understood, inarguably valid patent rights that were easy to obtain (i.e. without an attorney) and universally respected, so there was no need for litigation. In other words, property rights without transaction costs.
posted by jedicus at 8:31 AM on April 18, 2012
Hence my emphasis on greater use of fee-shifting. That substantially addresses the problem of litigation cost and greatly upsets the patent troll business model, which often relies on "cost of suit" nuisance settlements to bankroll other cases. What it doesn't do is throw the baby out with the bathwater.
I also happen to think that attorneys should generally make far, far less money, more on par with what they did 30 or 40 years ago. In 1967 the starting salary at a large firm in New York was $65k in 2010 dollars. Compare that with the current rate of $160k. As bad as the starting salary inflation has been, it's been even worse for partners. This despite the fact that the supply of lawyers has only increased and many of the costs have gone down due to computerization and automation (e.g. electronic research, case management, and discovery tools). Fixing that insanity would be beneficial in all areas of society, not just those directly affected by patents.
I really do not care about how patent policy affects patent attorneys. In an ideal world we would have well-defined, easily-understood, inarguably valid patent rights that were easy to obtain (i.e. without an attorney) and universally respected, so there was no need for litigation. In other words, property rights without transaction costs.
posted by jedicus at 8:31 AM on April 18, 2012
Compare that with the current rate of $160k.
OK, where do I have to apply for immigration?
posted by Skeptic at 8:44 AM on April 18, 2012
OK, where do I have to apply for immigration?
posted by Skeptic at 8:44 AM on April 18, 2012
Most attorneys do not start at large firms in New York and start at much lower salaries, although still good salaries to be sure. Compare with engineering salaries and the law salaries don't look that much better except for those few working in large firms in large cities.
posted by caddis at 9:20 AM on April 18, 2012
posted by caddis at 9:20 AM on April 18, 2012
OK, where do I have to apply for immigration?The key is "Large New York Firms". The average pay for lawyers has an Enormous bi-modal distribution. But most lawyers in 2007 made either around $40k or around $135k And I would guess that it's gotten worse for the bottom half of the hump since then.
posted by delmoi at 10:55 PM on April 19, 2012
« Older Bang, Bang, it's a free country | THE ONE TRUE WANKER OF THE DECADE Newer »
This thread has been archived and is closed to new comments
posted by crapmatic at 1:17 PM on April 17, 2012 [11 favorites]