I always thought Hotz an interesting character. This paragraph adds to that impression:
"What he didn’t take into account is that comma isn’t run by rational actors in suits sitting on a committee. It’s run by me, George Hotz. I’m willing to lose $1M before I give him $10k. We will hire an amazing legal team, fight this, and while doing so invalidate his patents so they can’t be used against anyone else. Not because it’s rational, but because it’s the right thing to do. No patent troll will ever get a dollar from comma."
Take for instance the problem of patent trolls, which could be much more comprehensively solved by regulatory/legal reform than by one guy with a well-funded legal team.
Obviously I don't think that premise false, else I wouldn't have said it.
You say you want regulatory/legal reform. Do you not think this starts with legislation? Do you not think that this requires action by the current legislative system? Do you think that the current legislative system is not broken?
Very importantly, patents are federal law. That means federal legislation. That means Congress. An institution that about 93% of Americans have fairly low confidence in. (See https://news.gallup.com/poll/394283/confidence-institutions-... for a source.)
It explains the economic reasons why it is hard to get collective action to happen. It also explains why any government organization organized for collective action in time is likely to undergo regulatory capture, meaning that the people you wanted to regulate are in control of the regulations and manipulate them for their own purposes. It also studies the various approaches that have been taken to solve this, and how effective they have or haven't been.
You'll learn WHY we don't simply all self-organize to act in our collective best interest.
The description of the book seems to take "acting as a group is necessary" as a given, actually, by my read. Right up front! "Economists have long understood that defense, law, and order were public goods that could not be marketed to individuals, and that taxation was necessary."
So if the alternative is more self-motivated individual action, you obviously don't have to go far to find tons of examples of how that results in bad outcomes. Here's someone motivated to say "fuck you" to patent trolls. But the patent trolls themselves are motivated by self interest to be a troll in the first place, so "individual agency" here is at best a wash. Difficult collective action to reform the mechanisms patent trolls take advantage of still seem far more likely to work here than just hoping for more individuals willing to light their money on fire on principle.
If the book merely said that the cause was hopeless, I would not recommend it.
We are surrounded by successful examples of collective action. But they also tend to be interestingly suboptimal in lots of ways. The book provides a framework through which you can understand why this happens.
A public good is a good which, if it is provisioned, is available to all regardless of how much they contributed to its provisioning. There are many, many examples of public goods, but I'll use clean air as an example.
The problem with public goods is that it is only in your personal interest to contribute to the extent that you personally reap the reward from your contribution. So, for example, catalytic converters cost about $1500. If you live in a city, you personally benefit more than $1500 in life quality from the fact that everyone around you has a catalytic converter. But if your catalytic converter goes, you don't benefit $1500 from YOUR catalytic converter. And so almost nobody would, without some other motivation, bother spending $1500 on a catalytic converter.
And yet, public goods do get provisioned. Here are the basic ways.
1. A single large entity benefits enough from the good to provision it for everyone else, who gets to free ride. The book calls this, the exploitation of the large by the small. If ever you see someone arguing that some big country or company should do something nice for everyone, they're hoping for this.
2. A small group combined can provision it. This usually results in complex negotiations, cheating, and so on as each tries to do the minimum necessary. The history of negotiations within OPEC on controlling the supply of oil is a good example.
3. A fundamentally coercive organization which exists for some other reason can ensure the provisioning of it. For my catalytic converter example, that is the government. If you fail to have a $1500 catalytic converter, the government will take steps up to and including putting you in jail.
4. People seem willing to put a small amount of energy out for a good cause. For example a lot of people are willing to put the energy out to vote, even though your personal vote is unlikely to swing any election in your lifetime. Some people are willing to put a lot of energy out.
5. The good may wind up provisioned for some other reason. For example open source software is a public good, which programmers voluntarily create and maintain for a wide variety of reasons. But very few programmers are creating it out of some notion of acting to maximize the common good.
But that is a very short take with very few examples. The book is quite readable and offers a lot more on all of this.
We definitely need more collective action, but a lot of evil happens because people in positions of power hesitate to do the right thing out of a sense of fiduciary duty.
Collective action could align the incentives better, but leaders willing to risk some of the company's assets to do the right thing should, IMO, also be encouraged.
Hmm. Fiduciary duty isn't a "sense"; it's a legal obligation on anyone who occupies a position of trust. I can't see how being aware of your legal obligations causes "a lot of evil".
Well, yeah. Either he inspired someone else to do it or he did it himself and used the power of STFU and impeccable opsec to get away with it. Here is the timeline.
On February 12, 2011, Hotz posted a one-minute diss track against Sony
On April 11, 2011, ... settlement out of court. This included a permanent injunction against Hotz doing any more hacking...
The attack occurred between April 17 and April 19, 2011... The outage lasted 23 days
To date, there is no confirmed evidence any credit card or personal information has been misused
"But that's purely circumstantial!"
Yes, deliciously so. This is all thoroughly speculative and had better remain that way.
His Lex Fridman interview was fascinating. I knew this would be his response just from the headline. Surprised the troll didn't do even a little research.
It is amazing how many lawyers are completely incompetent at everything other than billing. I expect that they will write a letter complaining that they are being harassed by Hotz and his associates and are cancelling the lawsuit for their own safety.
Old Newegg. That lawyer has moved on along with the rest of the leadership after a change in ownership.
Also George Hotz fought and cowed Sony’s legal team when he was a kid in a saga involving ‘marcan42 of Asahi Linux. These patent trolls must be very foolish to try to intimidate him.
In my opinion, for someone his age that’s essentially a victory if it doesn’t involve any payments or criminal repercussions. In theory, marcan42 would have been subject to a very similar or same settlement but he went on to continue working on Sony stuff without repercussion:
“In 2016, he ported Linux to the PlayStation 4 and demonstrated that at the 33rd Chaos Communication Congress by running Steam inside Linux.”
Considering that Sony’s primary claims were:
Violating the Digital Millennium Copyright Act (17 U.S.C. § 1201)
Violating the Computer Fraud and Abuse Act (18 U.S.C. § 1030(a)(2)(c))
Violating California Comprehensive Computer Data Access and Fraud Act (§ 502)
I’d say that the outcome failed to reflect Sony’s perspective on the situation and much more closely aligns with “George Hotz faced nearly no actual consequences”.
Hey, in general it's better to assume good faith. It's snide to just assume I haven't looked at the facts or attempted to understand anything.
According to Ars Technica the DOJ indictment stated:
* There was a screenshot of the SX OS menu with commercial games in it.
* A public changelog for SX OS publicly trumpeted the feature of "launching your Retro Game ROMs for emulators directly from gDrive,"
* They were "attempt[ing] to cloak its illegal activity with a purported desire to support homebrew enthusiasts who wanted to design their own games."
* the product description for the team's SX devices notes that you can "play all your favorite game backups straight off of the microSD card (or external USB storage!)"
So, yeah, I agree they were selling circumvention devices with at least some level of piracy intent (and that was used to send him to prison after all.) But I think you can also agree the delta between what they were doing and what geohotz did with the PS3 is not very large.
Yeah, and not a week later anonymous hacker(s) took down the Play Station network for 23 days. Either he did it himself and got away with it, which is impressive, or he inspired someone else to do it and get away with it, which is also impressive.
“anonymous hackers”? Publicly unidentified hackers you mean?
That hack almost certainly had nothing to do with the jailbreaking scene. Sony pulled down their infrastructure to investigate an extensive compromise.
I'm torn between considering it a power move of mentioning yourself and linking to the Wikipedia entry about you, or not. But I appreciate the troll hunt.
There very well may be a second order game theory at play - not only you broadcast far and wide that litigation will be exceedingly costly for anyone that tries, but you add an explicit threat that any attempt will result in invalidation risk to plaintiff patent portfolio.
For patent trolls, time spent on legal is mostly opportunity cost, but killing their patent portfolio has a real dollar cost associated to it.
Many trolls like I think Myhrvold's have something close to a new shell LLC for each major patent and sometimes it can be hard to link it all together (the recent changes to LLC/corporate ownership privacy should help here). It can be hard to even know their full portfolio.
Next thing we know, the officers of the patent troll company die in a horrific car crash involving 4 empty vehicles running comma.ai software, quickly ruled an accident after the ruling judge notices an awful amount of empty self driving vehicles accompanying them anywhere they go.
Any engineer who has ever filed a software patent soon realizes that the entire patent system in America (at least with respect to software) is a scam. It seems to be designed to enrich lawyers and tax engineers.
It's almost like all the kids in school who failed out of CS 101 went to law school and cooked up a scheme to tax their classmates who actually stuck it out and built something.
Why the hell do we have a system that gives a 20 year monopoly to assholes for shit they cook up on a cocktail napkin over lunch with their lawyer? This is how we used to brainstorm stuff to patent. It's utter bullshit. I refuse to participate in creating more bullshit patents.
I've filed multiple (hardware) patents in previous companies. Which part, exactly, is a scam?
I think the only reasonable criticisms of the patent system is a lack of intelligent lawyers in the areas of discipline that often fail finding similar claims in other filings, and that patents only benefit large companies, not DIY garage makers like the rest of the world thinks.
I also think your shitting on lawyers is typical of people that don't really understand legal frameworks. Would you find it reasonable if I said: "software engineers are such a joke, everything they do is so over complicated just so they can get rich like those SanFran dudebros, and for what, so that I can shit in a smart toilet with a web browser and AI? How stupid." ... Oh, shit. Wait a second... :)
I've been on the receiving end of one of these (for my own invention, no less), helped bust another. Much good luck to George Hotz on this one. I can't stand patent trolls and even as an atheist I secretly hope there is a hell so they will get what's coming to them.
To play devil's advocate, there is nothing about being a patent troll inherently illegal. Hotz feels the patent is invalid, but it might not be. Our system does not require you to ever build the object patented before extracting fees from it. This is in fact the way the US patent system works.
The patent in the complaint doesn't even have anything to do with what we do as a company! It describes a system that calls 911 when it detects a crash (like onstar). Ignore the (very generic) title and read on in the actual patent. https://portal.unifiedpatents.com/patents/patent/10454707
But for the purposes of attempting to extract a settlement, it doesn't matter. It will only matter after a long and expensive trial.
(I would like to invalidate the patent entirely, which several of their patents already have been. But even if it is valid, it also has to apply to us)
Once you’re done crushing this patent troll, go after the USPTO for granting bogus patents in the first place - you shouldn’t have to bear the costs of fixing their fuckups.
The disclosures in the patent aren't what they are suing you over. Read the claims. Patents like this often have a lot of stuff shoved into the disclosure so that they can keep the patent application alive to write more patents based on the original disclosure. The claims appear to be written to read on devices like yours: My knowledge of your system is a bit dated, but I thought you had a device that you plugged into a port in the car instead of a manufacturer-installed thing. This is why this patent has a 2007 priority date, but a 2018 application date: they essentially wrote a ridiculously broad disclosure, and kept filing patents based on it.
Dataspeed did appear to have tried to invalidate this patent in 2019 (see the IPR documents linked there), and succeeded on all but claim 20. If your device does not have a relay which shuts off its electrical connection within the car, you're probably safe. You could try to invalidate it, but you should read the docs from the IPR to see what Dataspeed did and what you might need to do.
My gut-check guess is that claim 20 is probably not going to get invalidated unless you can find some weird after-market car part from before 2007 which does a safety shut-off of a car component.
This is a good start but not fully there. Ideally you would like a bigger published technical description of that electronic stop device, so you can map the device onto the claims to show that a claimed device existed before the patent was filed.
A lead would need to teach significantly more to even be considered.
Does the vehicle have (1) an after-market device installed that (2) detects that a first component stopped sending commands to a second component and (3) uses relay switch to disconnect the failed first component from the second component and (4) starts sending commands to the second component on its own that are indistinguishable from those sent by the first component?
Technically, the above requirements are needed against claim 20 of '707. Claim 6 requires fewer features.
That doesn't help us because:
(1) claim 20 is asserted in the complaint [0], and
(2) claim 20 was upheld in a previous challenge to its validity (an "IPR"), so it's got some staying power.
You are describing one scenario where this invention could be used ("a system that calls 911 when it detects a crash").
The legal protection provided by this patent is _not_ limited to any specific context or scenario.
Let's take '707 claim 6. _Any_ commercial product with each and every one of the following bullet points ("limitations" in lawyer-speak) infringes and must purchase a license to this patent.
[1] A vehicle, comprising:
[2] a factory-installed first apparatus configured to generate an electrical signal;
[3] a factory-installed second apparatus configured to receive the electrical signal; and
[4] a retrofit apparatus electrically connected to the factory-installed second apparatus,
[5] wherein the retrofit apparatus generates a mimicked electrical signal independently of the electrical signal generated by the factory-installed first apparatus, and
[6] wherein the factory-installed second apparatus receives the mimicked electrical signal.
And at any rate, "illegal" is not the same as "immoral". Even if the troll's actions are within the law, in my opinion they're utterly immoral and scummy.
I like that statement. "I'll be the devil's advocate" is right up there with everyone stating they will defend hate speech "to the death" because of free speech. Why not take your time to say you disagree with someone instead?
There's a time and place for it, like "devil's advocate: what if we decide we don't trust this vendor's performance and want to delete our data after we've already started using them?" But when it's shorthand for "I'm going to be deliberately contrarian, and ...", then I tire of it.
I'm not sure he's making a legal claim when he says they're not building anything. I think he's just trying to establish that they are not adding anything to the universe and are just parasites. Companies that are actually adding value are being preyed on by patent trolls. It's a kind of capitalist entropy
The most irritating thing about these suites, is often they know setting the IP lawyers on them costs more than the claims. However, if you have an extra $60k in tax deductions to burn, than enjoy the show with popcorn.
In addition to taking on and crushing patent trolls, someone needs to also go after the USPTO for causing these people/companies financial hardship by granting these bogus patents in the first place.
I think patents are usually granted if filed in the correct format. The expectation is, challenges are where weak patents are discovered. The patent office hasn't got enough lawyers to examine every application.
Also, trolls rarely are the ones creating these patents. They just buy up some basket of idle patents, then spend their morning looking for any company announcing anything that has a couple words in common with something in their 'portfolio'. They send a boilerplate royalty request or whatever. Rinse and repeat.
Yep happened to me. The parent company was an e-cigarette business and they use the profits to buy random patents that have nothing to do with their core business. It was like getting a shakedown call from a mobster.
Patent offices aren’t full of specialist in every single area, and they don’t pretend to be. You need to meet few fairly simple requirements and patent is granted.
If patent is invalid the way to prove that is to challenge that in court.
That system, that makes filling patents easy, is a whole reason why patent trolls exists. But it’s not a fault of patent offices, that they’re mandated to work that way.
The above link is a published application rather than a granted patent. US20170170984A1 by itself carries no legal rights whatsoever.
In this case it was granted with largely similar claims (as US10027505B2). However, claims typically get significantly revised by the time they are granted, and less than half of applications are granted at all.
In every western country - except the US and England - the loser pays for the cost of the trial (all parties' lawyer fees + the court).
Just this reduces frivolous lawsuits. If you have a good case, sure, go ahead. If your case is weak, you run a real risk paying not only for your lawyers but the other side as well.
(Note that there are disadvantages to loser-pays as well. Image you legitimately want to sue corporation X because they did something atrocious. Big corporation X can now run up lawyer cost that you have to pay if you lose.)
Edit: Sorry I was wrong about England (which leaves the US only).
Edit 2: In the US you can file for your lawyer fees to be returned, but that in itself is risk (I have some lawyer friends)
No, in the UK, the loser will pay for court costs, in general. I had a bit of legal trouble a few years ago - the other side was trying to play "solicitor" to use the law as a way of extorting me (aka threatening to make criminal complaints over various (bullshit) things) so I had to hire someone to write a polite 'go fuck yourself' letter eventually. Solicitors are duty-bound to try and get things to settle out of court, so dealing with wanna-be solicitors is always a problem for them.
What's unique about the UK is that unlike actually civilized countries (incl. the US), there is no legal requirement to negotiate in good faith. It is expected for you to do so, but if the other side figures out that you haven't and tries to take you to court over that, then the English court system will say "tough luck". Manipulating the other side to your advantage should be expected, they said. There have been High Court cases that have established this principle. so doing business with the English is always possibly problematic (e.g. Brexit and all the lies, the mentality behind all that etc). Anecdotally, I have heard from foreigners that doing business with English people is more of a headache than with ones from other countries. And I have lived here long enough to 'understand' the English better than most Americans...
In England, the losing side usually pays at least a large portion (if not all) of the costs of the winning side. 100% cost awards are rare, but generally the costs burden is placed on whoever loses.
This seems to reduce frivolous lawsuits (depending on who is suing/being sued) but you can ask for a cost capping order to limit the amount you have to pay out if you lose, which partially helps address the issue of "they're too big/expensive to sue".
An important detail: it shouldn't be (and as far as I know isn't) the loser who pays but the party who behaved unreasonably. For example, if you damage someone's property, offer to pay £100, but they refuse your offer and instead sue you and win .... £100, then you won't be paying their legal costs. In fact, they may have to pay your costs because they were being unreasonable in bringing the case to court after you had made a reasonable offer out of court.
I trhink it's more complicated than that (IANAL). Isn't it most common to make such a settlement offer "without prejudice"? Then the fact that offer was made can't be brought up in court, so it can't affect the costs ruling.
I’m not sure about England, but in the US how it works is:
The party making the offer can make a part 36 offer (or state equivalent if it exists), which grants automatic presumptions of cost sharing if rejected and the case ends with a more favorable (to the offering party) settlement or reward.
They can do a less formal without prejudice save as to costs offer, which is one the courts can use to decide cost charging, but does not get automatic presumptions of how the sharing should work.
Lastly they can make just plain without prejudice settlement offers. Which won’t factor into costs at all.
Which type of offer to make depends on who (claimant or defendant) is making the offer, and how certain they are about the amount that will be awarded at trial.
For example if a defendant is making an offer that they are absolutely convinced is more than will be awarded at trial (with the idea being that it would cost less than doing the trial without any offer, and having the normal each party pays own costs rule apply), it makes sense to make it a part 36 offer.
However when making a lowball offer, or a reasonable one but that could well be less than what gets awarded at trial, a without prejudice offer is probably better.
On the claimants side, when making a offer that they are convinced is below what they will actually win it may make sense to make it a Part 36 offer. If making an offer closer to or above what they will probably actually win, a without prejudice offer could be more sensible.
A term often used is actually "Without prejudice, save as to costs".
The meaning they're going for is "by making this offer I'm not admitting guilt, but when we come to awarding costs the fact that I tried to end this without going to court should count in my favour".
Unreasonableness will have an impact as to whether costs are awarded and/or how much will be awarded but the general principle is still that the losing side pays. Obviously, that will vary depending on the circumstances of the case but most cases don't encounter costs objections on reasonableness grounds.
In the US, if you lose a patent lawsuit, you will often have to pay legal fees and costs (within reasonable limits). It's the same as the rest of the western world for these specific suits.
These suits usually take place in east Texas, where people are unfamiliar with tech. Am from east Texas, have seen literal patent trolls open an office next to the court house. They wanted to build the web 3.0 (not web 3) bit only wanted to hire part time students. They made it look like big tech was stopping them from being successful. Their pitch to me was they had a strong patent portfolio.
It should be noted that atleast in Germany, you can't just run up the lawyer bills. There are standard rates and you only pay "necessary costs". That means one lawyer. At standard rate. And only for what they did in the court room. That means what you pay the other side in legal costs is usually less than what you paid your lawyer. Though you pay other fees too and in total IIRC it usually amounts to about 150-200% of what you paid your lawyer after everything is said and done. If you are low income, the state can pay both lawyer fees, in fact, you can always request the state pay your lawyer, that's how we run the right to a lawyer to defend yourself (the state does not assign you one, you have to get one yourself).
This is entirely done so that the costs of a court case don't ruin someone's life forever. IIRC if you are low income, running a lengthy court proceeding won't be more expensive than a single lawyer in terms of the legal costs the court has you carry if you end up loosing. And even that depends because it's somewhat in the court to decide that, because we don't want to make people feel like they shouldn't sue for legitimate reasons.
Correct. I sued a party in Germany, won the case, got awarded 10K in legal costs (which actually was quite high, but it was a complete travesty) and another 25K in damages as well as annulment of whatever the suit was about.
Then the party I sued absconded to the US (he already had a house there and his family was already living there) and simply never returned.
The real legal bill was 50K. But at least the annulment stood and that was the main thrust of the suit.
Did you sue them as a company or as a person? Because AFAIK it is always a good idea to have lawyer insurance in Germany, however that would only work if you are a person, not a company.
As a company. Long story short, someone angled for years to have a partnership with us, finally got it and the day after the ink dried on the contract sold the rights to the source code to another company. That's not how that works...
In the US, the constitutional right to a lawyer only extends to defendants in criminal trials. Civil action, even you are sued by the government, is your responsibility regardless of income
Yeah in Germany that doesn't work. You have the right to a lawyer, even in civil action, even if you are suing, even if you are suing the government or the government is suing you. If you cannot afford a lawyer, the state pays it, you can pick your own lawyer. The argument is that if the state appoints a lawyer, the appointment is obviously not in your favor. Anyone should have the right to bring a slight or disagreement to court and anyone should have the right to get a lawyer to defend that.
I don't understand how this is not enough to deter patent trolls. Aren't they afraid companies (who tend to have deeper pockets than individuals anyway) will just fight the law suit and, y'know, win?
> Big corporation X can now run up lawyer cost that you have to pay if you lose
This is very much a consequence of the adversarial and very procedural (and costly) common law system as implemented in the US.
There's nothing stopping the US from passing new legislation to address exactly this downside.
The main issue is that the "big corp" can file seemingly endless motions to dismiss at any step (and then file for appeal, etc). But it seems perfectly possible to have a law that instructs the courts to by-default consider the merits of the other side without requiring that other side to also file whatever counter-motion to "yes, please keep the fucking suit going, because some random precedent from 1832 that google coughed up is almost surely bullshit, thanks y'honor"
Of course in many cases judges are elected. Which has its own very serious problems. [0]
The current workaround for the "big corp" problem is "class action", and they happen very frequently in the US. See also the small claims courts.
That said, the real problem is that the whole US system of torts/liability shifts a lot of responsibility on individuals, but in a way that's perfectly aligned with "mo' money = less worry" (Oh really? Who would have guessed.), because there are very few proactive, upfront, visible, legible pro-consumer regulations.
The most visible part is probably the endless EULA/ToS-style texts that nobody takes seriously [1] ... except courts. (See the famous case of a cruise ticket attached to a dissenting SCOTUS opinion https://en.wikipedia.org/wiki/Carnival_Cruise_Lines,_Inc._v.... )
US has fee shifting for frivolous patent litigations. for example, the NXIVM founder Keith Raniere paid out IBM's costs after he sued them for infringing one of his patents.
Automatic fee shifting has the opposite effect of discouraging anyone bringing forward a case that isn't a total slam dunk.
I frankly don't understand why a judge or jury would ever find for a company which has no revenue, other than through litigation settlements. I mean if they're not even selling a license or royalties for their patents, any cases should be dismissed.
We can call it the "use it or lose it" rule for patents.
Not every company that has no revenue other than through lawsuits got there by choice. Someone can copy your invention and sell it for less (or give it out for free), and you will be screwed.
There are lots of calls for exponentially escalating fees for patents over their life, and I think that is a much better idea than "use it or lose it."
I’m not convinced there’s any value in the patent system in the 21st century. Maybe it should be abolished. Who is it now helping? Certainly not the individual inventors it was meant to encourage.
Software patents should absolutely be prohibited altogether. They help no one; they’re not useful for small inventors and they’re just a pointless tax paid to patent trolls on innovation for everyone else. (I am a named inventor on many software patents.)
Patents have maintenance fees that have to be paid periodically, otherwise the invention becomes public domain early. They are only a few thousand dollars, so it is a no-brainer to pay them.
If they were exponential, it would be reasonable to see most patents have shorter lives, as patent owners would probably decide that it is not worth it to keep the patent around if they are not using it.
Also, most software patents are invalid. See the Alice Supreme Court case.
He's referring to the Eastern District of Texas which is infamous for patent trolls. A small number of judges find in favor of patent trolls far more often than the rest of the nation.
As the link notes, the Eastern District of Texas forum-shopping strategy for patent lawsuits was more or less shut down by the Supreme Court in a 2017 ruling [1]. This suit against comma.ai is, probably as a result, filed in the District of Delaware, since comma.ai is a Delaware corporation.
It’s not corrupt for a jury to follow what the law says—they swear an oath to do exactly that. And the law doesn’t have a carve-out saying you can only win if you have a product you’re making. (Although under current judge-made law, it is harder for a troll to get an injunction.)
If you don’t like the law, talk to Congress. Don’t blame the juries or the judges.
> I frankly don't understand why a judge or jury would ever find for a company which has no revenue, other than through litigation settlements
They don't. These guys all prey on settlements, the article even mentioned this troll has lost every single court battle and is surviving off of settlements.
Most people know that lawyers are expensive, but very, very few know just how expensive it is to go to trial. Even the simplest case can run into the hundreds of thousands of dollars.
Justice is only available for extremely rich folks.
I suspect this is more common than one might think especially for smaller companies that have enough cash/revenue to be collectable but too small to afford litigation or to take a principled stance like comma.ai plans to do. Ten years ago my business was a pioneer in smartphone/flip-phone enabled home-automation. We had barely been in business for a few months and doing ~$4k in monthly sales. A troll claimed he owned patent that covered using a phone to send a command to any electronic device/appliance. He eventually stopped calling me and his patent was for landline based cordless phones but I kept wondering if he would resurface years later once I had momentum but he never did.
Oddly I am now fighting a trademark dispute with a company that claims "non-metal shelves and storage" in their trademark. What that has to do with my home-automation smartphone app, who knows. This is all a reminder that we live in the most litigious country in the world.
I can't speak generally, but looking at the court filings for this suit, it is relating to techniques for retrofitting stuff onto vehicles, so I imagine merely the installation instructions for comma.ai would be enough to tell them what they need to know.
I don't really know though, my eyes glaze over when I read patents and think about patent law.
In prior history Hotz was targeted by Sony for his independent research into running homebrew applications on the PS3 game-console. Unfortunately Sony objected to this and launched legal action[1] against Hotz and others which sadly ended in Hotz settling to not research Sony products again and did not set a good precedent towards such work.
Different situation but hopefully better outcome this time...
Filing an invalid intellectual property claim should be considered fraud. Especially invalid patents with existing prior art and automated DMCA claims. Granting invalid patents should also be fraud. If a patent is granted despite the existence of prior art, whoever granted it has clearly caused damage to society.
"comma isn’t run by rational actors in suits sitting on a committee. It’s run by me, George Hotz. I’m willing to lose $1M before I give him $10k."
This stance is admirable but comma.ai is partly owned by investors. Isn't he obliged to find the lowest cost solution to this ? Or is George Hotz planning to spend his personal money on the lawsuit ?
comma was profitable pre-investment, so Geohot has a lot of leeway to do what he wants.
Besides, there is no fiduciary requirement to handle situations like this in any particular way. Fiduciary duty is to keep the business operating. Maximizing profit and minimizing costs are generally good practices; but they are not (legally) fiduciary duties.
It's easy to argue that spending $1M on deterrence once is cheaper than paying each patent troll who asks $10k.
It's also likely that this post is part of the defense strategy. If you were the patent troll, what would you do at this point - withdraw the claim and try again with an easier victim, or try to find out whether the "if this continues" part is true?
"If you were the patent troll, what would you do at this point ?"
If I could get the trial moved to a jurisdiction where I'm friends with a judge, which I recall was part of the strategy, I'd keep trolling. While geohot gets love on a hacker forum like HN, most courts (and people in general) as far as I can tell just assume guilt.
Courts don't appreciate having their time wasted. It's going to be a lot easier for a court to dismiss a frivolous patent troll than to "assume guilt" and go through the absolutely hellish process of extended litigation.
Pardon my ignorance, but why, then, is East Texas considered a hot-bed for these types of lawsuits? Are the judges there an exception to the rule about not wanting their time wasted?
Fucks ? It's one fuck, singular, over the past 6 years. There is this idea that an army of trolls are standing in attack formation and will swarm if he gives in this one time. But that claim seems silly.
geohot is rich and if he chooses to use those riches to end a stressful event and start sleeping well again, I wouldn't judge him one bit. Personally I'd rather be in the hospital that be in court.
317 comments
[ 2.7 ms ] story [ 163 ms ] thread"What he didn’t take into account is that comma isn’t run by rational actors in suits sitting on a committee. It’s run by me, George Hotz. I’m willing to lose $1M before I give him $10k. We will hire an amazing legal team, fight this, and while doing so invalidate his patents so they can’t be used against anyone else. Not because it’s rational, but because it’s the right thing to do. No patent troll will ever get a dollar from comma."
I find that relatable.
Most of the problems with software patents trace back, in one way or another, to that court.
What makes you think the current broken legislative system has a chance at a reform that makes things better in a way that you'd like?
That's a false premise.
You say you want regulatory/legal reform. Do you not think this starts with legislation? Do you not think that this requires action by the current legislative system? Do you think that the current legislative system is not broken?
Very importantly, patents are federal law. That means federal legislation. That means Congress. An institution that about 93% of Americans have fairly low confidence in. (See https://news.gallup.com/poll/394283/confidence-institutions-... for a source.)
It explains the economic reasons why it is hard to get collective action to happen. It also explains why any government organization organized for collective action in time is likely to undergo regulatory capture, meaning that the people you wanted to regulate are in control of the regulations and manipulate them for their own purposes. It also studies the various approaches that have been taken to solve this, and how effective they have or haven't been.
You'll learn WHY we don't simply all self-organize to act in our collective best interest.
So if the alternative is more self-motivated individual action, you obviously don't have to go far to find tons of examples of how that results in bad outcomes. Here's someone motivated to say "fuck you" to patent trolls. But the patent trolls themselves are motivated by self interest to be a troll in the first place, so "individual agency" here is at best a wash. Difficult collective action to reform the mechanisms patent trolls take advantage of still seem far more likely to work here than just hoping for more individuals willing to light their money on fire on principle.
We are surrounded by successful examples of collective action. But they also tend to be interestingly suboptimal in lots of ways. The book provides a framework through which you can understand why this happens.
The problem with public goods is that it is only in your personal interest to contribute to the extent that you personally reap the reward from your contribution. So, for example, catalytic converters cost about $1500. If you live in a city, you personally benefit more than $1500 in life quality from the fact that everyone around you has a catalytic converter. But if your catalytic converter goes, you don't benefit $1500 from YOUR catalytic converter. And so almost nobody would, without some other motivation, bother spending $1500 on a catalytic converter.
And yet, public goods do get provisioned. Here are the basic ways.
1. A single large entity benefits enough from the good to provision it for everyone else, who gets to free ride. The book calls this, the exploitation of the large by the small. If ever you see someone arguing that some big country or company should do something nice for everyone, they're hoping for this.
2. A small group combined can provision it. This usually results in complex negotiations, cheating, and so on as each tries to do the minimum necessary. The history of negotiations within OPEC on controlling the supply of oil is a good example.
3. A fundamentally coercive organization which exists for some other reason can ensure the provisioning of it. For my catalytic converter example, that is the government. If you fail to have a $1500 catalytic converter, the government will take steps up to and including putting you in jail.
4. People seem willing to put a small amount of energy out for a good cause. For example a lot of people are willing to put the energy out to vote, even though your personal vote is unlikely to swing any election in your lifetime. Some people are willing to put a lot of energy out.
5. The good may wind up provisioned for some other reason. For example open source software is a public good, which programmers voluntarily create and maintain for a wide variety of reasons. But very few programmers are creating it out of some notion of acting to maximize the common good.
But that is a very short take with very few examples. The book is quite readable and offers a lot more on all of this.
Collective action could align the incentives better, but leaders willing to risk some of the company's assets to do the right thing should, IMO, also be encouraged.
Rap Battle: https://www.youtube.com/watch?v=9iUvuaChDEg
When the rap battle didn't work, (probably) causing a $171M outage and getting away with it: https://en.wikipedia.org/wiki/2011_PlayStation_Network_outag...
Absolute legend.
Do you have more information on this? The Wikipedia page doesn't name Hotz.
Yes, deliciously so. This is all thoroughly speculative and had better remain that way.
https://www.newegg.com/insider/newegg-vs-patent-trolls-when-...
https://blog.cloudflare.com/project-jengo-redux-cloudflares-...
Also George Hotz fought and cowed Sony’s legal team when he was a kid in a saga involving ‘marcan42 of Asahi Linux. These patent trolls must be very foolish to try to intimidate him.
Edit: https://en.m.wikipedia.org/wiki/Sony_Computer_Entertainment_...
“In 2016, he ported Linux to the PlayStation 4 and demonstrated that at the 33rd Chaos Communication Congress by running Steam inside Linux.”
Considering that Sony’s primary claims were:
Violating the Digital Millennium Copyright Act (17 U.S.C. § 1201)
Violating the Computer Fraud and Abuse Act (18 U.S.C. § 1030(a)(2)(c))
Contributory copyright infringement (17 U.S.C. § 501)
Violating California Comprehensive Computer Data Access and Fraud Act (§ 502)
I’d say that the outcome failed to reflect Sony’s perspective on the situation and much more closely aligns with “George Hotz faced nearly no actual consequences”.
"Conspiracy to Circumvent Technological Measures and to Traffic in Circumvention Devices", and "Trafficking in Circumvention Devices" were two charges
He was, in fact, selling circumvention devices for the explicit purpose of enabling piracy.
According to Ars Technica the DOJ indictment stated:
* There was a screenshot of the SX OS menu with commercial games in it.
* A public changelog for SX OS publicly trumpeted the feature of "launching your Retro Game ROMs for emulators directly from gDrive,"
* They were "attempt[ing] to cloak its illegal activity with a purported desire to support homebrew enthusiasts who wanted to design their own games."
* the product description for the team's SX devices notes that you can "play all your favorite game backups straight off of the microSD card (or external USB storage!)"
So, yeah, I agree they were selling circumvention devices with at least some level of piracy intent (and that was used to send him to prison after all.) But I think you can also agree the delta between what they were doing and what geohotz did with the PS3 is not very large.
It was absolutely massive. geohots jailbreak did not enable piracy.
It’s like the delta between selling kratom and heroin is not very large, both are drugs that will get you high.
That hack almost certainly had nothing to do with the jailbreaking scene. Sony pulled down their infrastructure to investigate an extensive compromise.
[1] https://www.youtube.com/watch?v=9iUvuaChDEg
I'm torn between considering it a power move of mentioning yourself and linking to the Wikipedia entry about you, or not. But I appreciate the troll hunt.
For patent trolls, time spent on legal is mostly opportunity cost, but killing their patent portfolio has a real dollar cost associated to it.
By turning the former into the latter, it should make people think twice. But then people aren't necessarily rational.
Don't mess with Geohot.
Me too - nobody would accuse me of being a rational actor.
It's almost like all the kids in school who failed out of CS 101 went to law school and cooked up a scheme to tax their classmates who actually stuck it out and built something.
Why the hell do we have a system that gives a 20 year monopoly to assholes for shit they cook up on a cocktail napkin over lunch with their lawyer? This is how we used to brainstorm stuff to patent. It's utter bullshit. I refuse to participate in creating more bullshit patents.
I think the only reasonable criticisms of the patent system is a lack of intelligent lawyers in the areas of discipline that often fail finding similar claims in other filings, and that patents only benefit large companies, not DIY garage makers like the rest of the world thinks.
I also think your shitting on lawyers is typical of people that don't really understand legal frameworks. Would you find it reasonable if I said: "software engineers are such a joke, everything they do is so over complicated just so they can get rich like those SanFran dudebros, and for what, so that I can shit in a smart toilet with a web browser and AI? How stupid." ... Oh, shit. Wait a second... :)
One of the patents is here: https://portal.unifiedpatents.com/patents/patent/10454707
Method, Apparatus and System for Retrofitting a Vehicle. This does in fact describe what Comma does. Looks like it might be infringing!
Now Comma must prove this patent cannot be valid.
We write open source ADAS software, and it doesn't have any functionality like that: https://github.com/commaai/openpilot
But for the purposes of attempting to extract a settlement, it doesn't matter. It will only matter after a long and expensive trial.
(I would like to invalidate the patent entirely, which several of their patents already have been. But even if it is valid, it also has to apply to us)
Dataspeed did appear to have tried to invalidate this patent in 2019 (see the IPR documents linked there), and succeeded on all but claim 20. If your device does not have a relay which shuts off its electrical connection within the car, you're probably safe. You could try to invalidate it, but you should read the docs from the IPR to see what Dataspeed did and what you might need to do.
My gut-check guess is that claim 20 is probably not going to get invalidated unless you can find some weird after-market car part from before 2007 which does a safety shut-off of a car component.
Maybe one of the competitors used an existing (i.e. not custom made) car part for this?
———
Found this from 2004:
“A DARPA-provided electronic stop device allowed a remote control shut-down of the vehicle in case of an emergency”
Is that a lead?
[0] https://www.researchgate.net/profile/Reinhold-Behringer/publ...
Does the vehicle have (1) an after-market device installed that (2) detects that a first component stopped sending commands to a second component and (3) uses relay switch to disconnect the failed first component from the second component and (4) starts sending commands to the second component on its own that are indistinguishable from those sent by the first component?
Technically, the above requirements are needed against claim 20 of '707. Claim 6 requires fewer features. That doesn't help us because: (1) claim 20 is asserted in the complaint [0], and (2) claim 20 was upheld in a previous challenge to its validity (an "IPR"), so it's got some staying power.
[0] https://blog.comma.ai/img/20220824085232055.pdf
Let's take '707 claim 6. _Any_ commercial product with each and every one of the following bullet points ("limitations" in lawyer-speak) infringes and must purchase a license to this patent.
[1] A vehicle, comprising: [2] a factory-installed first apparatus configured to generate an electrical signal; [3] a factory-installed second apparatus configured to receive the electrical signal; and [4] a retrofit apparatus electrically connected to the factory-installed second apparatus, [5] wherein the retrofit apparatus generates a mimicked electrical signal independently of the electrical signal generated by the factory-installed first apparatus, and [6] wherein the factory-installed second apparatus receives the mimicked electrical signal.
And at any rate, "illegal" is not the same as "immoral". Even if the troll's actions are within the law, in my opinion they're utterly immoral and scummy.
I like that statement. "I'll be the devil's advocate" is right up there with everyone stating they will defend hate speech "to the death" because of free speech. Why not take your time to say you disagree with someone instead?
Is this because the devil is a qualified attorney himself?
The process -- your "day in court" (years of fees and stress) -- is the punishment, turning your "win" into only a Pyrrhic victory.
Also, trolls rarely are the ones creating these patents. They just buy up some basket of idle patents, then spend their morning looking for any company announcing anything that has a couple words in common with something in their 'portfolio'. They send a boilerplate royalty request or whatever. Rinse and repeat.
If you cut through the lawyer-speak, claim #1 is basically about MITM-ing the vehicle's internal message bus. Hardly novel, even in 2007.
If patent is invalid the way to prove that is to challenge that in court.
That system, that makes filling patents easy, is a whole reason why patent trolls exists. But it’s not a fault of patent offices, that they’re mandated to work that way.
What I would like to see is a deposition of the patent examiner who approved this ridiculous patent.
In this case it was granted with largely similar claims (as US10027505B2). However, claims typically get significantly revised by the time they are granted, and less than half of applications are granted at all.
https://patents.google.com/patent/US10027505B2/en
Just this reduces frivolous lawsuits. If you have a good case, sure, go ahead. If your case is weak, you run a real risk paying not only for your lawyers but the other side as well.
(Note that there are disadvantages to loser-pays as well. Image you legitimately want to sue corporation X because they did something atrocious. Big corporation X can now run up lawyer cost that you have to pay if you lose.)
Edit: Sorry I was wrong about England (which leaves the US only). Edit 2: In the US you can file for your lawyer fees to be returned, but that in itself is risk (I have some lawyer friends)
https://www.ashurst.com/en/news-and-insights/legal-updates/q...
What's unique about the UK is that unlike actually civilized countries (incl. the US), there is no legal requirement to negotiate in good faith. It is expected for you to do so, but if the other side figures out that you haven't and tries to take you to court over that, then the English court system will say "tough luck". Manipulating the other side to your advantage should be expected, they said. There have been High Court cases that have established this principle. so doing business with the English is always possibly problematic (e.g. Brexit and all the lies, the mentality behind all that etc). Anecdotally, I have heard from foreigners that doing business with English people is more of a headache than with ones from other countries. And I have lived here long enough to 'understand' the English better than most Americans...
I needed a good laugh today, so thanks.
This seems to reduce frivolous lawsuits (depending on who is suing/being sued) but you can ask for a cost capping order to limit the amount you have to pay out if you lose, which partially helps address the issue of "they're too big/expensive to sue".
The party making the offer can make a part 36 offer (or state equivalent if it exists), which grants automatic presumptions of cost sharing if rejected and the case ends with a more favorable (to the offering party) settlement or reward.
They can do a less formal without prejudice save as to costs offer, which is one the courts can use to decide cost charging, but does not get automatic presumptions of how the sharing should work.
Lastly they can make just plain without prejudice settlement offers. Which won’t factor into costs at all.
Which type of offer to make depends on who (claimant or defendant) is making the offer, and how certain they are about the amount that will be awarded at trial.
For example if a defendant is making an offer that they are absolutely convinced is more than will be awarded at trial (with the idea being that it would cost less than doing the trial without any offer, and having the normal each party pays own costs rule apply), it makes sense to make it a part 36 offer.
However when making a lowball offer, or a reasonable one but that could well be less than what gets awarded at trial, a without prejudice offer is probably better.
On the claimants side, when making a offer that they are convinced is below what they will actually win it may make sense to make it a Part 36 offer. If making an offer closer to or above what they will probably actually win, a without prejudice offer could be more sensible.
The meaning they're going for is "by making this offer I'm not admitting guilt, but when we come to awarding costs the fact that I tried to end this without going to court should count in my favour".
https://en.m.wikipedia.org/wiki/American_rule_(attorney%27s_...
This is entirely done so that the costs of a court case don't ruin someone's life forever. IIRC if you are low income, running a lengthy court proceeding won't be more expensive than a single lawyer in terms of the legal costs the court has you carry if you end up loosing. And even that depends because it's somewhat in the court to decide that, because we don't want to make people feel like they shouldn't sue for legitimate reasons.
Then the party I sued absconded to the US (he already had a house there and his family was already living there) and simply never returned.
The real legal bill was 50K. But at least the annulment stood and that was the main thrust of the suit.
This is very much a consequence of the adversarial and very procedural (and costly) common law system as implemented in the US.
There's nothing stopping the US from passing new legislation to address exactly this downside.
The main issue is that the "big corp" can file seemingly endless motions to dismiss at any step (and then file for appeal, etc). But it seems perfectly possible to have a law that instructs the courts to by-default consider the merits of the other side without requiring that other side to also file whatever counter-motion to "yes, please keep the fucking suit going, because some random precedent from 1832 that google coughed up is almost surely bullshit, thanks y'honor"
Of course in many cases judges are elected. Which has its own very serious problems. [0]
The current workaround for the "big corp" problem is "class action", and they happen very frequently in the US. See also the small claims courts.
That said, the real problem is that the whole US system of torts/liability shifts a lot of responsibility on individuals, but in a way that's perfectly aligned with "mo' money = less worry" (Oh really? Who would have guessed.), because there are very few proactive, upfront, visible, legible pro-consumer regulations.
The most visible part is probably the endless EULA/ToS-style texts that nobody takes seriously [1] ... except courts. (See the famous case of a cruise ticket attached to a dissenting SCOTUS opinion https://en.wikipedia.org/wiki/Carnival_Cruise_Lines,_Inc._v.... )
[0] https://www.propublica.org/article/these-judges-can-have-les...
[1] https://www.abajournal.com/news/article/chief_justice_robert...
Automatic fee shifting has the opposite effect of discouraging anyone bringing forward a case that isn't a total slam dunk.
We can call it the "use it or lose it" rule for patents.
There are lots of calls for exponentially escalating fees for patents over their life, and I think that is a much better idea than "use it or lose it."
I’m not convinced there’s any value in the patent system in the 21st century. Maybe it should be abolished. Who is it now helping? Certainly not the individual inventors it was meant to encourage.
Software patents should absolutely be prohibited altogether. They help no one; they’re not useful for small inventors and they’re just a pointless tax paid to patent trolls on innovation for everyone else. (I am a named inventor on many software patents.)
If they were exponential, it would be reasonable to see most patents have shorter lives, as patent owners would probably decide that it is not worth it to keep the patent around if they are not using it.
Also, most software patents are invalid. See the Alice Supreme Court case.
https://en.wikipedia.org/wiki/United_States_District_Court_f...
[1] https://en.wikipedia.org/wiki/TC_Heartland_LLC_v._Kraft_Food...
If you don’t like the law, talk to Congress. Don’t blame the juries or the judges.
They don't. These guys all prey on settlements, the article even mentioned this troll has lost every single court battle and is surviving off of settlements.
Justice is only available for extremely rich folks.
Oddly I am now fighting a trademark dispute with a company that claims "non-metal shelves and storage" in their trademark. What that has to do with my home-automation smartphone app, who knows. This is all a reminder that we live in the most litigious country in the world.
How do they figure out whether stuff is truly infringing a patent in the first place? Surely some corporate espionage is required to figure this out?
I don't really know though, my eyes glaze over when I read patents and think about patent law.
Different situation but hopefully better outcome this time...
[1] https://en.wikipedia.org/wiki/Sony_Computer_Entertainment_Am...
This stance is admirable but comma.ai is partly owned by investors. Isn't he obliged to find the lowest cost solution to this ? Or is George Hotz planning to spend his personal money on the lawsuit ?
Besides, there is no fiduciary requirement to handle situations like this in any particular way. Fiduciary duty is to keep the business operating. Maximizing profit and minimizing costs are generally good practices; but they are not (legally) fiduciary duties.
They released new HW, that gives them lower margins and volume didn’t pickup as much as they planned.
I’m guessing this stance will be popular among engineers who will hear about the company and apply for jobs.
It's also likely that this post is part of the defense strategy. If you were the patent troll, what would you do at this point - withdraw the claim and try again with an easier victim, or try to find out whether the "if this continues" part is true?
If I could get the trial moved to a jurisdiction where I'm friends with a judge, which I recall was part of the strategy, I'd keep trolling. While geohot gets love on a hacker forum like HN, most courts (and people in general) as far as I can tell just assume guilt.
(Source: Worked as a litigator for 7 years.)
If you invest in a geohot-owned business, you should probably know who you're bankrolling. This type of behavior by him is hardly surprising.
geohot is rich and if he chooses to use those riches to end a stressful event and start sleeping well again, I wouldn't judge him one bit. Personally I'd rather be in the hospital that be in court.