Most Software Patent Trolls Lose Lawsuits 108
An anonymous reader writes "A new study is out concerning patent trolls and software patents, which found the rather surprising news that the most litigated patents tend to lose nearly 90% of the time. When broken down into different categories, patent trolls and software patents lose their lawsuits most often. While some may suggest this means 'the system is working,' that's not really true. The data suggests that most companies, when threatened with a lawsuit, end up settling or licensing to avoid the high costs of litigating. But the fact that so few software patents and patent trolls do well at trial may be more incentive to fight back. Either way, what does seem pretty obvious is that all those ridiculous patents you see in patent lawsuits are, in fact, bad patents."
Marshall, Texas (Score:3, Funny)
The checks and balances don't work for software (Score:1)
The problem is, in pharma or the car industry, patent threats get met by a legal team and a well-financed company, so patent holders don't launch as many spurious lawsuits. In software, where you don't need a legal team and a six digit bank balance, many developers get shafted because they can't use the court system. Patents shouldn't exist in domains where individuals and non-commercial entities can be mass producers.
Re:The checks and balances don't work for software (Score:5, Insightful)
No.
The underlying problems are twofold.
First, that a major number of patents are being granted that never should have been granted. Either because they are overworked, or because they are not correctly able to evaluate patents due to lack of training in the fields they are analyzing, or because they have been indoctrinated into a "just grant it the courts can sort it out later" (and a friend I know who works in the USPTO has been told that several times by his direct superiors over the years) mentality, the USPTO is granting things that never should have been granted.
Second, that the US court system is so fucked up and overburdened that most people who get hit with a troll lawsuit choose to settle, because fighting it is going to take years upon years, resources upon resources, and probably it's cheaper for them to just pay up. The legal system has ceased to be a venue where fair and equitable analysis of this sort of thing can take place, and instead is just a bludgeon for bullies with money and shyster lawyers willing to throw away the ethical codes to beat up on everyone else.
Re:The checks and balances don't work for software (Score:5, Insightful)
"Law" is simply disembodied violence, physical force transferred to the domain of the mind.
The notion of "fair and equitable analysis" is simply the system's own inflated self-image, like "all men are created equal" in 1776, or "freedom" in 2010.
I agree with your statements, btw. I just want to make sure that in moving to something new we don't repeat the mistakes of our past.
There's free as in speech, free as in beer, and free as in range. Americans are "free" in the final sense.
Re: (Score:3, Insightful)
The "law" has always been just another way for those who write the law to control those who are required to follow it. "Law" is simply disembodied violence, physical force transferred to the domain of the mind.
Intriguing rhetoric, but what alternative do you propose to Law? (And I'd call it more "passive aggressiveness" than "disembodied violence", but perhaps that's just a different face of the same coin. )
Re: (Score:2)
Not to be rhetorical, but in my country's history (and others), regardless of politics, guns seem to do well whether enforcing or abolishing "law".
Re: (Score:1)
Guns favor the guy with more guns and men. Guns certainly aren't the solution to organized crime. Having a police force means at least the most guns are in the hands of people who are at least somewhat accountable to the people.
Re: (Score:3, Interesting)
In other words, were a small group of people, no matter how gifted and altruistic, to create a framework for others to follow without participation and consent of those others during the framing itself, then eventually the society will come to reflect that inequity of that pro
Re: (Score:3, Insightful)
That sounds like a great bit of logic there, if you ignore various facts along the way.
Many of those founders owned slaves. Does American society today support slavery?
Women were considered unfit to enter politics. Are American women still kept out of political offices and denied the ability to vote?
Society's observed reflection of its laws is not the result of some poorly-understood natural process, but the intentional result of the society's trust in its governing body. Laws are merely standards for behav
Re: (Score:2)
To use the infamous car analogy, consider the laws requiring drivers to periodically stop and allow crossing traffic to move. If everyone follows the law, a few people get slowed down by a few seconds.
I routinely see four-minute red lights at an intersection whose metal detector fails to sense my bike. Can you fit this into your car analogy?
That's why American citizens (should) vote for representatives that they believe will support their views.
The major publishers of non-free fictional entertainment own the means that representatives use for communicating to the general public. Can you fit MPAA news [pineight.com] into your car analogy?
Want a bigger voice out of the 300 million Americans? Write your representatives!
Who write back with a form letter stating that an expansionist position on copyrights and patents is good for America.
Run for office!
Rep. Ron Paul (R-TX) tried this. His campaign got buried by the mainstr
Re: (Score:2)
I routinely see four-minute red lights at an intersection whose metal detector fails to sense my bike. Can you fit this into your car analogy?
Yes, actually. Occasionally, people get screwed by laws that don't account for them. The best solution is to propose an alternative that works for everybody. If you can invent a detector that works for both cars and bikes, yet still has the lifespan and approximate cost of the current detectors, go suggest it to the appropriate folks.
The major publishers of non-free fictional entertainment own the means that representatives use for communicating to the general public. Can you fit MPAA news [pineight.com] into your car analogy?
Yes, actually. Occasionally, there are greedy people who will do bad things that are not actually against the written rules, such as parallel parking in the exact middle of two
Re: (Score:2)
If you can invent a detector that works for both cars and bikes, yet still has the lifespan and approximate cost of the current detectors, go suggest it to the appropriate folks.
I managed to get one misbehaving intersection's sensitivity increased by calling 311 and reporting it. But the crack at the other, at the intersection of a low-traffic street and a major traffic artery, appears to be too long for such a quick fix. The long-term solution is to replace ring-shaped detectors with quadrupole detectors [humantransport.org]. This figure 8 shape concentrates the sensitivity in the middle of the lane, allowing traffic engineers to crank up the detector's sensitivity while still rejecting spurious signa
Re: (Score:3, Interesting)
If you look at the real linguistic roots of the word anarchy, it means, not "no law", but "no rulers" - In essence, phrases such as "government by laws and not men" describe a form of anarchy. Having one group that is much less than the whole of the populace frame the laws, and be relatively immune from their penalties, is one, mostly inferior, option among some quite more acceptable alternatives.
Supposedly, the Incans had a legal system where the penalties for acts grew if the
checks and balances work if there (Score:1)
Re: (Score:2)
If people weren't afraid to defend themselves in court, you'd see a lot fewer patent trolls.
However, it's usually cheaper just to roll over and settle.
Re: (Score:1, Interesting)
So isnt this just a case of the need for general tort reform ? It's not only patents that suffer from this kind of thing. Everything from liability lawsuits for slip and falls, to frivolous medical malpractice lawsuits burden the justice system and drive all costs up. It's rather narrow to focus on patents alone, when its a much wider problem.
Re: (Score:2)
Fields they are analyzing? (Score:2)
because they are not correctly able to evaluate patents due to lack of training in the fields they are analyzing
As an aside to your main argument - I've seen this meme come up repeatedly, and I'm not sure where it started, though I suspect it's on the patent reform side: most anti-patent people seem to think that biologists are reviewing networking patents, and that GUI programmers are reviewing chemistry patents. Two words: Art Groups. The USPTO divides Examiners into groups based on their specific areas of expertise, divides applications the same way, and only has people working on stuff they're qualified on.
You c
Re: (Score:2)
For a time, while software patents were being granted, a computer science degree was not allowed as a prerequisite for becoming either a patent examiner or a patent lawyer. So just who was examining those early software patents?
Re: (Score:2)
For a time, while software patents were being granted, a computer science degree was not allowed as a prerequisite for becoming either a patent examiner or a patent lawyer. So just who was examining those early software patents?
Computer Engineers and Electrical Engineers. For a time, when software was first being written, it was engineers doing the work, rather than computer science majors.
But it's also finer grained than that - it's not one USPTO art group for "software"... Instead, there's more than a hundred, subdivided into areas of specialty, including AI, cryptography, network acceleration, UI design, etc: http://www.uspto.gov/patents/resources/classification/art/index.jsp [uspto.gov].
Re: (Score:2)
Computer Science was not accepted until the 1990s. The field was quite well-established by then. They're still very restrictive on which CS programs they accept, requiring an accreditation that has only very recently become common, whereas they are not so restrictive with other technical degrees.
Re: (Score:2)
Computer Science was not accepted until the 1990s. The field was quite well-established by then. They're still very restrictive on which CS programs they accept, requiring an accreditation that has only very recently become common, whereas they are not so restrictive with other technical degrees.
That's because some CS programs are little more than liberal arts programs plus some coding, populated by neckbeards who will never amount to anything more than a Helpdesk monkey. Others are serious engineering programs. Until there's some standardization, it's reasonable to not accept at face value someone who holds up a CS degree.
There's no evidence, IMO, that any of the patent examiners in any of those hundred subfields are skilled in those subfields.
I would disagree. Many of them seem quite skilled to me. How many have you talked to or otherwise communicated with? Or are you just basing your opinion off the fact that softw
Re: (Score:2)
Re: (Score:2)
Re: (Score:2)
Maybe software patents are simply the most requested, and the USPTO is so fee hungry that it's cracking down on people actually giving a fuck about prior art and obviousness and whatnot. Pissing off the corporate sector would mean less application fee bling, and that brings down the budget.
Re:Just ninety percent? (Score:5, Insightful)
Yes, but if you're going to claim to know something about the law, it helps if you can pick some arbitrary criterion for why your rule works. In other words:
"Patent claims tend to fail 90% of the time" is factually incorrect.
But:
"Patent claims by trolls tend to fail 90% of the time." is so fuzzy that no one can dispute it.
Re:Just ninety percent? (Score:5, Informative)
"Troll" has a precise definition in their study - a non-practicising entity. So if the patent being sued over isn't used by the entity doing the suing in any of their products/etc, they are labelled a "patent troll" in the study.
Re: (Score:2)
"Troll" has a precise definition in their study - a non-practicising entity. So if the patent being sued over isn't used by the entity doing the suing in any of their products/etc, they are labelled a "patent troll" in the study.
Accordingly, the only way that "Patent claims tend to fail 90% of the time" could be factually correct is if 100% of patent owners were non-practicing entities. Since that's not true, the 90% claim is false for all patents.
Patents are expensive to obtain... Most companies need a revenue stream to support them. I suspect that the number of non-practicing entities asserting patents is closer to 5-10%. So, if my estimation is correct (and this is fully open to debate), we're really talking about .5-1% rather
Re: (Score:2)
Sure, but no one other than you has made the claim that "patent claims tend to fail 90% of the time", so that being not factually correct is pretty much irrelevant.
The actual article is talking about various subsets of patents and that some of them have very high "failure" rates when they make it through to a judgement. The "patent troll" subset is one such high failure rate group.
Of course there are holes in their data, heck they even point some of them out.
Re: (Score:2)
No, It is a classic case of conditional probability:
Claims that are litigated fail 90%. This is conditional on not settling and licensing before it goes to court.
Just ask any licensing professional - if you end up litigating it is considered a failure. This means that you have not managed to present a sufficiently good case to the "customer" which shows that it you have good enough IPR and it is worth licensing at the rate you are offering. It is generally more cost effective for any _REAL_ patents than dea
Re: (Score:2)
My opinion, having looked into patenting my own algorithm recently:
No. Absolutely not.
Patents in general are intended to protect an idea from the time it's conceived until the inventor has had the chance to at least recoup his research investment. Note that I said "chance", not "guarantee". In the world of physical devices, it takes a long time to turn an invention into a commercial success. First there's several rounds of prototyping (beyond what the USPTO has ever required), then manufacturing, distributi
Re: (Score:2)
Re: (Score:2)
You don't really seem to be presenting an argument for software patents on their own merits, but rather that other patents are accepted and that you don't see too many differences.
The AC's comment was regarding only software patents. I was staying on topic.
You need a considerable sum of money to manufacture a device on an industrial scale. With software, it's quite possible for a single person to do everything themselves, and the distribution network can have virtually no actual cost.
So your argument is that software is easier to steal, so it shouldn't be protected? That seems to miss the fact that research is expensive, regardless of the medium. My whiteboard markers, notebooks, and especially my food still cost money.
Perhaps more importantly, the idea that other non-software patents are good is not self evident.
I'll rephrase my first sentence in simple terms, then:
When greedy people see good ideas, they often copy those ideas. Since they're greedy, they don't pay anything to the inventor. The inventor
Re: (Score:2)
I didn't say you were off topic. I was saying that you aren't actually presenting an argument of why software patents work, but rather that software is similar to other patents, which may or may not work. Similarities between the two might just mean that all patents suck.
No, my argument is that there is less overhead in production, so there is a
Re: (Score:1)
A problem with SW patents is that they are broader than physical patents. Physical patents cover implementations, SW patents often cover mere ideas. A patent covering an implementation can be worked around by devising another implementation but patenting an algorithm means there is no way around the patent and whoever patented it first can screw over the whole world. Patenting MP3 encoding should only be able to patent a specific implementation (not limited to a specific language but specifying e.g. using a
Re: (Score:2)
That's a problem with the USPTO's reviewers, most of whom need to spend a few years in training (but good luck finding the funding for that). It's not an inherent problem with software patents themselves.
The equivalent statements in physical patents would be something like "use a thingy to move the stuff near the gadget, and a widget does something". A reviewer with an IQ over 80 should see the problems, but since they're often unfamiliar with computer science, the ridiculous patent gets approved anyway.
My
On the other hand... (Score:5, Insightful)
You can lost most of the time, but when you win, can't the winnings quite overshadow the actual effort in trolling the system?
It's not completely unlike walking up to many random women and saying, "nice shoes, want to have sex?" Sure, the vast majority will think you are a schmuck and maybe even give you a slap, but to some having an extremely tiny percentage agree makes it worth their effort. ;-)
Re: (Score:2)
Re: (Score:2)
Because more than 90% never go to trial--that's 90% of less than 10% that are losing, because going to trial for most defendants is more expensive, even if they win, than paying off the extortion racket.
Re: (Score:2)
And that's exactly it. RIM lost its lawsuit to the tune of several hundred million dollars. It doesn't matter if the patent troll ever won a lawsuit before or afterwards, they just received a lifetime payment for everyone who works there.
That's the bad part about patent trolls. They're only business is lawsuits, so engaging in one isn't a cost as it is for other businesses. Losing one is just an outcome of their daily work. Winning one is the jackpot they're after.
Re: (Score:3, Insightful)
And what REALLY sucks is that anyone who actually worked on the patent gets precisely Jack Shit.
The CEO, CFO, and the lawyers will suck up all the money. The rest get told "here's a grand or so bonus, now get back to work, slaves."
Re: (Score:1)
Patent trolls usually buy up inventions of other people. They're purely parasitic on society.
Re: (Score:2)
The judge in the losing cases should be making sure it's expensive to lose, as well. Countersuits should come into play, too.
Re:On the other hand... (Score:4, Interesting)
I bet a lot of patent trolls don't have much money in the bank.
Acacia does a lot of troll suits under other names: IP Innovations, Acacia Technologies, or Acacia Media Technologies, among others.
I wonder if this is a way to protect Acacia from having the pay the losses. Maybe when they lose, they just tear up the piece of paper on which that legal entity existed. ...and make two more.
Re: (Score:2)
I'm surprised I don't know this, but I wonder what happens to the patent once you lose a case; can the judge invalidate it completely if he decides it's just wrong? Certainly you could never press the case against anyone making whatever the guy you just sued was making, but it's unlikely anyone would, so you could still sue people who are making slightly different things.
As for the shell company bankruptcy thing, that's a major flaw in corporate law, IMO. If the owners of corporations were more on the hoo
Re: (Score:3, Informative)
Depends. The patent holder can lose either by the judge invalidating the patent, or by the judge saying that the accused's usage doesn't violate the patent. The latter obviously leaves the patent intact.
If the judge invalidates the patent, the patent holder can still go back to the USPTO and ask to change the wording and for a new patent with a different/narrower scope to be granted. (I'm not 100% sure, but I think this is a Reissued patent [swpat.org].)
For example (although it's not the same situation), the Bilski
Re: (Score:2)
The patent holder can lose either by the judge invalidating the patent, or by the judge saying that the accused's usage doesn't violate the patent. The latter obviously leaves the patent intact.
But importantly, the latter also leaves a workaround for the patent in case law.
Re: (Score:2)
You do not have to win at a trail to "win" the lawsuit.
TomTom was a huge win for Microsoft, but MS did not win at a trial.
HTC was a huge win for Apple, but Apple did not win at a trial.
In fact, it's better to not go to trial, because then everything is nice and undisclosed.
Re: (Score:1)
What you say is so incredibly true its not even funny.
I had a friend that more or less did exactly that, he was slightly more polite than 'nice shoes, wanna fuck' but the conversation was never more than a minute long.
He pretty much could get laid any time he wanted, he just went through a LOT of women in the process, but he won often enough that you had to wonder if he didn't really have the right way to go about the process if you just wanted to have sex.
Bad Patents (Score:2)
Either way, what does seem pretty obvious is that all those ridiculous patents you see in patent lawsuits are, in fact, bad patents.
What seems obvious is that the *majority* of ridiculous patents are, legally, bad patents. This doesn't seem to suggest that *all* bad patents are caught by the judicial filter. It's a minor nitpick, for sure, but I'm 100% certain that the number of "bad patents" that have held up in court is non-zero.
Re: (Score:2)
Re: (Score:2)
Re: (Score:3, Insightful)
Given that the 10% usually results in a uber-massive payday, and that the company being litigated against usually settles anyway?
Not sure of the odds, but it likely adds up to a HUGE reason as to why patent troll 'companies' are realizing a lot of investment money, ne?
Re: (Score:2)
With a success rate of 10% it must be certainly worth it.
Actually, the success rate is much higher than 10%. They win 10% of what goes to trial. But for every suit that goes to trial, there are many more that are settled much earlier on. That is actually where the big money comes from. Patent trolls don't actually want to go to trial; they want a settlement. They end up having to go to trial because someone decides that they think they can win in court and refuse to be extorted for money. And as soon as they win at trial, they know that anyone else that was
In fact, the opposite of true (Score:2)
While some may suggest this means 'the system is working,' that's not really true.
How in the heck would a 90% reject rate indicate "the system is working"? If that number is correct, that supports the idea that the U.S. patent system under our current legal system is stifling innovation. I.e. "the system is broken" is a more sensible conclusion. With that kind of failure rate on a challenge, the patent trolls simply must not care that the patent is weak. They expect targets to just roll over and settle most of the time. Business model:
1. Buy up patent with a domain applicable to tar
Re: (Score:2)
It means the court system is working to correct the problems in the patent system. You can't build a system the size of the US government without some pathological failures, so arguing that the 'whole' system isn't working is like suggesting that water is wet. Yep.
Re:In fact, the opposite of true (Score:4, Insightful)
The number of people who settle rather than litigate - compared with people who later litigate and get the same patents thrown out - indicates that you are full of shit.
The system isn't working, because the cost to challenge a patent is so high, and the process takes so long, that most companies just pay into the extortion racket.
Re: (Score:2)
challenge (Score:2)
The costs aren't that high, I'm always curious why parties in litigation almost never file a re-exam.
http://en.wikipedia.org/wiki/Reexamination [wikipedia.org]
The filing fees are low and anyone can supply prior art that they feel invalidates the patent it.
The backlog for re-exams isn't that long either.
Re: (Score:3, Informative)
A thought (Score:4, Insightful)
Re: (Score:2)
maybe (Score:4, Insightful)
Either way, what does seem pretty obvious is that all those ridiculous patents you see in patent lawsuits are, in fact, bad patents
Oh sure, I'll accept that they are bad patents, but I'm not going to accept that the law recognizes them as bad until Amazon's one-click patent is invalidated. Obvious ideas can be, and have been, patented, and that is a serious problem.
uhh (Score:5, Insightful)
Of course we don't get a link to the paper itself, but the fact that where defendants decide their case is strong enough to go to trial, they tend to win, is not especially surprising.
Re: (Score:2)
s/their case is strong enough/they are rich enough to afford a lawyer long enough to survive the trial/
Re: (Score:2)
Only 90%, huh? (Score:4, Insightful)
Turning that 90% figure on its head, if the average ROI for buying up a patent and sucessfully suing some suitably wealthy potential infringers in court is more than nine times the outlay then unfortunately being a patent troll is still a viable business model.
The summary of the summary of TFA misses the point (Score:5, Insightful)
Re: (Score:2)
The fact is that the statistics are limited, and the most solid statistics are about final judgments, because it's hard to
In the UK they're smarter than us (Score:1)
Re: (Score:2)
So what about the ones that settle? (Score:2)
All that proves... (Score:2)
All the proves is that defendents settle unless they have a case they believe has a 90% chance of winning.
Statistics... (Score:4, Insightful)
The majority of people in marathons do fairly well. That cannot be extrapolated to say that the majority of people would do well in a marathon. Its stupid. This study is no different.
Disclaimer: I didn't RTFA because stats scare and confuse me.
Selection bias (Score:1)
Fight back! (Score:1)
Yes, I'll break out my vast fortune, right now... I mean it... I'm headed to be bank... You'll be sorry...
Patent trolls have a 100% success rate of draining your account... They should have to post a bond to pay your expenses if you win
Timely (for me) (Score:1)
Expected Value (Score:4, Insightful)
The probability of success is irrelevant if the expected value (success probability * average profit) is still positive.
Why "winning" the suit is not important (Score:2)
Consider the scox-scam. The lawsuit was ridiculous from day one. There was no way scox could "own" UNIX. But, the scam is now well into it's eighth year, with no end in sight.
If, like IBM, you happen to have around $50 million to $100 million to defend yourself against a bogus lawsuit, maybe that's not a big problem.
But if you are a smaller company, like TomTom, it may be better to just cave in. Maybe it's easier for HTC to cave in to a bogus lawsuit. Those suits were not officially "won" at trial, but they
The lawyers on both sides win, always (Score:2)
Re: (Score:2)
At first if you don't succeed, try, try again? (Score:2)
Isn't this the little engine that could?
the human struggle?
Never going to give you up?
err, sorry, guess i'll leave the rickroll out of this.
Here's a real-life example... (Score:1)
In Litigation, Final Outcomes Aren't Everything (Score:2)
The problem with this study is that it ignores several key advantages enjoyed by patent trolls over their targets:
They are spamming a lot of lawsuits (Score:2)
Effective (Score:1)
What a waste of time.
Innocent until speculated guilty (Score:2)
In the case of Microsoft it seems to be the case anyway.
They have used the news of a competitor being sued for IP infringement in their own interests numerous times now. Just recently with Google and Oracle over the Android Java implementation.
Maybe if they concentrated on making their products better and gave people a genuine reason to upgrade then they wouldn't need so much FUD?
Selection Bias (Score:2)
90% of software patents that went to trial were invalidated. Obviously, the ones that go to trial are usually the ones where the defendant has a high expectation of success, otherwise they'd just settle.
Nothing to see here, folks. Move along...
Re: (Score:2)
Re: (Score:2)