The US has done far, far worse than expected, ICANN has shown that they really can't and the FCC has utterly destroyed any possibility of it doing anything by treating the Internet as not a communication system. The major ISPs are acting like gangsters, using extortion and running protection rackets via the death of network neutrality.
Leaving it where it is WILL kill the Internet as we know it. You WILL lose what freedoms you still have, if power doesn't shift soon.
I don't know if the UN will do any better, but they sure as hell can't do worse and there are no other international organizations capable of the task.
GPR is very low power, as transmissions go. The higher-power high-end 2.4 GHz and 5 GHz radars out there can manage 10 feet through soil but typical sets will manage about half that. Although people will talk about GPR not working well when there's water around, I've no particular interest in trudging through muddy fields in heavy rain. That kind of stuff only really affects corporate and military users, where there's very specific time constraints involved.
My interest is in building GPR systems that are specifically for archaeology, where your time window is measured in centuries not hours, so waiting until the ground is dry is a non-issue. However, excavations are slow and expensive, so you've got to pick where you look with a lot of care.
The crucial difference is that in archaeological surveys, you want a LOT of data. You care about shadows, reflections at weird angles, the refractive indexes of different materials and other details that "regular" GPS users consider anomalies. Because the regular users don't want that data, it's generally eliminated from the data set. Which is fine for what they're doing. If you know there are N pipes and one has a fracture, you want to see N lines with one line being different (doesn't matter how) from the rest. That's your typical corporate and local authority use case. The military want to spot mines, IEDs and UXBs, but they don't need to know if the objects surrounding it are remnants of a Greek statue or just randomly-shaped lumps of rock.
High frequencies = high resolution, but limited depth; Low frequencies = low resolution, but greater depth. In theory, you'd want a mix to learn everything but nobody actually does that.
UHF, SHF and EHF should all be fine. The government position is that civilians don't need GPR, that it's a military technology, and to hell with scientists who say otherwise. However, as far as I know only the FCC actually restricts the technology. Pointing a ham antenna at the ground is an interesting idea, might actually work very well. Incorporating the call sign into the signal would be easy.
Since the lower frequencies can pass through rock, it's safe to say line of sight isn't necessary. However, the limited range is a problem and as 100 Gb/s is already achievable by Ethernet (300 Gb/s for Infiniband), it's not clear what the benefit is of adopting a technology that will max out (in the future) at the speeds you can get now, when most of the benefit of wireless really won't apply.
So it'll be useless in England or the US' Pacific Northwest.
Although it can penetrate a wide range of materials, that doesn't mean there's zero reflection. I'm much less interested in this for Wifi and much MORE interested in it for building a GPR, since it is essentially illegal in the US for garage developers to experiment with GPR technology (you can't even get the license to operate one unless you've a provable corporate need or are military). If these frequencies aren't regulated, FCC rules prohibiting any kind of private GPR research don't apply.
The MIPS won't, since Microsoft doesn't write for it, so that's 3 of the CPUs. Same for the Itanium, since Microsoft has abandoned that. It's very unlikely Microsoft is developing for both the A9 and A15, so that eliminates half of what's left. Most ARMs won't be running a MS OS, it'll be a minority OS for a long time on that chip. So really only the AMD CPU even has the potential for vendor lock-in by Microsoft.
However, in the more abstract sense, you can only extrapolate models so far beyond the furthest point for which you have data before the models break down. But you don't know when that will happen, it depends on how good the model is and you can't know that in advance.
Specs are therefore reasonably conservative. They'll go a little bit beyond what's feasible right now, but only a little. Just enough to give wiggle-room and space for progress, but not to the point where there's a serious risk of problems developing.
Examples of reasons why they need to be careful: both electron tunneling and thermal noise will generating errors, yields at lower scales aren't always predictable, alternative techniques for performing the same function at the higher speed can be incompatible with accelerating the original technique, etc.
A spec is supposed to work for anything it is a spec for - manufacturers do revise/debug specs but they vastly prefer to release upgraded versions as compatibility issues and implementation details can all be carefully documented and properly presented.
AMD has now low-power cores due out later in the year. I imagine their timing is designed to make it possible to have manufacturers design motherboards with the new memory in mind, rather than have them aim for DDR3 and only develop a DDR4 board later.
I would imagine DDR4's higher performance will be important to Intel now that they've designed an even faster Itanium chip (8-core, multithreaded). The power requirement probably wouldn't be much of a factor, though.
Sir Clive Sinclair gave very similar reasons for using the 68008 (the 8-bit version of the 32-bit 68000) - he stated that nobody needed a 32-bit computer, but that you needed a machine to be advertisable as a 32-bit system to compete with the Mac and Atari ST.
Actually, they could. The processors of the time used two interwoven 16-bit registers with a 12-bit overlap. (This meant some memory locations had multiple addresses.) The total sort-of-linear address space was therefore 20 bits, or 1 megabyte.
Plenty of extended memory cards existed at the time, which used a memory banking system to produce the illusion of larger machines. The only restriction with banked memory is that it slows the machine down as the CPU can only see one bank at a time. Well, that and reads/writes can't span multiple banks.
10% by page count or some percentage of one chapter of a book (IIRC it's 25% but it has been a while), whichever is lesser, being "fair use" has been placed on copyright information notices in British University libraries since at least the late 1980s, maybe earlier. Oxford and Cambridge have no excuse for not knowing what their own libraries are instructing, so I've little sympathy for any copyright claim that ran foul of fair use law.
The judge's argument was essentially that if the flaw exists for software patents then it exists for ALL patents and that burying the symptoms merely allows the flaws to continue untreated, that by making the problems blatantly obvious allows them to be fixed for everything.
There is a lot to that. I don't think that the argument, in and of itself, is that unsound.
The problem is that the flaws are so deep and so entrenched, it's unclear that fixing the problems would allow patents to survive at all.
True, and that would meet the sum of three odd primes requirement. If you include 2, then 7+2=9, making it the sum of two primes where one prime is NOT a Goldbach Number.
Looked the conjecture up on Wikipedia. It's actually a little more specific still - every even number is a Goldbach Number, where a Goldbach number is a number that can be written as the sum of two odd primes.
That means that every odd number can always be written as the sum of three primes or less. Numbers like 9 are the sum of two primes but are NOT Goldbach numbers since one of the primes is 2 and the requirement is that both primes be odd.
Errors in this post are due to Wikipedia, blame them if there are any.
Bitcoin's creators are completely ignorant about what makes for a useful currency, what makes for a secure system and what makes for a workable system. The FBI's concerns are immaterial in all of this, Bitcoin is simply a very badly-designed system.
Current central currencies are no better, they have many of the same defects and a whole host of different ones.
What is needed is a replacement system, sure. A decentralized system, with no tracking. But Bitcoin isn't it. Bitcoin just makes power generating companies very wealthy.
No system is truly robust unless it meets not only the trivial requirements (such as strong encryption) but also meets the harder requirements (it must meet the reliability threshold outlined in the Byzantine General's Problem, for example, where the nodes in the problem are not simply the people but also the computers concerned).
That is a tough one. The problem is not solved in centralized banking systems, it is merely better-hidden. Banks are reputed to lose many billions a year to people getting in through backdoors, or breaking horribly bad security systems (CitiBank letting you access anyone's account by editing the URL you go to -- that likely cost them a lot of money).
Decentralized currency is better at making problems visible, but it needs considerable security in areas that BitCoin has neglected. BitCoin makes forgery hard, but that's as far as their security goes. Bank of England's paper currency also makes forgery hard, but that doesn't make it worth using.
Old, wizened lawyer #1: "If we agree to Oracle's naive, stupid and utterly doomed plan, we will make ourselves look extremely stupid to the geeks."
Old, wizened lawyer #2: "The geeks can't afford us anyway, so who cares what they think? We'll make gob-loads of money off Oracle. Imagine what we can buy with that?"
Middle-aged, blue, slightly-fuzzy lawyer that looks remarkably like a sock puppet: "Coooooookies!!!!! Mrumrumrumrumrumrumrum"
The lawyers will be paid by Oracle, win or lose. They won't get paid if they're not involved in a lawsuit. Oracle wants them involved in a lawsuit that they will probably lose. Turning Oracle down is an expensive way to be ethical.
The "ideal" would be to add a stipulation that no provider (be they medical or otherwise) is permitted to cause the person to exceed the stipulated yearly dose without informed consent, with total dosage accumulated to that point being recorded by the provider before and after a scan.
What I do not see is any way you could possibly achieve this, without being incredibly invasive and/or potentially causing worse side-effects.
Nonetheless, the goal should be to not merely ensure individual scans are given responsibly but that the cumulation of scans is also being done responsibly. That's much harder to do, since human memory isn't reliable and patients can't possibly know what they've been exposed to up to that point (especially in the case of TSA scans). It doesn't matter if individual scans are relatively harmless, it's the dosage per unit time that matters.
Slashdot Mirror
The US has done far, far worse than expected, ICANN has shown that they really can't and the FCC has utterly destroyed any possibility of it doing anything by treating the Internet as not a communication system. The major ISPs are acting like gangsters, using extortion and running protection rackets via the death of network neutrality.
Leaving it where it is WILL kill the Internet as we know it. You WILL lose what freedoms you still have, if power doesn't shift soon.
I don't know if the UN will do any better, but they sure as hell can't do worse and there are no other international organizations capable of the task.
GPR is very low power, as transmissions go. The higher-power high-end 2.4 GHz and 5 GHz radars out there can manage 10 feet through soil but typical sets will manage about half that. Although people will talk about GPR not working well when there's water around, I've no particular interest in trudging through muddy fields in heavy rain. That kind of stuff only really affects corporate and military users, where there's very specific time constraints involved.
My interest is in building GPR systems that are specifically for archaeology, where your time window is measured in centuries not hours, so waiting until the ground is dry is a non-issue. However, excavations are slow and expensive, so you've got to pick where you look with a lot of care.
The crucial difference is that in archaeological surveys, you want a LOT of data. You care about shadows, reflections at weird angles, the refractive indexes of different materials and other details that "regular" GPS users consider anomalies. Because the regular users don't want that data, it's generally eliminated from the data set. Which is fine for what they're doing. If you know there are N pipes and one has a fracture, you want to see N lines with one line being different (doesn't matter how) from the rest. That's your typical corporate and local authority use case. The military want to spot mines, IEDs and UXBs, but they don't need to know if the objects surrounding it are remnants of a Greek statue or just randomly-shaped lumps of rock.
High frequencies = high resolution, but limited depth; Low frequencies = low resolution, but greater depth. In theory, you'd want a mix to learn everything but nobody actually does that.
UHF, SHF and EHF should all be fine. The government position is that civilians don't need GPR, that it's a military technology, and to hell with scientists who say otherwise. However, as far as I know only the FCC actually restricts the technology. Pointing a ham antenna at the ground is an interesting idea, might actually work very well. Incorporating the call sign into the signal would be easy.
Since the lower frequencies can pass through rock, it's safe to say line of sight isn't necessary. However, the limited range is a problem and as 100 Gb/s is already achievable by Ethernet (300 Gb/s for Infiniband), it's not clear what the benefit is of adopting a technology that will max out (in the future) at the speeds you can get now, when most of the benefit of wireless really won't apply.
I am Icarus, and I approve of this message.
So it'll be useless in England or the US' Pacific Northwest.
Although it can penetrate a wide range of materials, that doesn't mean there's zero reflection. I'm much less interested in this for Wifi and much MORE interested in it for building a GPR, since it is essentially illegal in the US for garage developers to experiment with GPR technology (you can't even get the license to operate one unless you've a provable corporate need or are military). If these frequencies aren't regulated, FCC rules prohibiting any kind of private GPR research don't apply.
It goes very fast in a straight line, corners horribly and requires enormous amounts of energy. :)
The MIPS won't, since Microsoft doesn't write for it, so that's 3 of the CPUs. Same for the Itanium, since Microsoft has abandoned that. It's very unlikely Microsoft is developing for both the A9 and A15, so that eliminates half of what's left. Most ARMs won't be running a MS OS, it'll be a minority OS for a long time on that chip. So really only the AMD CPU even has the potential for vendor lock-in by Microsoft.
As others have noted, the tech wasn't there.
However, in the more abstract sense, you can only extrapolate models so far beyond the furthest point for which you have data before the models break down. But you don't know when that will happen, it depends on how good the model is and you can't know that in advance.
Specs are therefore reasonably conservative. They'll go a little bit beyond what's feasible right now, but only a little. Just enough to give wiggle-room and space for progress, but not to the point where there's a serious risk of problems developing.
Examples of reasons why they need to be careful: both electron tunneling and thermal noise will generating errors, yields at lower scales aren't always predictable, alternative techniques for performing the same function at the higher speed can be incompatible with accelerating the original technique, etc.
A spec is supposed to work for anything it is a spec for - manufacturers do revise/debug specs but they vastly prefer to release upgraded versions as compatibility issues and implementation details can all be carefully documented and properly presented.
AMD has now low-power cores due out later in the year. I imagine their timing is designed to make it possible to have manufacturers design motherboards with the new memory in mind, rather than have them aim for DDR3 and only develop a DDR4 board later.
I would imagine DDR4's higher performance will be important to Intel now that they've designed an even faster Itanium chip (8-core, multithreaded). The power requirement probably wouldn't be much of a factor, though.
Sir Clive Sinclair gave very similar reasons for using the 68008 (the 8-bit version of the 32-bit 68000) - he stated that nobody needed a 32-bit computer, but that you needed a machine to be advertisable as a 32-bit system to compete with the Mac and Atari ST.
Actually, they could. The processors of the time used two interwoven 16-bit registers with a 12-bit overlap. (This meant some memory locations had multiple addresses.) The total sort-of-linear address space was therefore 20 bits, or 1 megabyte.
Plenty of extended memory cards existed at the time, which used a memory banking system to produce the illusion of larger machines. The only restriction with banked memory is that it slows the machine down as the CPU can only see one bank at a time. Well, that and reads/writes can't span multiple banks.
10% by page count or some percentage of one chapter of a book (IIRC it's 25% but it has been a while), whichever is lesser, being "fair use" has been placed on copyright information notices in British University libraries since at least the late 1980s, maybe earlier. Oxford and Cambridge have no excuse for not knowing what their own libraries are instructing, so I've little sympathy for any copyright claim that ran foul of fair use law.
The judge's argument was essentially that if the flaw exists for software patents then it exists for ALL patents and that burying the symptoms merely allows the flaws to continue untreated, that by making the problems blatantly obvious allows them to be fixed for everything.
There is a lot to that. I don't think that the argument, in and of itself, is that unsound.
The problem is that the flaws are so deep and so entrenched, it's unclear that fixing the problems would allow patents to survive at all.
http://mathworld.wolfram.com/GoldbachNumber.html
Wolfram says the same thing. So Nyah! Pttttthhhhhpt!
True, and that would meet the sum of three odd primes requirement. If you include 2, then 7+2=9, making it the sum of two primes where one prime is NOT a Goldbach Number.
ObTrivia: Found this page, it looks like it has some interesting information: http://homepage.mac.com/billtomlinson/primes.html
Looked the conjecture up on Wikipedia. It's actually a little more specific still - every even number is a Goldbach Number, where a Goldbach number is a number that can be written as the sum of two odd primes.
That means that every odd number can always be written as the sum of three primes or less. Numbers like 9 are the sum of two primes but are NOT Goldbach numbers since one of the primes is 2 and the requirement is that both primes be odd.
Errors in this post are due to Wikipedia, blame them if there are any.
Bitcoin's creators are completely ignorant about what makes for a useful currency, what makes for a secure system and what makes for a workable system. The FBI's concerns are immaterial in all of this, Bitcoin is simply a very badly-designed system.
Current central currencies are no better, they have many of the same defects and a whole host of different ones.
What is needed is a replacement system, sure. A decentralized system, with no tracking. But Bitcoin isn't it. Bitcoin just makes power generating companies very wealthy.
I have gold, but that has made bugger all difference. I'm as ignored by women now as I was before. So that's clearly not the deciding factor.
No system is truly robust unless it meets not only the trivial requirements (such as strong encryption) but also meets the harder requirements (it must meet the reliability threshold outlined in the Byzantine General's Problem, for example, where the nodes in the problem are not simply the people but also the computers concerned).
That is a tough one. The problem is not solved in centralized banking systems, it is merely better-hidden. Banks are reputed to lose many billions a year to people getting in through backdoors, or breaking horribly bad security systems (CitiBank letting you access anyone's account by editing the URL you go to -- that likely cost them a lot of money).
Decentralized currency is better at making problems visible, but it needs considerable security in areas that BitCoin has neglected. BitCoin makes forgery hard, but that's as far as their security goes. Bank of England's paper currency also makes forgery hard, but that doesn't make it worth using.
Old, wizened lawyer #1: "If we agree to Oracle's naive, stupid and utterly doomed plan, we will make ourselves look extremely stupid to the geeks."
Old, wizened lawyer #2: "The geeks can't afford us anyway, so who cares what they think? We'll make gob-loads of money off Oracle. Imagine what we can buy with that?"
Middle-aged, blue, slightly-fuzzy lawyer that looks remarkably like a sock puppet: "Coooooookies!!!!! Mrumrumrumrumrumrumrum"
The lawyers will be paid by Oracle, win or lose. They won't get paid if they're not involved in a lawsuit. Oracle wants them involved in a lawsuit that they will probably lose. Turning Oracle down is an expensive way to be ethical.
...the asteroid belt IS a failed planet after all?
If so, then it demonstrates why Real Scientists (not ones that kill cute puppies like Pluto for amusement) are wary of definitive statements.
The TSA is a cult to Dread Cthulhu. What makes you think there's a lesser evil?
The "ideal" would be to add a stipulation that no provider (be they medical or otherwise) is permitted to cause the person to exceed the stipulated yearly dose without informed consent, with total dosage accumulated to that point being recorded by the provider before and after a scan.
What I do not see is any way you could possibly achieve this, without being incredibly invasive and/or potentially causing worse side-effects.
Nonetheless, the goal should be to not merely ensure individual scans are given responsibly but that the cumulation of scans is also being done responsibly. That's much harder to do, since human memory isn't reliable and patients can't possibly know what they've been exposed to up to that point (especially in the case of TSA scans). It doesn't matter if individual scans are relatively harmless, it's the dosage per unit time that matters.