Scaled Composites is probably close to the point where they look like they can make a very real change to the expectations of the market [...]
The simpler explanation (thank you Occam) is that Scaled Composites has created a completely new market out of thin air, and Northrop wants a bigger piece of the action.
That's a reasonable question, and you're right, it's only tangentially related. But there's a principle here.
In both our countries, we're supposed to be ruled by law, not humans. Legislatures are good at passing overbroad laws with the assurance that they're intended only to have a limited effect. We're only going to use this anti-terrorist legislation against terrorists, right? But humans apply them to other kinds of criminal, or to some guy who gave an old SIM card that he wasn't going to use any more to someone.
That's why you should strongly resist any piece of law (be it legislation or executive order) which is worded broadly. Because it will get misused.
And here, ladies and gentlemen, we have the proverbial "other side" of the "letting the government do whatever the hell it wants" coin. We call this other side of the coin, the "legislation by public opinion" side of the coin.
No, that's not it at all. My point is twofold.
One is that civil liberties do you no good if you're dead. More on that below.
The other (and I didn't state this explicitly) is that it's not enough for you to believe in your civil liberties. Other people have to, too. I'm waiting for the day when the NRA stands up for the rights of an unpopular group, like the victims of the "War on Drugs". It won't happen, because the NRA doesn't believe in civil rights. They believe in exactly one civil right.
Men who have been wrongly incarcerated or had all their belongings "frozen" without due process have no civil rights either.
If they're alive, they have a chance to bring those responsible to justice.
It's not good enough for the purposes of this order to be engaged in transactions with people in Iraq--it also requires that the purpose of those transactions is to either directly or indirectly support violence in Iraq.
Are you so sure? Let's see that quote again...
... to have materially assisted, sponsored, or provided financial, material, logistical, or technical support for, or goods or services in support of, such an act or acts of violence or any person whose property and interests in property are blocked pursuant to this order; or...
I don't know if you've been keeping up with what's happening in Australia at the moment, but we have a man currently being held (though that's a story in itself) on a charge of giving his mobile phone SIM card (which had some free minutes left on it; he was leaving the country) to his mother's cousin whom he barely knew. See, he recklessly didn't bother to ask if his mother's cousin whom he barely knew if he was going to try to blow up the terminal at Glasgow International Airport at any point in the future.
This, apparently, is part of the definition of "material assistance".
If you're worried about your rights being trampled upon, do something about it. I suggest becoming familiar with the phrase "cold dead hands".
I find it truly amusing that you say this right after mentioning Waco and Ruby Ridge. If taking away your rights requires your hands being dead and cold, then depending on the government agency that doesn't like you, they might just be happy to comply.
Thought experiment: What if the Japanese-Americans rounded up for internment during WW2 had instead chosen to defend themselves? Surely, in public opinion, it would have proven the government's point, right?
Right, which is precisely my point. In addition to the point that we don't know what such a mechanism should look like yet.
But I don't think the problem is unsolvable. Importantly, CPUs are transactional at the microarchitecture level, precisely to support parallelism. Some ISAs support LL/SC, which is a very rudimentary kind of transaction. It may only be a matter of time before transactions become more pervasive at the ISA level, at least in research. Imagine, for example, if you supported small basic block-sized transactions which "roll back" if an interrupt occurrs, or if a branch is mispredicted.
Sure you can do full blown STM with crazy commits and rollbacks that are large and complex but that isn't what causes the problems with most threading issues. Really the primary benefit of STM is just to give an understandable and intuitive means to manage simple things that programmers now do with locks, e.g., making sure the other thread doesn't update part of the object while your thread is making some small change to it.
STM isn't just a compiler issue, it's actually a whole other model for synchronisation (though it's part of a large family of lock-free models, such as RCU).
Today, mutexes, condition variables, semaphores and so on are implemented inside the operating system for a very good reason. We want the same for lock-free models.
The system is, unfortunately, rigged such that a modern startup needs patents. First, they need to stockpile them against existing companies who would rather litigate than compete. Secondly, they need to be able to measure, in some tangible way, how much "innovation" they've done, for the benefit of investors.
The latter point is critical. The value of a startup should be based on how valuable the products are. A patent is an asset which increases the value of the company, even if it's a loss-maker by itself. It's used as a measure of how much innovative stuff is in your product, even though the only value of the innovation is in the product itself.
I'd like to hear some suggestions as to how we could show the value of innovation without patents. I'm sure there must be a better way.
I think that the problem with STM is actually that there is little direct hardware support.
And by that, I don't mean fully transactional caches, though it'd be great if that could be done efficiently. The problem is that existing thread synchronisation mechanisms can be implemented fairly efficiently on top of existing hardware (atomic test-and-set, compare-and-swap or LL/SC). We don't yet know what the appropriate hardware operations are to support STM efficiently.
So I think that pervasive STM will happen, even on desktop hardware, and it'll be very efficient when it does happen. But it won't be for a decade or so at least.
Yes. So while a collection entitled "Love Sonnets of Shakespeare" might be copyrightable (creativity in the collection), "The Complete Works of Shakespeare" probably isn't.
I'm going to define that as meaning "What was the first movie nominated for a special effects/visual effects Academy Award(R), on the strength of the use of computer graphics for anything that wasn't meant to be a computer display?"
(The "on the strength of" qualification rules out, say, the star fields from Star Wars, which were largely incidental. The "non-computer-display" qualification rules out Westworld and Looker (not that they were nominated), and elements such as the hologram from Return of the Jedi.)
The answer is 2010 (from 1984), which used CGI for shots of Jupiter, especially for the monoliths. Another notable mention is Young Sherlock Holmes (1985), with the stained glass knight sequence.
One notable movie that didn't get a nomination is The Last Starfighter (1984), as this was really the first movie which took the whole CGI thing seriously. All space shots were done by computer.
Willow (1988) is significant because it's the first movie that used digital manipulation of live-action plates. It might have even been the first to use digital compositing, but I'm not certain of that.
The Poles had cracked earlier versions before the war.
If you include Poland as an "ally", then it's correct.
The Germans improved the security of Enigma from the original patent bit by bit during the 1930s. The amusing part is that the Poles only just managed to crack previous improvements as the new ones were added. If it wasn't for the piecemeal approach to security, Enigma would likely have remained unbroken until a machine was captured.
When something is measured, it collapses it... What causes the collapse?
There is no such thing as "wavefunction collapse".
Apart from the second law of thermodynamics (which, it would be fair to say, we don't really understand), all of the laws of physics are time-symmetric. In quantum theory, causality works backwards in time just as well as it does forwards, and that includes interactions that leak quantum information.
The many-worlds hypothesis does have some serious problems, such as how a universe with probability p and one with probability -p cancel each other out. (The branching would have to happen "after" the cancellation.)
People designed planes, nuclear bombs and all sorts of engineering/science marvels without computers.
In Japan, they still teach kids how to use an abacus. In fact, I think it's an advantage in the early years, because it gives a tactile meaning to what would otherwise be marks on paper.
I think the biggest problem is with what they didn't do [...]
I think this is a key point.
I'm as critical of the Catholic Church as anyone, but most of the supposed "evils" in living memory, from the connection with Nazism to abuse by clergy, is more a problem of mishandling, negligence or generally being out of touch with the modern world than actual evil acts.
In the case of abuse, for example, the church seemed to be of the opinion that the matter was largely one of internal discipline and could be dealt with privately. In a previous era (before we really understood the psychology of sexual abuse, both for the perpetrator and the victim), that made a certain amount of sense, but it's wholly wrong by today's standards.
In Mexico we have a saying that goes:
"El enemigo de mi enemigo es mi amigo" and means something like "The enemy of my enemy is my friend".
Just so you know, the proverb is Arabic, although I was surprised to learn in Wikipedia that there's an identical Chinese proverb.
And if I may digress for a moment, double negative elimination is not a theorem of intuitionistic logic. (It's related to the law of the excluded middle; the proverb can be thought of as an example of false dichotomy.) However, double negative introduction is. So in intuitionistic logic, the enemy of my enemy may not be my friend, but the enemy of the enemy of my enemy is my enemy.
Read back over what I wrote. What I said was pretty much a tautology: If time is money, then spend your time wisely.
And for the record, I said what I said to agree with your main point! For many kinds of application, Java/Python/Ruby/Perl/etc are closer to optimal for programmer performance, so can often win in a cost/benefit analysis.
Incidentally, P.J. Plauger pointed out that hard real-time performance is also always important. If your monthly payroll processing program takes more than a month to run, it's useless. And if a program takes more than about two years to run on today's hardware, it's going to complete sooner if you wait six months and buy a faster computer.
Slashdot Mirror
We Australians don't even understand that. Who puts a bath on a plane?
Damn, you blew my cover.
The simpler explanation (thank you Occam) is that Scaled Composites has created a completely new market out of thin air, and Northrop wants a bigger piece of the action.
That's a reasonable question, and you're right, it's only tangentially related. But there's a principle here.
In both our countries, we're supposed to be ruled by law, not humans. Legislatures are good at passing overbroad laws with the assurance that they're intended only to have a limited effect. We're only going to use this anti-terrorist legislation against terrorists, right? But humans apply them to other kinds of criminal, or to some guy who gave an old SIM card that he wasn't going to use any more to someone.
That's why you should strongly resist any piece of law (be it legislation or executive order) which is worded broadly. Because it will get misused.
No, that's not it at all. My point is twofold.
One is that civil liberties do you no good if you're dead. More on that below.
The other (and I didn't state this explicitly) is that it's not enough for you to believe in your civil liberties. Other people have to, too. I'm waiting for the day when the NRA stands up for the rights of an unpopular group, like the victims of the "War on Drugs". It won't happen, because the NRA doesn't believe in civil rights. They believe in exactly one civil right.
If they're alive, they have a chance to bring those responsible to justice.
It depends what you mean by "prevalent", but the Cell Broadband Engine is probably the biggest at the moment.
Are you so sure? Let's see that quote again...
I don't know if you've been keeping up with what's happening in Australia at the moment, but we have a man currently being held (though that's a story in itself) on a charge of giving his mobile phone SIM card (which had some free minutes left on it; he was leaving the country) to his mother's cousin whom he barely knew. See, he recklessly didn't bother to ask if his mother's cousin whom he barely knew if he was going to try to blow up the terminal at Glasgow International Airport at any point in the future.
This, apparently, is part of the definition of "material assistance".
I find it truly amusing that you say this right after mentioning Waco and Ruby Ridge. If taking away your rights requires your hands being dead and cold, then depending on the government agency that doesn't like you, they might just be happy to comply.
Thought experiment: What if the Japanese-Americans rounded up for internment during WW2 had instead chosen to defend themselves? Surely, in public opinion, it would have proven the government's point, right?
Dead men have no civil rights.
Right, which is precisely my point. In addition to the point that we don't know what such a mechanism should look like yet.
But I don't think the problem is unsolvable. Importantly, CPUs are transactional at the microarchitecture level, precisely to support parallelism. Some ISAs support LL/SC, which is a very rudimentary kind of transaction. It may only be a matter of time before transactions become more pervasive at the ISA level, at least in research. Imagine, for example, if you supported small basic block-sized transactions which "roll back" if an interrupt occurrs, or if a branch is mispredicted.
STM isn't just a compiler issue, it's actually a whole other model for synchronisation (though it's part of a large family of lock-free models, such as RCU).
Today, mutexes, condition variables, semaphores and so on are implemented inside the operating system for a very good reason. We want the same for lock-free models.
The system is, unfortunately, rigged such that a modern startup needs patents. First, they need to stockpile them against existing companies who would rather litigate than compete. Secondly, they need to be able to measure, in some tangible way, how much "innovation" they've done, for the benefit of investors.
The latter point is critical. The value of a startup should be based on how valuable the products are. A patent is an asset which increases the value of the company, even if it's a loss-maker by itself. It's used as a measure of how much innovative stuff is in your product, even though the only value of the innovation is in the product itself.
I'd like to hear some suggestions as to how we could show the value of innovation without patents. I'm sure there must be a better way.
Personal opinion:
I think that the problem with STM is actually that there is little direct hardware support.
And by that, I don't mean fully transactional caches, though it'd be great if that could be done efficiently. The problem is that existing thread synchronisation mechanisms can be implemented fairly efficiently on top of existing hardware (atomic test-and-set, compare-and-swap or LL/SC). We don't yet know what the appropriate hardware operations are to support STM efficiently.
So I think that pervasive STM will happen, even on desktop hardware, and it'll be very efficient when it does happen. But it won't be for a decade or so at least.
Yes. So while a collection entitled "Love Sonnets of Shakespeare" might be copyrightable (creativity in the collection), "The Complete Works of Shakespeare" probably isn't.
I'm going to define that as meaning "What was the first movie nominated for a special effects/visual effects Academy Award(R), on the strength of the use of computer graphics for anything that wasn't meant to be a computer display?"
(The "on the strength of" qualification rules out, say, the star fields from Star Wars, which were largely incidental. The "non-computer-display" qualification rules out Westworld and Looker (not that they were nominated), and elements such as the hologram from Return of the Jedi.)
The answer is 2010 (from 1984), which used CGI for shots of Jupiter, especially for the monoliths. Another notable mention is Young Sherlock Holmes (1985), with the stained glass knight sequence.
One notable movie that didn't get a nomination is The Last Starfighter (1984), as this was really the first movie which took the whole CGI thing seriously. All space shots were done by computer.
Willow (1988) is significant because it's the first movie that used digital manipulation of live-action plates. It might have even been the first to use digital compositing, but I'm not certain of that.
The first naval Enigma wasn't captured until mid-1940.
If you include Poland as an "ally", then it's correct.
The Germans improved the security of Enigma from the original patent bit by bit during the 1930s. The amusing part is that the Poles only just managed to crack previous improvements as the new ones were added. If it wasn't for the piecemeal approach to security, Enigma would likely have remained unbroken until a machine was captured.
There is no such thing as "wavefunction collapse".
Apart from the second law of thermodynamics (which, it would be fair to say, we don't really understand), all of the laws of physics are time-symmetric. In quantum theory, causality works backwards in time just as well as it does forwards, and that includes interactions that leak quantum information.
The many-worlds hypothesis does have some serious problems, such as how a universe with probability p and one with probability -p cancel each other out. (The branching would have to happen "after" the cancellation.)
In Japan, they still teach kids how to use an abacus. In fact, I think it's an advantage in the early years, because it gives a tactile meaning to what would otherwise be marks on paper.
The Future Fund isn't supposed to just sit there gathering dust. It's supposed to be invested so it makes money.
Investing it in broadband is not ridiculous. It makes money directly (the telcos will pay rent on it) and indirectly by growing the economy.
I'm not saying that Labor's plan is a good one, merely that it's not ridiculous (or an abuse of powers).
I think this is a key point.
I'm as critical of the Catholic Church as anyone, but most of the supposed "evils" in living memory, from the connection with Nazism to abuse by clergy, is more a problem of mishandling, negligence or generally being out of touch with the modern world than actual evil acts.
In the case of abuse, for example, the church seemed to be of the opinion that the matter was largely one of internal discipline and could be dealt with privately. In a previous era (before we really understood the psychology of sexual abuse, both for the perpetrator and the victim), that made a certain amount of sense, but it's wholly wrong by today's standards.
American Baptist, sure. If it was a Southern Baptist Church, it might be an improvement.
Just so you know, the proverb is Arabic, although I was surprised to learn in Wikipedia that there's an identical Chinese proverb.
And if I may digress for a moment, double negative elimination is not a theorem of intuitionistic logic. (It's related to the law of the excluded middle; the proverb can be thought of as an example of false dichotomy.) However, double negative introduction is. So in intuitionistic logic, the enemy of my enemy may not be my friend, but the enemy of the enemy of my enemy is my enemy.
Not only has the front-end been rewritten several times, they've released the framework that they use as open source.
Read back over what I wrote. What I said was pretty much a tautology: If time is money, then spend your time wisely.
And for the record, I said what I said to agree with your main point! For many kinds of application, Java/Python/Ruby/Perl/etc are closer to optimal for programmer performance, so can often win in a cost/benefit analysis.
Incidentally, P.J. Plauger pointed out that hard real-time performance is also always important. If your monthly payroll processing program takes more than a month to run, it's useless. And if a program takes more than about two years to run on today's hardware, it's going to complete sooner if you wait six months and buy a faster computer.