Why Red Hat Is Wrong That It's Better To Just Pay Patent Trolls Sometimes
from the that's-why-you-get-more dept
Last year, people got reasonably upset when it came out that Red Hat had settled one of many patent lawsuits filed against it, this time from patent troll giant Acacia. People were upset both at the decision to settle and then to keep the terms secret. Red Hat, after all, has spoken out many times against software patents and patent trolls. Slashdot points us to the news that Red Hat’s CEO has explained that sometimes it’s just easier to pay up. It’s clear that he really doesn’t want to and thinks the whole practice is distasteful, but there are times when he recognizes it’s just cheaper to pay up:
“When it’s so little money, at some point, bluntly, it’s better to settle than fight these things out.”
He does say that they fight on bigger cases or cases they feel are especially ridiculous. But, in others, it’s just cheaper and easier to settle. I certainly understand the reasoning. And I definitely understand the short-term cost-benefit analysis. If you can pay off the patent holder for less than it’ll take to fight the case, even if you win, that seems like a good deal. Except… in the long run, this may be penny-wise and pound-foolish, because as you build up the reputation as a company who will fold as long as the settlement demands are under a certain level, then all you do is encourage more trolling behavior, leading to more new lawsuits with more patent holders demanding a handout.
Again, I can certainly understand the basic reasoning for settling, and can’t really begrudge any company that decides to settle to avoid a lawsuit, but it is a little disappointing that this just perpetuates the problem.
Filed Under: patent trolls, patents, settlements
Companies: red hat
Comments on “Why Red Hat Is Wrong That It's Better To Just Pay Patent Trolls Sometimes”
Kidnapping
Deciding whether to settle out of court with a patent troll is similar to deciding whether to pay a ransom to a kidnapper. In both cases, you know that what you’d be doing would contribute to the overall problem, but you’re balancing that abstract, overall, long-term negative against the immediate, personal, short-term benefit.
The difference, I think, with settling with a troll is that you’re not just making it worse for your fellow citizens, but for yourself. As Mike pointed out, settling with trolls may be good in the short term, but you’re setting yourself up for being sued for the lifetime of your company.
I know that it’s the nature of people (and corporations) to focus on short term gains, but I still don’t see why companies don’t draw a line in the sand and say “No, we will never settle with anybody, ever over a frivolous patent lawsuit.” Sure, there’d be some additional costs up front, but eventually the trolls would just move onto easier targets i.e. companies that were stupid and shortsighted enough to actually give into the demands of the trolls.
Re: Kidnapping
I agree. You give a blackmailer a little money and they will always be back for more. If everyone refused, the risks for a patent troll would increase and the rate of return would decrease. Right now, there is not a lot of negative incentive for a troll.
They threaten… a lot of people license for a lawsuit is filed.
They file suit… a lot of people will settle before it goes to the jury.
No settlement… if it looks bad for the patent troll, they drop the suit and look for the next target.
Re: Kidnapping
I think a better analogy would be a younger sibling annoying the hell out of you. If you ignore them for long enough, they will go away.
Re: Kidnapping
Reminds me of a behavioral study. Three groups of rats had a lever in each group’s cage. Group 1 would always get a food reward when the lever was pressed. Group 2 would never get a food reward. Group 3 would sometimes get a food reward (random between 1 and 10 presses, IIRC).
After a month, all three levers were disabled. Results:
Group 1 — tried for a day or two, then quit.
Group 2 — never got conditioned to trying, so no change
Group 3 — pressed the lever thousands of times because they were conditioned to believe that *eventually* they would get something
Inconsistent (or inconsistently enforced) policy is usually the worst solution.
Although it would be redundant and unnecessary, I think they should redo the study with lawyers in the cages.
Re: Kidnapping
Of course, the assumption is that the suit is frivolous and has no chance of winning.
I know that assumption seems to apply to just about every patent lawsuit here at Techdirt, but the actual parties might not see it that way.
This is a collective action problem, where litigating would produce positive externalities that are not easily internalized.
If RedHat fights, it pays more in legal fees, and yet the benefits of deterring future trolls accrues to the entire industry (including competitors). Moreover, in some of those cases, RedHat could be enjoined from distributing software, which is the commercial equivalent of the death penalty.
So they settle, and hope the trolls sue your competitors. The trolls are counting on the same thing. That’s why they fought against patent reform in Washington.
This is why we need to fire all of the lawyers.
Out of a cannon and into the sun.
Re: Re:
Futurama FTW!
Paying up should be illegal
Paying ransom and settling bogus lawsuits have something in common with dumping toxic waste in Love Canal. They might be beneficial to the participants, but they’re harmful to society at large.
Somebody was wrong, either the plaintiff or the defendant.
To me you should not be able to settle without admitting wrongdoing. That’s bad for everyone. The innocent get shaken down after having done nothing wrong, and the guilty get off for just cash. Once a case has been filed it becomes public interest, and the participants shouldn’t be able to walk away without a public accounting of the facts.
Re: Paying up should be illegal
I think the simplest patent reform we could possibly implement that would have the most immediate and powerful impact is to change the law so that all civil proceedings require the losing party to pay the winning parties legal fees.
I actually think this would discourage a great deal of bogus lawsuits. I also think it would encourage people with meaningful lawsuits to file, even if they don’t have the means to do so when the case begins.
What do you mean?
Appeasement worked out great for France right before WWII!
RedHat's strategy can be reasonable
I think it depends. BS lawsuits are filed hoping for a payoff settlement. If you are consistently fighting lawsuits under a certain threshold of BS, then you are creating a disincentive to file similar lawsuits. For lawsuits above that threshold, where they have a chance at actually winning, the disincentive created by fighting is weaker. So assuming the fight/settle threshold is set correctly, I don’t think you inevitably make more trouble for yourself in the long run.
Perhaps with unlimited legal resources, one could create the strongest disincentive to trolling. With limited resources however, a choice has to be made – is the ROI on the disincentive to troll greater than the ROI on development. Put another way, RedHat could get a reputation as a company that you should never sue because they will always fight, but without better products that reputation will do them little good.
RedHat's strategy can be reasonable
I think it depends. BS lawsuits are filed hoping for a payoff settlement. If you are consistently fighting lawsuits under a certain threshold of BS, then you are creating a disincentive to file similar lawsuits. For lawsuits above that threshold, where they have a chance at actually winning, the disincentive created by fighting is weaker. So assuming the fight/settle threshold is set correctly, I don’t think you inevitably make more trouble for yourself in the long run.
Perhaps with unlimited legal resources, one could create the strongest disincentive to trolling. With limited resources however, a choice has to be made – is the ROI on the disincentive to troll greater than the ROI on development. Put another way, RedHat could get a reputation as a company that you should never sue because they will always fight, but without better products that reputation will do them little good.
Re: RedHat's strategy can be reasonable
Could you please make sense ? ”For lawsuits above that threshold, where they have a chance at actually winning, the disincentive created by fighting is weaker.”
I assume with ‘they’ you are talking about the trolls. So lawsuits above a certain bullshit level, have a likelihood of being won by the trolls. So, if redhat would fight such an above-average-lawsuit, it is stil not clear whether redhat would win. So, indeed, if redhat would fight it, it is not such a good disincentive to the troll, who might win. On the other hand, if redat is certain to win, it creates a higher disincentive to the trolls. So, I cannot understand why you then conclude that ‘the treshold is set correctly’.
Differntly stated: if redhat might loose and did not choose to fight because they might loose, how does that disincentivize (what a word) the trolls ?
Re: Re: RedHat's strategy can be reasonable
An assumption is not a conclusion. “Assuming the threshold is set correctly” does not mean concluding the threshold is set correctly. I made no conclusions about what the threshold should be or whether RedHat had set it correctly. Rather, I made an assumption about something for the sake of argument because I had insufficient data on that point. Similarly, you assumed that “they” referred to trolls so that you could make your argument.
RedHat’s decision not to fight where trolls have a decent chance of success does not provide a disincentive to troll. It preserves resources to develop the products which are the company’s primary purpose.
The argument is that a strategy which involves fighting where there is a high chance of success (and therefore a high ROI in the form of trolling disincentives) and capitulating where the opportunity costs of fighting outweigh the potential benefit (meaning the lower chance of success is less valuable than product development) is a reasonable strategy. Otherwise you risk developing excellent disincentives to troll, but at the cost of losing the ability to create new tech.
Additionally, trolls with good cases are less susceptible to intimidation through fighting back, as they have a chance of a payday through a verdict. So the troll-fighter will have decreasing gains in trolling disincentives as troll-cases get stronger.
Therefore, if you create a disincentive to troll with BS cases while settling good cases, you maximise the bang for your buck in creating a disincentive to troll while minimizing the drag on resources you get from paying lawyers.
Re: Re: Re: RedHat's strategy can be reasonable
Are you thus saying that I should read your first line as: ”For lawsuits above that threshold, where they [redhat] have a chance at actually winning, the disincentive created by fighting is weaker.”
That makes even less sense because then you say that if Redhat would win, the trolls would not see that as a disincentive ?
Otherwise, I fail to understand your reasoning. Whether the case is BS or non-BS, we are still talking about software (as far as I see it), and settling is never an option. These are abstractions, and those should not be patentable. So you don’t want to create disincentives for patent trolls, you want to get rid of the entire patent system. So following your argument I could then argue that rolling over to every troll and giving them your company is the best solution. No money goes to the legal system and nothing will be created. The system will come to a standstill and trolling will no longer exist. Problem solved.
Re: Re: Re:2 RedHat's strategy can be reasonable
Man, for a guy claiming someone else is making little sense, you’re making precious little yourself.
I honestly don’t know where to begin, but I think we can start with this: As I see it, the only way to read his sentence sensibly is that “they” = “the plaintiffs” not Redhat.
Re: Re: Re:2 RedHat's strategy can be reasonable
Are you intentionally misreading my comments to yank my chain?
Pronoun free version:
Trolls with weak claims don’t have a chance of winning their lawsuits. RedHat should fight when RedHat will definitely win. That way trolls with weak claims who only sue because they think they can get a quick settlement will not sue because they know they won’t get that quick settlement.
Trolls with stronger cases will not be intimidated by RedHat fighting back. When a troll lawsuit is strong enough, RedHat may not win and the cost of litigation may be high. In that case Redhat should invest RedHat’s money in software development instead of lawyers because investing in lawyers will not deter trolls with strong cases from suing RedHat.
Whether RedHat chooses to settle or fight with patent trolls has absolutely no effect on whether software can be patented.
Arguing the total invalidity of all software patents in a trial for patent infringement will lose every time. Arguing a given patent is invalid may win, but will do absolutely nothing to invalidate software patents in general. They are so well established under current precedent that only Congressional legislation or a decision by SCOTUS can eliminate them. A trial court wouldn’t even think about it.
Re: RedHat's strategy can be reasonable
I’m not sure what you’re saying either; Mind if I try re-wording it to see if I have it right?
Red Hat will not exist infinitely; Even assuming that to be true, the patent system will eventually change and the problem will shift/fail to be significant at some point in the future.
On an infinite timeline, discouraging trolls, (and all others), is always profitable, no matter the short-term cost.
On a finite timeline, the cost and strength of each lawsuit could be weighed to produce a threshold above which it is worthwhile to fight, and below which, the lifetime cost of the lawsuits will likely fall beneath the short-term more expensive cost of fighting and thus discouraging them permanently.
(This is assuming troll lawsuits).
The threshold is further raised because if the money is, instead of fighting, spent developing new software & techniques, there will be less ‘strong’ lawsuits in the first place.
Is this the gist of what you were saying?
Re: Re: RedHat's strategy can be reasonable
I think that you have the gist of what I was saying. Thanks for restating it in an illuminate way. I Although I will quibble a bit – I think that even on an infinite timeline, fighting everytime will not create a perfect disincentive. There will always be trolls who have a high enough chance of success to make it worth it to roll the dice. Or alternatively a troll may only care about creating trouble for a competitor ..cough.. Apple.. cough.. and be willing to throw enough money into it to guarantee a pyrrhic victory for a weaker opponent.
It’s like negotiating with kidnappers or terrorists..
First, getting a “reputation” of taking every case to the mat has not panned out for advocates of such strategy. It has been largely debunked. The flaw is to assume that patent holders will blindly assume past behavior is a predictor of future behavior when the facts and circumstances of each case is different including, among other things (i) breadth of claims; (ii) success of previous litigation licensing; (iii) perceived “value” of the patent; (iv) perceived validity of a particular patent… and so on and so forth.
Second, the idea of “taking one for the team” doesn’t make much commercial sense either. At some level, it is rational for a competitor to cut its losses especially if that means greater costs down the road for its competitors.
Third, taking a “loser” of a case to trial is expensive and risky.
Fourth, big, public companies are accountable to shareholders and board members who may not share their “principles” if it meant saving the company millions in legal fees and potential damages claims.
Fifth, proving invalidity of even the most “obvious” patents is expensive and time consuming and far from a sure thing when put to a lay judge/jury or even if bringing it back to the PTO in reexam. It can take a long time for a court to get to rule on invalidity and a significant chunk of invalidity questions are not resolved at summary judgment but at trial. It’s usually easy for a patent owner to create some issue of fact regarding invalidity to preserve it for trial.
Finally, patent cases are not unique. Product liability, discrimination cases, and “nuisance” suits are frequently settled even if the case is weak. Such behavior is a function of the system itself.
Re: Re:
So…everything is fine? Nothing to see here? You actually support a system that allows leeches to thrive off the success of others?
Investors
My guess is it has a lot to do with RedHat being a publicly traded company and just not wanting to have to explain these expenses to investors.
If each quarter at the earnings call you have to say you paid $X to lawyers to fight cases, and you don’t know when the case will end, that’s more unsettling to investors then saying we paid $X, a one-time fee, to Acacia this quarter for licensing.
I doubt RedHat has the legal budget of the big-wigs like Google. While they may *want* to fight every case on principal, at the end of the day they need to be profitable if they want to stay in business. Hemorrhaging money on these cases not only has the potential to lose them a ton of money on lawyers fees, but at the same time will drive away investors.
This strategy worked out great for RIM.
The reason they settle because the costs are DRAMATICALLY lower than fighting. A fight doesn’t guarantee a win (See Microsoft vs. i4i.)
It’s not just about costs, it’s about time/resources. Putting money into a 5+ patent suit will impact every bottom line for 5+ years, scaring off investors/growth. Look at RAMBUS/Nvidia for recent examples of investors pulling away due to suits that dragged on for many years.
Business is also least-cost oriented. If there was a way to go to court and have a 90%+ assurance of a win, and have it resolved within a year, sure they’d go to court. That system won’t ever exist in modern times.
NPE / PAE
One thing I find interesting in a recent FTC report is the FTC’s distinction between “good” NPEs and patent trolls (which it refers to as “PAEs”). Many have long noted that there is a need to differentiate between NPEs such as universities and those other entities who abuse the system through arguably-excessive patent litigation. Distinguishing between those bad actors and other NPEs may be helpful in narrowing the focus and the terms of the debate over patent trolls.