Lone wolf is particularly bad. Wolves are horrible hunters when solitary. They're only a significant threat to anything in packs. (The German U-Boat fleet discovered that, too, when they became incapable of forming their own wolfpacks.)
If you want a meme that describes solitary madmen, "lone lolcat" would be more effective. Cats are dangerous on their own. Just ask a cheezburger.
I disagree, in that I believe firmly that if education were massively improved (and I mean massively), many of the other problems would become obvious to the majority of people as would the solutions. People function with split realities, but not well, and the more extreme the split the greater the discomfort. Deliberately worsen that discomfort through high quality (and maximal quantity) education. These will be the people who do the voting in 20 years time and who will also be the candidates then as well. Superior voters with superior candidates to choose will necessarily improve the situation as a whole.
Those will also be the businesspeople in that timeframe and thus will be making more rational decisions on what jobs are appropriate to be overseas and what jobs are appropriate here.
An informed electorate, or so Plato tells us, will also be less eager to go to war and less eager to blindly follow populist leaders. Indeed, he made it an essential criterion for a functional democracy. The experience of the last decade tells us he was right on the dangers of ignorance, so it seems worth testing whether he was right on the benefits of knowledge and wisdom.
Education alone won't fix all the issues, but I see no reason why - over time - it wouldn't solve a good number. Combine it with quality public healthcare and you solve many of the problems that ill-health cause (weakened economy, reduced opportunity, reduced flexibility, inferior mobility, desires for feel-good politics and/or substance-abuse, etc.)
That last one is worth reflecting on a bit. Religion may be the opiate of the masses, but feel-good politics is crack cocaine. Neither is good for you, both should be avoided where possible, but populist politicians are infinitely more dangerous than populist preachers. Jim Jones killed less than a thousand in total, fanatical politicians in 1914-1918 were managing to average that per day per nation for four years.
After his example of the Chinese Room, in which he failed to grasp the obvious (that the observation bears multiple interpretations, including the one he doesn't want to make), I ceased to regard the Emperor's New Mind as being all that interesting. His points on teleportation in that book were more interesting and more valid. A book on those would have had greater value.
Yes but that doesn't mean everything was compiled with RHEL's standard compile flags or that different patches weren't included/excluded. I haven't looked at the specifics of Rocks' packages for a while, but I can trust they're tried-and-tested in cluster environments. RHEL may be 100% identical in all respects, but then it might not be. All I'm really sure of is that Red Hat's QA won't have put clustering as high on their list of things to test against.
That's no criticism of Red Hat. They've a finite number of testers in-house and a finite time before a release. It's natural that they test as much as they can but slant it towards those buying RHEL. Generally, academics and scientists building moderately-coupled clusters on a shoestring budget won't be RHEL's biggest customers. CentOS - much the same applies, in that biggest user groups are going to take priority and smaller (albeit highly interesting) groups will get some support but not as much.
I'm fairly certain MOSIX compiled against RHEL is available, so that should be just fine from that standpoint if MOSIX is the way the person wants to go. I just don't know what distros have tested against kernel patches for clustering, or which ones they have tried if they have. There are so many and there's a high risk that some will conflict with patches the distro does use. (It's what made the FOLK project such a bugger to maintain. Conflicts are rife on patches that aren't even in the popular patchsets.)
Rocks is another good distro for this. It's designed specifically for cluster use, with packages pre-built with that in mind.
It also depends some on what clustering system you're using. If you're wanting to use MOSIX or Kerrighed, then use a distro the one you want to use is well-tested on. Kernel patch conflicts can otherwise make things very difficult.
No, the Constitution is not a Federal law. In fact, there is nothing in the Constitution that applies to private citizens at all. It is a law of governance, not the governed.
People believe in 20 contradictory things before breakfast. It's entirely plausible that 50% of Americans would simultaneously want others patted down for security reasons provided they themselves had legal protection against it, without even realizing that a contradiction even exists.
Actually, yes, I do. The cost of reaching escape velocity is fixed, so whether go to the ISS or Mars doesn't change the cost of the launcher at all. The cost of =staying= in orbit is high (you've got to overcome drag, move away from space debris, etc). The cost of staying on the ground is nil, since the base isn't going anywhere. The cost of replacing the men on the ISS is high, the cost of not replacing anyone whilst on-route, at, or returning, from Mars is nil. The cost of assembling hundreds of modules packed with scientific equiptment, launching and assembling them is high. The cost of three modules, packed with food and spare parts, is very close to nil.
Way ahead of state-of-the-art? This is Werner von Braun had all figured out in the 70s. Don't blame me if NASA is four decades behind where it was.
Well, it depends. The examples in the article were kinda pathetic, I have to say. However, here are some things we have lost that are a bit (to me, at least) more meaningful:
Content Addressable Memory Battery-backed RAM (none of the limited writes or speed issues of Flash) Deep pipelining Sprites (actual, meaningful, sprites) Definable sound envelopes (though some JACK-compatible modules can provide this feature) Offline virtual memory (early machines could extend virtual memory onto mobile media, so processes unlikely to be needed soon could be offloaded to free space and reloaded later)
I would consider "divisible" for these purposes to mean "if you impose sufficient and appropriate forces, you will see evidence of underlying structure" (since you can't have a composite if there's nothing doing the composing) and to mean "if you split the particle, you will yield something other than energy" (since all matter is ultimately energy) PLUS "if you reduce some particle to a single particle (that is not the original particle) and energy, there exists at least one particle that can be constructed from less than the energy released AND which fits the underlying structure you have revealed" (ie: if you can only see part of the underlying structure, you can nonetheless deduce part or all of the remainder from what you do observe with no contradictions).
So, if an electron has both an internal structure AND you can identify at least one component directly AND any components you cannot directly observe CAN be directly deduced from the observations, THEN an electron is divisible.
In the case of a proton, for example, you can break it into a quark-gluon soup (which would be the ideal, since you can then observe quarks and gluons independently) or you can observe the quarks directly and deduce the gluons from the fact that what remains fits the model for gluons correctly.
That the electron should, under extreme conditions, alter shape tells me that there is something that can obtain a lower-energy state than a sphere. You can't distort points, however, as there's nothing to distort. You can't achieve a lower energy state if there's one item present. How do you sort a unitary item? As for mass - an electron has a mass that is approximately 1/1836 that of the proton. Since ONLY the Higgs boson* gives the property of mass, you would require enough such bosons for the electron to have the mass it does.
*Agreed, this assumes the Higgs boson exists, which is not the case in alternatives to the Standard Model. However, under accepted methods, you assume H0 to be what you would expect under the accepted theory and H1 to be that the accepted theory is broken in some way. Thus, it is correct to assume the Higgs boson is valid and then work to find a case where it is NOT valid. The accepted theory, again under accepted methods, is ALWAYS the the theory that is as simple as possible but no simpler. ie: the simplest theory that still works.
The problem with gravity being weaker than other fields is, in my opinion, of no significance as far as whether the electron has structure. A fact can have many interpretations, which is why there are so many theories trying to explain it. With no meaningful information at present on why the fields appear to be of different strength, we can't even be sure they actually are. Some versions of supergravity suggest that they are not different but that the relative distance makes for the illusion.
Self-maintaining machines have been drempt about since the 40s. None exist. Pit a current UAV against a current fighter pilot and I'll tell you which is the more likely to get the job done. Sure, UAVs will eventually get smarter but AI is simply not advanced enough - and won't be for many, many decades - to replace humans for anything more than the most trivial of functions. Current AI can run through identification keys faster or run through simple herustics faster, sure, but that's it.
I was told in the 70s that knowing electronics was futile as computers would someday repair computers, and that computer programming was a useless skill as computers would someday program computers. Neither has happened and both skills are still very much the province of humans.
What may be possible a century from now - that's different. Maybe you could have robot repair-men on Mars by then. I'm not talking about then, though.
As for Bush, pffft. He had no head for numbers and NASA screwed up most of theirs under his regime. A manned mission to Mars would cost 10-15x as much as an unmanned mission, and given the survival rate of unmanned missions to Mars (about 50%), you would want to halve that difference since you've got to have the unmanned mission actually reach the ground safely.
Because building and running a manned base is (a) relatively cheap, and (b) more reliable as there's fewer components to go wrong.
A manned base takes adequate radiation shielding (not hard), oxygen scrubbers (NASA's got those), a vehicle capable of going from Martian surface to orbit (similar to the lunar landers), two years of food and a kindle with enough books to last the one year in, one year back (big deal, torrents should work fine using the Delay Tolerant Protocol NASA developed).
A thousand launchers, alone, would outstrip the cost of such a venture by enough orders of magnitude to plunge the Earth's economy into permanent recession. A thousand rovers alone, ignoring all operating costs, would so vastly exceed the cost of switching from a relatively light launcher for a rover to two or three heavy launchers for the couple of humans, two sets of spare parts and essential supplies, that you'd be insane to try it.
Can humans remain in almost total isolation for that long? Sure, you probably want to opt for people who are on the heavier side of Aspergers or even HFA, since they're best-equipt to handle it. Old-Series Trekkies could manage it as well, since the DVDs for the episodes won't cost as much as a year's supply of books, but you'd get too many complaints that the rovers were illogical.
The formal edge of the solar system is at the end of the heliopause. Who cares how many you need, we've something like 3 or 4 trillion of them. It is essential that we design a GPU capable of handling floating point numbers to this accuracy.
The easiest explanation is that the particle has no fixed position but exists over a probability wave. It can be anywhere on that wave, with the probability equal to the value of the wave at that point. Thus, it is a particle and a wave, but cannot be treated as both at the same time (which is exactly what we observe). The certainty with which we know the position also decreases the certainty of its velocity since it will reside ANYWHERE on the wave on the next instant. Since velocity is rate of change of displacement, we cannot know the velocity. However, if your concern is with velocity, you can measure how far the wave has moved. In so doing, the exact position of the electron on the wave is known only by the probability and so is unknown. For all intents and purposes, it doesn't have both a simultaneous velocity and position in this description (which is also what we observe).
This is not the "standard model" (well, not unless I just reverse-engineered it in under a minute - and, if so, you can sent the Nobel prize to the following addres....) but it IS an easy-to-master description that explains Heisenberg's uncertainty principle, wave-particle duality, diffraction patterns with extremely low photon counts, quantum tunneling and many of the other strange effects described at that level.
Slashdot Mirror
2MASS publish their full catalogue on their FTP server. It's painful to download, even on broadband.
Lone wolf is particularly bad. Wolves are horrible hunters when solitary. They're only a significant threat to anything in packs. (The German U-Boat fleet discovered that, too, when they became incapable of forming their own wolfpacks.)
If you want a meme that describes solitary madmen, "lone lolcat" would be more effective. Cats are dangerous on their own. Just ask a cheezburger.
I disagree, in that I believe firmly that if education were massively improved (and I mean massively), many of the other problems would become obvious to the majority of people as would the solutions. People function with split realities, but not well, and the more extreme the split the greater the discomfort. Deliberately worsen that discomfort through high quality (and maximal quantity) education. These will be the people who do the voting in 20 years time and who will also be the candidates then as well. Superior voters with superior candidates to choose will necessarily improve the situation as a whole.
Those will also be the businesspeople in that timeframe and thus will be making more rational decisions on what jobs are appropriate to be overseas and what jobs are appropriate here.
An informed electorate, or so Plato tells us, will also be less eager to go to war and less eager to blindly follow populist leaders. Indeed, he made it an essential criterion for a functional democracy. The experience of the last decade tells us he was right on the dangers of ignorance, so it seems worth testing whether he was right on the benefits of knowledge and wisdom.
Education alone won't fix all the issues, but I see no reason why - over time - it wouldn't solve a good number. Combine it with quality public healthcare and you solve many of the problems that ill-health cause (weakened economy, reduced opportunity, reduced flexibility, inferior mobility, desires for feel-good politics and/or substance-abuse, etc.)
That last one is worth reflecting on a bit. Religion may be the opiate of the masses, but feel-good politics is crack cocaine. Neither is good for you, both should be avoided where possible, but populist politicians are infinitely more dangerous than populist preachers. Jim Jones killed less than a thousand in total, fanatical politicians in 1914-1918 were managing to average that per day per nation for four years.
Upwards in Australia, relative to Europe.
After his example of the Chinese Room, in which he failed to grasp the obvious (that the observation bears multiple interpretations, including the one he doesn't want to make), I ceased to regard the Emperor's New Mind as being all that interesting. His points on teleportation in that book were more interesting and more valid. A book on those would have had greater value.
Yes but that doesn't mean everything was compiled with RHEL's standard compile flags or that different patches weren't included/excluded. I haven't looked at the specifics of Rocks' packages for a while, but I can trust they're tried-and-tested in cluster environments. RHEL may be 100% identical in all respects, but then it might not be. All I'm really sure of is that Red Hat's QA won't have put clustering as high on their list of things to test against.
That's no criticism of Red Hat. They've a finite number of testers in-house and a finite time before a release. It's natural that they test as much as they can but slant it towards those buying RHEL. Generally, academics and scientists building moderately-coupled clusters on a shoestring budget won't be RHEL's biggest customers. CentOS - much the same applies, in that biggest user groups are going to take priority and smaller (albeit highly interesting) groups will get some support but not as much.
I'm fairly certain MOSIX compiled against RHEL is available, so that should be just fine from that standpoint if MOSIX is the way the person wants to go. I just don't know what distros have tested against kernel patches for clustering, or which ones they have tried if they have. There are so many and there's a high risk that some will conflict with patches the distro does use. (It's what made the FOLK project such a bugger to maintain. Conflicts are rife on patches that aren't even in the popular patchsets.)
Rocks is another good distro for this. It's designed specifically for cluster use, with packages pre-built with that in mind.
It also depends some on what clustering system you're using. If you're wanting to use MOSIX or Kerrighed, then use a distro the one you want to use is well-tested on. Kernel patch conflicts can otherwise make things very difficult.
"Don't mess with Texas" apparently only applies on non-Thursdays.
No, the Constitution is not a Federal law. In fact, there is nothing in the Constitution that applies to private citizens at all. It is a law of governance, not the governed.
Arizona is illegal under the supremacy clause.
People believe in 20 contradictory things before breakfast. It's entirely plausible that 50% of Americans would simultaneously want others patted down for security reasons provided they themselves had legal protection against it, without even realizing that a contradiction even exists.
What did you expect? Once it appeared on Slashdot, there was no saving it.
Wouldn't those be Untested Tubes (or Liberal Arts Tubes)?
Actually, yes, I do. The cost of reaching escape velocity is fixed, so whether go to the ISS or Mars doesn't change the cost of the launcher at all. The cost of =staying= in orbit is high (you've got to overcome drag, move away from space debris, etc). The cost of staying on the ground is nil, since the base isn't going anywhere. The cost of replacing the men on the ISS is high, the cost of not replacing anyone whilst on-route, at, or returning, from Mars is nil. The cost of assembling hundreds of modules packed with scientific equiptment, launching and assembling them is high. The cost of three modules, packed with food and spare parts, is very close to nil.
Way ahead of state-of-the-art? This is Werner von Braun had all figured out in the 70s. Don't blame me if NASA is four decades behind where it was.
Well, it depends. The examples in the article were kinda pathetic, I have to say. However, here are some things we have lost that are a bit (to me, at least) more meaningful:
Content Addressable Memory
Battery-backed RAM (none of the limited writes or speed issues of Flash)
Deep pipelining
Sprites (actual, meaningful, sprites)
Definable sound envelopes (though some JACK-compatible modules can provide this feature)
Offline virtual memory (early machines could extend virtual memory onto mobile media, so processes unlikely to be needed soon could be offloaded to free space and reloaded later)
I would consider "divisible" for these purposes to mean "if you impose sufficient and appropriate forces, you will see evidence of underlying structure" (since you can't have a composite if there's nothing doing the composing) and to mean "if you split the particle, you will yield something other than energy" (since all matter is ultimately energy) PLUS "if you reduce some particle to a single particle (that is not the original particle) and energy, there exists at least one particle that can be constructed from less than the energy released AND which fits the underlying structure you have revealed" (ie: if you can only see part of the underlying structure, you can nonetheless deduce part or all of the remainder from what you do observe with no contradictions).
So, if an electron has both an internal structure AND you can identify at least one component directly AND any components you cannot directly observe CAN be directly deduced from the observations, THEN an electron is divisible.
In the case of a proton, for example, you can break it into a quark-gluon soup (which would be the ideal, since you can then observe quarks and gluons independently) or you can observe the quarks directly and deduce the gluons from the fact that what remains fits the model for gluons correctly.
That the electron should, under extreme conditions, alter shape tells me that there is something that can obtain a lower-energy state than a sphere. You can't distort points, however, as there's nothing to distort. You can't achieve a lower energy state if there's one item present. How do you sort a unitary item? As for mass - an electron has a mass that is approximately 1/1836 that of the proton. Since ONLY the Higgs boson* gives the property of mass, you would require enough such bosons for the electron to have the mass it does.
*Agreed, this assumes the Higgs boson exists, which is not the case in alternatives to the Standard Model. However, under accepted methods, you assume H0 to be what you would expect under the accepted theory and H1 to be that the accepted theory is broken in some way. Thus, it is correct to assume the Higgs boson is valid and then work to find a case where it is NOT valid. The accepted theory, again under accepted methods, is ALWAYS the the theory that is as simple as possible but no simpler. ie: the simplest theory that still works.
The problem with gravity being weaker than other fields is, in my opinion, of no significance as far as whether the electron has structure. A fact can have many interpretations, which is why there are so many theories trying to explain it. With no meaningful information at present on why the fields appear to be of different strength, we can't even be sure they actually are. Some versions of supergravity suggest that they are not different but that the relative distance makes for the illusion.
I dunno. I try to stay clear of the people who apparently do live in test tubes.
The silicon can run Linux, sure. All you need then is an iron whisker to let you pass signals between the silicon and the brain.
Since it's a mix of brains and silicon, aren't these Borg?
Self-maintaining machines have been drempt about since the 40s. None exist. Pit a current UAV against a current fighter pilot and I'll tell you which is the more likely to get the job done. Sure, UAVs will eventually get smarter but AI is simply not advanced enough - and won't be for many, many decades - to replace humans for anything more than the most trivial of functions. Current AI can run through identification keys faster or run through simple herustics faster, sure, but that's it.
I was told in the 70s that knowing electronics was futile as computers would someday repair computers, and that computer programming was a useless skill as computers would someday program computers. Neither has happened and both skills are still very much the province of humans.
What may be possible a century from now - that's different. Maybe you could have robot repair-men on Mars by then. I'm not talking about then, though.
As for Bush, pffft. He had no head for numbers and NASA screwed up most of theirs under his regime. A manned mission to Mars would cost 10-15x as much as an unmanned mission, and given the survival rate of unmanned missions to Mars (about 50%), you would want to halve that difference since you've got to have the unmanned mission actually reach the ground safely.
Because building and running a manned base is (a) relatively cheap, and (b) more reliable as there's fewer components to go wrong.
A manned base takes adequate radiation shielding (not hard), oxygen scrubbers (NASA's got those), a vehicle capable of going from Martian surface to orbit (similar to the lunar landers), two years of food and a kindle with enough books to last the one year in, one year back (big deal, torrents should work fine using the Delay Tolerant Protocol NASA developed).
A thousand launchers, alone, would outstrip the cost of such a venture by enough orders of magnitude to plunge the Earth's economy into permanent recession. A thousand rovers alone, ignoring all operating costs, would so vastly exceed the cost of switching from a relatively light launcher for a rover to two or three heavy launchers for the couple of humans, two sets of spare parts and essential supplies, that you'd be insane to try it.
Can humans remain in almost total isolation for that long? Sure, you probably want to opt for people who are on the heavier side of Aspergers or even HFA, since they're best-equipt to handle it. Old-Series Trekkies could manage it as well, since the DVDs for the episodes won't cost as much as a year's supply of books, but you'd get too many complaints that the rovers were illogical.
The formal edge of the solar system is at the end of the heliopause. Who cares how many you need, we've something like 3 or 4 trillion of them. It is essential that we design a GPU capable of handling floating point numbers to this accuracy.
Since all matter is ultimately reducible to energy, and energy must travel at C, then it follows the brain must explode at 3.pi.C/7 m/s.
The easiest explanation is that the particle has no fixed position but exists over a probability wave. It can be anywhere on that wave, with the probability equal to the value of the wave at that point. Thus, it is a particle and a wave, but cannot be treated as both at the same time (which is exactly what we observe). The certainty with which we know the position also decreases the certainty of its velocity since it will reside ANYWHERE on the wave on the next instant. Since velocity is rate of change of displacement, we cannot know the velocity. However, if your concern is with velocity, you can measure how far the wave has moved. In so doing, the exact position of the electron on the wave is known only by the probability and so is unknown. For all intents and purposes, it doesn't have both a simultaneous velocity and position in this description (which is also what we observe).
This is not the "standard model" (well, not unless I just reverse-engineered it in under a minute - and, if so, you can sent the Nobel prize to the following addres....) but it IS an easy-to-master description that explains Heisenberg's uncertainty principle, wave-particle duality, diffraction patterns with extremely low photon counts, quantum tunneling and many of the other strange effects described at that level.
Since electrons have mass, they can't be indivisible. They have, at the very least, to be divisible into a Higgs boson and whatever is left over.