The plaintiffs would have to demonstrate how the company lost value by them doing that.
Also, if the articles of incorporation or the IPO documentation includes a statement indicating that increasing the use of electric vehicles is a corporate goal, alongside of, or even more important than, making money, then this move could be perfectly aligned with the stated corporate goals which the investors bought into, even if it can be proven that this move decreased share value.
There is a middle ground. They could issue a zero-cost, binding, globally-applicable patent license with an exception that withdraws the license from anyone who sues them. This is actually pretty common, and I think it would be a much better choice than placing the patents in the public domain.
I expect something like that will be forthcoming. So far all we have is a blog post; I imagine the more substantive version from the legal department is in progress.
If I were personally going to use one of Tesla's patents in my business, I'd want a signed zero-cost GPL-like license agreement with Tesla. For example, Musk's good will is nice, but what if someone else were to acquire Tesla's IP?
It wouldn't matter, I don't think. A clearer statement than "we took our plaques off our wall" is needed, but assuming there is a clear statement from Tesla that they will only use these patents defensively, anyone who takes them at their word should be safe.
Why? There's a legal concept called "promissory estoppel". In a nutshell, it means that if I make you a promise and you, in good faith, depend upon that promise and build your business on it, and I knew or should have known that you were going to do so, then I can't later change my mind, withdraw my promise and sue you for doing what I said you could do.
Most of the Internet-related patents are so utterly obvious the patents should never have been issued. The reason companies don't want to license them is because thay add no value: The companies' engineers independently reinvented all of those trivially-obvious inventions, and now the patent holders are trying to hold them up for a lot of money. This is evidence that the system is badly broken.
The problem is that the situation you describe fits about 0.1% of patents, or less. The dream of the lone inventor making it big with the help of patent law isn't a fantasy, but it's so rare that it might as well be. In the meantime the current patent law structure serves mostly to impede technological progress and enrich patent attorneys.
To be precise, this ruling established a binding precedent in the 11th circuit and a persuasive precedent elsewhere in the country, correct?
My understanding is that given a binding precedent a circuit judge must explain why the precedent does not apply to the facts in order to rule contrary, and that given a persuasive precedent the judge merely needs to explain (in some detail) why the precedent is in error. Is that right?
If people start losing their driver's licenses when they're caught doing commercial driving without being properly insured, I would guess fewer of them will take the risk.
Uber provides a $1M liability policy, so they are properly insured.
Update: As others have pointed out, Uber already addresses the question of liability coverage by providing the coverage itself... so the potential problem arising from drivers not having coverage is already handled via another mechanism, making the law irrelevant (not inapplicable in a legal sense, just irrelevant in the sense that it's not actually accomplishing anything).
If Uber were really offering legitimate competition, I would be more sympathetic. But they're partly undercutting existing taxis through ridiculous things like using drivers who lack commercial vehicle insurance, which is rather irresponsible.
Is it?
The fact that commercial vehicles must have said insurance by law doesn't mean it's actually necessary even for them. In many cases, those sorts of legal requirements are imposed not because they're actually necessary per-se, but because they establish obstacles to entry of service providers into the market, and aren't obviously arbitrary. They are arbitrary, just not obviously arbitrary.
In a market where information is highly asymmetric, establishing arbitrary obstacles to entry is actually a useful thing to do, because the mere act of bothering to clear the hurdles demonstrates a high probability of good faith on the part of the company trying to enter. Individuals or organizations trying to brand themselves as cabs, for example, in order to more easily rob unsuspecting patrons will wish to be able to establish themselves quickly and then disappear quickly.
The Internet, particularly the mobile Internet, however, breaks down information asymmetries. Uber uses different mechanisms that have much lower friction to establish trust in its drivers, eliminating the need for arbitrary obstacles. Do Uber's mechanisms work? I don't know. They appear to, and time will tell whether or not something more is required to make the service safe and effective.
This then means we have to revisit the basis for requiring commercial vehicle insurance. Since it's no longer required as an arbitrary (but not obviously arbitrary) obstacle, is it actually required? What problem does it solve that's not solved some other way?
FYI, I suspect I know what the problem is, and I think it probably is a real problem. My point is that all of the assumptions about what is necessary for ordinary commercial operations do have to be re-evaluated because in the new context they may not make sense, even if this one does. We shouldn't just blindly apply the old rules, but should instead see what problems arise and apply the rules that make sense to resolve the problems.
What bizarre reasoning. Looking at total violence rather than violence committed with one specific tool is "moving the goalposts"? Nonsense, it's looking at the whole picture. Eliminating gun violence is pointless if it's just replaced by the same amount of violence using other tools.
Narrowing the focus to consider only gun violence constitutes brazen moving of the goalposts. Shifting them sideways a few inches and then ignoring all of the goals that pass through the region where the goalposts previously weren't. The other team doesn't actually score any less, but we ignore a portion of the goals and call it a victory?
The only reasonable and valid goal is to reduce violence, period. If it could be proved that restricting guns reduced violence, then that would be a valid argument for restricting guns. But if restricting guns merely causes people to choose different tools, with the same results, then there is no case for restricting guns.
cool story bro, CBC is broken. Truecrypt uses XTS, and TLS added GCM.
CBC is not broken. It doesn't provide the authentication properties an AEAD mode like GCM does, and it's more subject to ciphertext tampering attacks than XTS, but it's a perfectly good mode when applied with understanding of its strengths and weaknesses -- which is also true of GCM (which is terribly insecure if tags are truncated too much; far worse than CBC) and XTS (which isn't authenticated and therefore still subject to ciphertext tampering). And if you want CBC to have authentication and tamper-resistance, they can easily be added by HMACing the ciphertext.
I imagine the problem is that these databases only hold collected data for a short period of time, say 30 or 90 days. The data scraped is massive, so it is constantly deleting old data to make way for more. IAA Intelligence Analyst, and I know of some imagery databases, for example, that only hold the last 30 days of imagery. If you forced them to hold all of it for years, it would mean increasing server space by orders of magnitude.
It's based on the rate that the owning entity (typically a university or the DoE) charges for time on it, which is (at least in theory) correlated with the machine's electric, AC (i.e. more electric) and ISP bills, the hardware manufacturer service and support contracts, and the price of the highly paid technicians and sysadmins who keep it working. All of these are nontrivial.
OTOH, the cost of central computer time is systematically overestimated, often by multiple orders of magnitude, and no one really pays the inflated costs.
25 years ago I was doing a lot of calculations on my campus mainframe, generating pretty pictures of the Mandelbrot and Julia sets. After doing this for a few weeks, I was confronted by the administration with a bill for over $30,000. After they let me freak out for a few seconds they said really it was no big deal, that the usage costs were just for internal accounting purposes (though they were based on the cost of the machine plus power, maintenance, etc.), but they needed me to stop slowing down other jobs. We agreed that I could run my jobs during the evening, at idle priority (during the day there was supposedly some impact on other jobs, even at idle priority, which still doesn't make any sense to me). From then on I occasionally looked at the bills for my usage. They eventually topped $100K, but it was all funny money.
I've talked to a lot of people in organizations with big central computing infrastructures and without fail heard the same story. There is a cost accounting for CPU time, but it's insanely high, and not real -- even though it really is related to the cost of the machine, nearly all of it is sunk cost, and often there's significant unused capacity just sitting idle at least some of the time. The marginal cost for using that idle capacity is low, but gets accounted at the full rate.
So... it's entirely likely that this $150K is quite a bit higher than the real cost of his mining.
The only reasonable response to mass CCTV is for everyone to wear a balaclava.
That's silly.
People willing to wear balaclavas to avoid being tracked are clearly willing to take the much less radical step of voting for and donating to the campaigns of politicians who oppose mass CCTV coverage. If you get a sufficiently large segment of the population willing to wear balaclavas that the CCTV system is useless, then you also have enough public opposition to CCTV cameras to remove them via the political process.
The truth of the matter is that most of the population doesn't care, and a substantial portion of the population thinks the cameras are a good thing. Changing those attitudes is what will eliminate CCTV, and if CCTV opponents start wearing masks in public they'll just be marginalizing themselves.
Honestly, I think it's a losing fight. Mass surveillance (and sousveillance; the same thing but from the bottom rather than the top) are going to be core features of future society; we'll adapt. Or maybe we'll get a massive public backlash against it... but I'm not holding my breath.
Er, yes, I meant to say that the US rate was 9X and is now 4X the UK's rate, not vice versa.
I agree that these things are hard to get a handle on. So hard that your claim of gun restrictions reducing violence in the UK is as unsupportable as the other comments in the thread.
Regarding guns and violence, I'm very interested in the topic (my personal opinion is that more normal, sane people with guns reduces violent crime, but that's a belief I can't support with statistics) so I follow all of the studies. So far it seems that every good study concludes that if there is an effect of gun control it's small to imperceptible, and that there's certainly no clear causal link between gun control and reduced violence.
Uh, no. The UK has 2000% and 800% higher property and violent crime than the US, respectively. The only thing they have the US beat on is murder, which is a bit of a special case as usually the victim and perpetrator know each other.
You're right that I was looking just at homicide rates. However, that's largely because it's really hard to compare other crimes. Classifications are so different -- and in particular the UK classifies a lot of things as violent crimes that the US does not -- that it's hard to compare meaningfully. Reporting rates also have a huge effect. Murder is nice in that the definition and reporting are both very consistent throughout the developed world.
So, I'm not disagreeing with you, just pointing out that you need to take those numbers with a grain of salt. The UK may have a higher overall violent crime rate than the US, but it's lower (per US definitions) than the 8X you cite, and could even be lower than the US, though that seems unlikely.
serious gun control laws work, as have been demonstrated by the countries in which they have been instituted
Do you have any evidence for that? I don't think it's true. As far as I can tell there are no countries that enacted tight gun restrictions and then had a significant decline in violence which there's reason to believe would not have happened anyway, per comparison with similar countries without restrictions. Merely observing restrictions followed by decrease in violence doesn't prove the point; if post hoc ergo propter hoc were valid, the significant relaxations in gun control in the US, followed by decreasing violence, would prove the opposite.
Slashdot Mirror
The plaintiffs would have to demonstrate how the company lost value by them doing that.
Also, if the articles of incorporation or the IPO documentation includes a statement indicating that increasing the use of electric vehicles is a corporate goal, alongside of, or even more important than, making money, then this move could be perfectly aligned with the stated corporate goals which the investors bought into, even if it can be proven that this move decreased share value.
There is a middle ground. They could issue a zero-cost, binding, globally-applicable patent license with an exception that withdraws the license from anyone who sues them. This is actually pretty common, and I think it would be a much better choice than placing the patents in the public domain.
I expect something like that will be forthcoming. So far all we have is a blog post; I imagine the more substantive version from the legal department is in progress.
Target shooting, competitive shooting, hunting and self-defense are other legal uses of handguns.
If I were personally going to use one of Tesla's patents in my business, I'd want a signed zero-cost GPL-like license agreement with Tesla. For example, Musk's good will is nice, but what if someone else were to acquire Tesla's IP?
It wouldn't matter, I don't think. A clearer statement than "we took our plaques off our wall" is needed, but assuming there is a clear statement from Tesla that they will only use these patents defensively, anyone who takes them at their word should be safe.
Why? There's a legal concept called "promissory estoppel". In a nutshell, it means that if I make you a promise and you, in good faith, depend upon that promise and build your business on it, and I knew or should have known that you were going to do so, then I can't later change my mind, withdraw my promise and sue you for doing what I said you could do.
I disagree.
Most of the Internet-related patents are so utterly obvious the patents should never have been issued. The reason companies don't want to license them is because thay add no value: The companies' engineers independently reinvented all of those trivially-obvious inventions, and now the patent holders are trying to hold them up for a lot of money. This is evidence that the system is badly broken.
The problem is that the situation you describe fits about 0.1% of patents, or less. The dream of the lone inventor making it big with the help of patent law isn't a fantasy, but it's so rare that it might as well be. In the meantime the current patent law structure serves mostly to impede technological progress and enrich patent attorneys.
It's called the influence of money on politics.
There is money fighting for both sides. Much of the tech industry is fighting for patent reform.
Translation: Google is about to introduce yet another web language into this Tower of Babel
Better translation: Some engineer who works for Google and is interested in web programming languages wants to create a new one.
To be precise, this ruling established a binding precedent in the 11th circuit and a persuasive precedent elsewhere in the country, correct?
My understanding is that given a binding precedent a circuit judge must explain why the precedent does not apply to the facts in order to rule contrary, and that given a persuasive precedent the judge merely needs to explain (in some detail) why the precedent is in error. Is that right?
Did you actually read that waiver? It doesn't waive liability arising due to accidents. It's all about not promising any service level guarantees.
Uber does not provide that level of insurance. It provides only insurance to the passenger, not the driver or third parties
No. Uber provides coverage up to $1M per accident in automobile liability insurance. So any liability arising from the auto accident is covered.
http://blog.uber.com/ridesharinginsurance
http://blogcdn.uber.com/wp-content/uploads/2014/02/commercial-insurance-policy.pdf
Or maybe they just decided it was the easiest way to make that particular issue go away.
If people start losing their driver's licenses when they're caught doing commercial driving without being properly insured, I would guess fewer of them will take the risk.
Uber provides a $1M liability policy, so they are properly insured.
Yes, that's the issue I suspected. However, Uber has already addressed it.
Update: As others have pointed out, Uber already addresses the question of liability coverage by providing the coverage itself... so the potential problem arising from drivers not having coverage is already handled via another mechanism, making the law irrelevant (not inapplicable in a legal sense, just irrelevant in the sense that it's not actually accomplishing anything).
If Uber were really offering legitimate competition, I would be more sympathetic. But they're partly undercutting existing taxis through ridiculous things like using drivers who lack commercial vehicle insurance, which is rather irresponsible.
Is it?
The fact that commercial vehicles must have said insurance by law doesn't mean it's actually necessary even for them. In many cases, those sorts of legal requirements are imposed not because they're actually necessary per-se, but because they establish obstacles to entry of service providers into the market, and aren't obviously arbitrary. They are arbitrary, just not obviously arbitrary.
In a market where information is highly asymmetric, establishing arbitrary obstacles to entry is actually a useful thing to do, because the mere act of bothering to clear the hurdles demonstrates a high probability of good faith on the part of the company trying to enter. Individuals or organizations trying to brand themselves as cabs, for example, in order to more easily rob unsuspecting patrons will wish to be able to establish themselves quickly and then disappear quickly.
The Internet, particularly the mobile Internet, however, breaks down information asymmetries. Uber uses different mechanisms that have much lower friction to establish trust in its drivers, eliminating the need for arbitrary obstacles. Do Uber's mechanisms work? I don't know. They appear to, and time will tell whether or not something more is required to make the service safe and effective.
This then means we have to revisit the basis for requiring commercial vehicle insurance. Since it's no longer required as an arbitrary (but not obviously arbitrary) obstacle, is it actually required? What problem does it solve that's not solved some other way?
FYI, I suspect I know what the problem is, and I think it probably is a real problem. My point is that all of the assumptions about what is necessary for ordinary commercial operations do have to be re-evaluated because in the new context they may not make sense, even if this one does. We shouldn't just blindly apply the old rules, but should instead see what problems arise and apply the rules that make sense to resolve the problems.
What bizarre reasoning. Looking at total violence rather than violence committed with one specific tool is "moving the goalposts"? Nonsense, it's looking at the whole picture. Eliminating gun violence is pointless if it's just replaced by the same amount of violence using other tools.
Narrowing the focus to consider only gun violence constitutes brazen moving of the goalposts. Shifting them sideways a few inches and then ignoring all of the goals that pass through the region where the goalposts previously weren't. The other team doesn't actually score any less, but we ignore a portion of the goals and call it a victory?
The only reasonable and valid goal is to reduce violence, period. If it could be proved that restricting guns reduced violence, then that would be a valid argument for restricting guns. But if restricting guns merely causes people to choose different tools, with the same results, then there is no case for restricting guns.
cool story bro, CBC is broken. Truecrypt uses XTS, and TLS added GCM.
CBC is not broken. It doesn't provide the authentication properties an AEAD mode like GCM does, and it's more subject to ciphertext tampering attacks than XTS, but it's a perfectly good mode when applied with understanding of its strengths and weaknesses -- which is also true of GCM (which is terribly insecure if tags are truncated too much; far worse than CBC) and XTS (which isn't authenticated and therefore still subject to ciphertext tampering). And if you want CBC to have authentication and tamper-resistance, they can easily be added by HMACing the ciphertext.
I imagine the problem is that these databases only hold collected data for a short period of time, say 30 or 90 days. The data scraped is massive, so it is constantly deleting old data to make way for more. IAA Intelligence Analyst, and I know of some imagery databases, for example, that only hold the last 30 days of imagery. If you forced them to hold all of it for years, it would mean increasing server space by orders of magnitude.
None of which in any way impacts the GP's point.
It's based on the rate that the owning entity (typically a university or the DoE) charges for time on it, which is (at least in theory) correlated with the machine's electric, AC (i.e. more electric) and ISP bills, the hardware manufacturer service and support contracts, and the price of the highly paid technicians and sysadmins who keep it working. All of these are nontrivial.
OTOH, the cost of central computer time is systematically overestimated, often by multiple orders of magnitude, and no one really pays the inflated costs.
25 years ago I was doing a lot of calculations on my campus mainframe, generating pretty pictures of the Mandelbrot and Julia sets. After doing this for a few weeks, I was confronted by the administration with a bill for over $30,000. After they let me freak out for a few seconds they said really it was no big deal, that the usage costs were just for internal accounting purposes (though they were based on the cost of the machine plus power, maintenance, etc.), but they needed me to stop slowing down other jobs. We agreed that I could run my jobs during the evening, at idle priority (during the day there was supposedly some impact on other jobs, even at idle priority, which still doesn't make any sense to me). From then on I occasionally looked at the bills for my usage. They eventually topped $100K, but it was all funny money.
I've talked to a lot of people in organizations with big central computing infrastructures and without fail heard the same story. There is a cost accounting for CPU time, but it's insanely high, and not real -- even though it really is related to the cost of the machine, nearly all of it is sunk cost, and often there's significant unused capacity just sitting idle at least some of the time. The marginal cost for using that idle capacity is low, but gets accounted at the full rate.
So... it's entirely likely that this $150K is quite a bit higher than the real cost of his mining.
The only reasonable response to mass CCTV is for everyone to wear a balaclava.
That's silly.
People willing to wear balaclavas to avoid being tracked are clearly willing to take the much less radical step of voting for and donating to the campaigns of politicians who oppose mass CCTV coverage. If you get a sufficiently large segment of the population willing to wear balaclavas that the CCTV system is useless, then you also have enough public opposition to CCTV cameras to remove them via the political process.
The truth of the matter is that most of the population doesn't care, and a substantial portion of the population thinks the cameras are a good thing. Changing those attitudes is what will eliminate CCTV, and if CCTV opponents start wearing masks in public they'll just be marginalizing themselves.
Honestly, I think it's a losing fight. Mass surveillance (and sousveillance; the same thing but from the bottom rather than the top) are going to be core features of future society; we'll adapt. Or maybe we'll get a massive public backlash against it... but I'm not holding my breath.
Er, yes, I meant to say that the US rate was 9X and is now 4X the UK's rate, not vice versa.
I agree that these things are hard to get a handle on. So hard that your claim of gun restrictions reducing violence in the UK is as unsupportable as the other comments in the thread.
Regarding guns and violence, I'm very interested in the topic (my personal opinion is that more normal, sane people with guns reduces violent crime, but that's a belief I can't support with statistics) so I follow all of the studies. So far it seems that every good study concludes that if there is an effect of gun control it's small to imperceptible, and that there's certainly no clear causal link between gun control and reduced violence.
Uh, no. The UK has 2000% and 800% higher property and violent crime than the US, respectively. The only thing they have the US beat on is murder, which is a bit of a special case as usually the victim and perpetrator know each other.
You're right that I was looking just at homicide rates. However, that's largely because it's really hard to compare other crimes. Classifications are so different -- and in particular the UK classifies a lot of things as violent crimes that the US does not -- that it's hard to compare meaningfully. Reporting rates also have a huge effect. Murder is nice in that the definition and reporting are both very consistent throughout the developed world.
So, I'm not disagreeing with you, just pointing out that you need to take those numbers with a grain of salt. The UK may have a higher overall violent crime rate than the US, but it's lower (per US definitions) than the 8X you cite, and could even be lower than the US, though that seems unlikely.
Did you seriously just cite a TV show?
It actually seems like a reasonable citation to me, demonstrating that Americans found the idea of fire department EMTs to be normal.
serious gun control laws work, as have been demonstrated by the countries in which they have been instituted
Do you have any evidence for that? I don't think it's true. As far as I can tell there are no countries that enacted tight gun restrictions and then had a significant decline in violence which there's reason to believe would not have happened anyway, per comparison with similar countries without restrictions. Merely observing restrictions followed by decrease in violence doesn't prove the point; if post hoc ergo propter hoc were valid, the significant relaxations in gun control in the US, followed by decreasing violence, would prove the opposite.