Humans are very good at creating artificial boundaries between subjects. It's useful for categorization, but it can be misleading. In this case, I think it's misleading. Anatomy is but one tiny field in human biology, and human biology is but a tiny speck in the field of cellular biology, which in turn is but one special case.
Is the whole field of living systems infinite? Quite probably yes. That does not mean each and every subcategory is also infinite, although it can be.
Of course, there are some questions. For example, is human anatomy so well understood that it can be called a "dead" subject? (Dead in the sense of it is stagnant, it won't evolve or grow, in the same way that Latin is a dead subject, even though it has a lot of value in some disciplines.) Maybe, maybe not. Depends on how detailed something can be and still be anatomy and not some other branch of human biology. We've not mapped out all the regions of the brain, and we know from studies of tetrachromats that the brain can form in multiple ways, reassigning portions of itself. The anatomy of the brain is therefore not absolute, but is much more amorphous. Unless we know every possible condition that can physically reshape the brain, that is definitely not a dead area of research.
The brain seems to have mechanisms for which we have no identified pathway. For example, aluminium in the brain will cause the formation of tau protein knots - apparently some archaic form of protective mechanism against the metal that is normally disabled. Nobody, as far as I know, has identified what part of the brain is responsible for this protection, what re-enables it or when in the evolutionary trail it would have had any value. (In modern humans, it causes the brain to crush itself from the inside.)
Anything else? The stomach still causes problems - and, no, antacid won't fix them. Going by current structural knowledge, it is impossible for a prion to traverse the stomach wall. Current research on new-variant CJD requires that they can. Is our structural understanding incomplete or wrong?
Then there is the issue of stem cells. The body produces a lot of adult stem cells, but stem cells aren't specialized and it is unclear to me that they should be considered part of a structure that they have not yet joined.
I'm inclined to say that anatomy is close to being a dead subject, but that there are significant areas of uncertainty that need to be cleared up before research should be considered truly finished.
Of course it's FUD. This is intended to scare corporations into paying Microsoft a leave-me-alone fee... which I believe puts it into the category of a Protection Racket. It may also be intended to scare the EU, as Linux is the main competitor to Microsoft in the corridors of power there: "Drop the fines and we'll not come after you".
These charges are clearly designed to intimidate non-technical people who lack the knowledge to understand the claims or the time to do the research. I seriously doubt the DOJ could be talked into investigating under racketeering laws, but with things as they stand, it might be possible for EU Linux organizations to file a "friend of the court" on Microsoft's appeals. It's awfully hard for Microsoft to claim a willingness to work with others on interoperability if anything interoperable is automatically claimed as Microsoft property.
(It might - just might - persuade the appeals court that Microsoft can have no defense against illegal monopolistic practices and anticompetitiveness, if they insist on claiming that they actually own their only meaningful competitor.)
And yes, it's blatantly absurd to imagine Microsoft can have any claims to IP on framebuffer support within the OS. That's what makes it a good illustration of their malpractices. (The framebuffer and KGI are the only two "Linux GUIs" - X11 isn't limited to running on Linux, predates Windows and is merely packaged with the kernel, it is not part of Linux itself.) Since they clearly and expressly state IP on something that is quite impossible for them to have IP on, and since their legal department is hardly stupid, it is clear that whatever it is that they are referring to has nothing to do with whatever they are saying.
Now, false accusations for the purpose of extorting money is a very serious matter. If they are deliberately confusing issues that have nothing to do with Linux with that OS, that would be grounds for a counter-suit from hell. The same would be true if Microsoft can be shown to have made a fictional claim for the purpose of intimidating customers. I imagine the OLPC project can't be happy with this - remember, customers won't be sued according to Microsoft, but the OLPC group isn't a customer. Dell can't be happy, either, and that is presumably why they're paying up the protection money.
The thing Dell has to remember, and what the English learned over one and a half thousand years ago, was that anyone who extorts successfully once will continue extorting. (The English call the money involved "Dane Gold", in honor of the fact that the Danes were very good at parking warships off the English coast and demanding gold in return for not attacking. The Danes got so good at it that they almost never needed to actually attack. They made far better money from threatening.)
Oh, I dunno about that. In this day and age, those kinds of skills are actually quite valuable. If you're looking in the right sectors. Look at the contortions ATLAS has to go through to optimize itself to different architectures and cache sizes. This couldn't be done at all if the maintainer wasn't in the borderlands of genius at hand-optimizing, as the optimized versions don't write themselves. The same is arguably true of compiler optimizations - you can't tell a computer how to do something you don't know how to do.
What about pile-of-pc supercomputers? Each node HAS to be tightly coded - you just can't get away with high per-node overheads on such systems (which is one reason it is very hard to make such systems work well). Fast graphics (where the libraries, code, GUI and kitchen sink are all loaded into the GPU) is not trivial by any means, either.
Yes, many people write in C. Some maths people still use Fortran, because the compilers have evolved for longer and so have better-understood optimizations. But if you want fast, you work in assembly or raw machine code. Compilers have improved enormously over the years and so have processors, but amazingly many programs are no faster now than they were in the 80s - or are even slower. Bloat has far surpassed the combined improvements in software and hardware.
Even if you don't code in raw machine code, having the skill is what makes the difference between a coder that writes code people crave versus a coder that writes code people merely use.
(For those who want to know what old-time machines were like to program, there are simulators for the Manchester Mk. 1 - a 40-bit system from the 1940s - and other early computers on various historic computer sites. Some of the Mk. 1 programs are amazing in their sophistication, given that there were a total of 7 instructions in the entire instruction set - and no add operator.)
It's the framebuffer device. Ok, it's not much of a GUI, but it is graphical (makes nice penguin pics at boot time) and it is used as a user interface. So there. At a stretch, you could also include KGI on much the same grounds, although it is provided as a distinct patch and not rolled into the main kernel.
It is perfectly true that X11 is not a part of Linux - although most of it is old enough to be considered prior art with respect to any patents Microsoft may have, so it wouldn't even matter if it was. About the only "new" GUI is Berlin and that is no longer under development.
I'm not convinced that that is a valid conclusion. There are many systems (such as chaotic systems) that are entirely deterministic but also entirely non-predictable. The two states are not mutually exclusive. It is entirely possible that the result of any given collision of particles is 100% deterministic, but that because we cannot measure the initial conditions, we cannot predict what the result would be.
An alternative view is that when you talk about a "probability" in quantum mechanics, you are really talking about the fraction of the wave function that describes something that exists in the state you are considering the probability for. In that case, all possible outcomes occur, but you only see the dominant one. (This is slightly better than the "many worlds" theory, in that it doesn't require the spontaneous creation of universes.) Here, there are not really "different outcomes", just different ratios, which may or may not be deterministic. This interpretation allows for "randomness", but in a different sense than is normally considered, but could also be treated as a chaotic system.
A third view is the "many worlds" theory. Since every possibility is mapped out into a gigantic n-ary tree of possibilities, it is entirely valid to say that every given path is fixed and that the tree is static. What varies is the path you take through it. Is that a fixed future or not? The path you take isn't fixed, but the result of whatever path you do take is.
Personally, I dislike the "many worlds" theory, it smacks too much of Fred Hoyle's "October the First Is Too Late". Ugh. Ghastly. Since the second option can be treated as chaos theory, it's really just a minor variant of the first. In either case, the future would indeed be entirely predetermined but it would also be unknowable.
Is that likely? Well, does it matter? If the future is not knowable, under the constraints of chaos, it is utterly unimportant for any practical purposes as to whether it is fixed. It may be interesting to know, but it's really not useful - either experimentally or experientially. The only value would be that some probabilities in QM may be replaceable by the corresponding set of Strange Attractors and chaotic systems. However, even if you could do that, unless this led to a simpler model, there really wouldn't be any point. If probabilities are a simpler way to describe things and produce the same results, then whether it is mechanically correct or not, it is the best way to describe things.
A point has zero dimensions, which means the real component of the Hausdorf Dimension cannot be strictly less than the real component of the Euclidian Dimension (it can only be equal), which in turn means that you cannot have zero-dimensional fractals.
There is a case for the existence of imaginary components to dimensions (it's how you avoid a singularity at the big bang), but it is unclear to me if an object at 0K would remain a point if viewed from a different perspective. This is important. For all fractals, not only does the size depend on the scale you are measuring with, but so does the number of apparent dimensions. In cases where the size is an abstract concept, it can be easier to measure the apparent number of dimensions from different perspectives and see if it varies.
The usual example Mandelbrot gives is a ball of string. As seen from far enough away, it is a point. Zero dimensions. Move closer, and it is a disk. Two dimensions are needed to describe everything you can define at this distance. Closer still, and it becomes a ball. Three dimensions are now required. Yet closer and you can make out the line of the thread. A line is one dimensional. There is no need to define anything other than the position on the line, in order to define the entire object, so the object now has one apparent dimension. Closer still and the line acquires a width. Back to two dimensions. As you zoom in, the thread becomes fully three dimensional.
Depends. If this means that the next time Microsoft is caught pirating software to include in their OS or office products, the DOJ will confiscate their property, bank accounts and license to operate as a corporate entity in the US, it might be interesting. If it means that the RIAA and MPAA will face stiffer penalties for copyright violations and intellectual property theft, I might be persuaded to be in favour.
You know what? Somehow, I don't see these as being the people who will get targeted. Odd, that, given that very few thieves are going to give a damn about a show-trial of some insignificant individual, and that said individual might be responsible for maybe a couple of dozen minor infractions, usually on a personal level. When a megacorp is involved, start thinking not in tens, but tens of millions. If some future version of Microsoft Office contains pirated software, stop to consider that 98% of all desktop users could be installing that.
Think also of the deterrent value. A twelve-year-old gets arrested. Probably won't even deter their friends. Might even encourage them to stick it to "the man". A multi-billion-dollar company gets skewered - especially if they've "friends" in high places - and people will notice.
So why is this effort more likely to target the twelve-year-olds? Easy. This isn't about IP. Never has been. If it was, half the Hollywood studios would have filed for bankruptcy years ago. No, this is about creating an atmosphere of fear and/or paranoia. Why? To control people? No. The Government is paranoid, and paranoids are incapable of controlling themselves, much less anyone else. However, paranoids are very good at spreading that paranoia around. That's all that's happening here. It's nothing more than a contagious mental disease being coughed around.
These "delusional cultists" appear to have been following the BBC reporter involved. Now, let us say for the sake of argument that 99.999% of them are perfectly rational, law-abiding folk. That's actually a damn sight better than any other group, so no matter what your viewpoint, this is generous.
There are therefore 0.001% of Scientologists who are NOT perfectly rational, law-abiding folk, who now have access to information on that reporter's every move, and who would have every excuse possible to deal with the heresy problem.
I dislike Scientology, but Scientology is actually not the issue here. The issue here is that a situation has been created that is extremely dangerous to all parties. It wouldn't even take a crazy inside the organization. All it would take is for someone - anyone - who could gain access to the information to do so and act on it. It would make almost perfect cover for a criminal - they wouldn't be the ones on the suspect list. By the time the police had figured out it wasn't those spying on him, the criminal would be long gone.
This is why fear and intimidation make extremely bad weapons. They can be used against the wielder as easily as they can against the target. No matter what the rights and wrongs of the situation, such weapons MUST be removed from the arena.
There are a number of cancers which leave free-floating cancerous clumps of cells in the bloodstream. Patients with such cancers often get extra chemotherapy injected into the spine to stop it reaching there. A free-floating cancer clump would seem to be easier to filter with this sort of sponge than an individual virus.
Would it make more sense for these folks to use the product on a market that actually exists right now, so that they can refine and develop the idea further for viruses who have not yet evolved to be transmissible between humans and therefore whose lethal form is as yet unknowable?
(It sounds a great idea, but great ideas need to be researched thoroughly, which isn't cheap. Free-floating cancers could be a potential source of revenue between now and when it's needed for a viral epidemic.)
Once the square kilometer array has been constructed, we will be able to get maps of this level of detail on Earth-sized planets at 1 AU from its sun at distances of around 50 light-years or so. If they made it a square mile, they could do the same at a distance of 100 light-years.
Oh, one small correction. We'd be able to get continuous maps. In other words, you'd be able to see temperature (and atmosphere composition) changes over time.
(At that point, something like SETI would get potentially much less of an intragalactic lottery and be much more interesting.)
I honestly don't know what resolution the NASA folks are working at, but the image presented is almost certainly some form of interpolation from available data, as even a super-Jovian planet is far too small to get more than a pixel or two resolution at any distance. I don't quite know what they're calculating, rather than directly observing, but there is simply no way they're getting that kind of resolution with direct measurement. Not of a planet.
There's nothing wrong with mathematical techniques, and if they're as good as they seem to be, then obviously the square kilometer array will be able to resolve Earth-sized planets at greater resolution than initially expected. Which is good, so long as the methods applied are valid.
(Let's face it - if anyone quibbles with these results, who do you think NASA would send over there to personally check? The person getting fame and glory for the organisation, or the person who is spoiling the party?)
Since heavy stars should not have formed that early on, it is clearly an illegal immigrant from another universe and therefore not eligible for Social Security. Indeed, reports are that a team of astrophysicists has been deployed to deport it to an acceptable parallel realm of existence.
That's the best you could find? Then consider what MSN's store is selling. NOW you know why nobody talks about it......if they know what's good for them.... Bwahahahahahahaha!
You are correct about the level of awareness and about the risks. Mobile IP and NEMO (mobile networks) are potentially serious hazards to the infrastructure, if implemented incorrectly - which seems to be a very likely thing to happen. I've offered Yet Another alternative below as to how you could do this without having to tamper too much with the way things are already done. I needed to make just a couple of minor tweaks and had no need for any kind of source-based routing or self-directing traffic. This is not intended as necessarily the "right solution", but merely one way you can do mobile IP without needing such hacks.
I always considered the one destination thing to be rather crude. But let's say that we keep that (for now), but modify the semantics. Instead of there being one physical endpoint, have one group endpoint in which exactly one member of the group will respond. (Then you can use all the conventional reliability mechanisms and all that carp.)
How would this work? The standard "unicast" packet would be transported a-la multicast to all members of the group. In all cases, the group will have exactly one physical member - your machine. Your machine gets the packet, responds as it would to any other inbound packet, and nothing changes. Everything that currently works would continue to work.
Your machine migrates to another network and rejoins the group. The packet temporarily gets sent to both destinations, but there is still only one recipient - your machine. Hence, there is still only one response to each inbound packet. The old connection gets pruned and ceases to operate.
This would absolutely positively require that the host machine is strongly authenticated, or you'd get people sniffing traffic simply by joining other people's groups. However, strong authentication is certainly doable. I don't see any serious problem there, IPSec, SSL and TLS can provide authenticated connections. Actually, these simplify things a little because then it really wouldn't matter who could wiretap. The strongest encryption ciphers and modes available for these have no (publicly) known flaws.
The next change would be that you would be using unicast semantics and unicast protocols but routing via multicast to ensure that the traffic got to your new endpoint. That one is trickier but I don't see any fundamental objection. Routing is routing. So you duplicate the traffic at times. Big deal.
Finally, you would need to own the group address. Your physical address could be assigned by the local network and you wouldn't care. Multicast routing doesn't need to transport endpoints, so so long as your LAN can identify your machine as the group owner, your physical address can remain totally invisible to the rest of the network.
Are things like this done? Sure. It's very similar to the self-optimizing routing scheme originally proposed for mobile IP, and has a lot in common with anycasting in that you are targeting a group in order to reach a single machine. It's also very similar to wireless routing techniques where the transmission has to be damn-near guaranteed to get to the endpoint where (for one reason or another) you can't be sure what path gets to that endpoint.
I seriously doubt anyone will implement this, not because of any obvious flaws (although there are probably some not-so-obvious ones), but because if you thought IPv6 was taking its time getting deployed, that would be a cinch compared to telling the Internet backbone to dump unicast routing entirely and add the compute power needed to handle dynamic groups at this kind of level. (Best guess is that it's going to double CPU requirements on average.)
They already deploy IPv6 nationally. Just because the US domestic market is more sluggish than a salted slug, it would be wrong to assume everyone else is as bad.
What's more, IPv4+NAT (as standard) doesn't give you half the features of IPv6. I've listed them before, I'll list them again here. Sure, not many use them NOW, but most of these are major areas of growth and Internet-aware devices will (sooner or later) have to use IPv6 to get the support they need.
IPSec
Anycasting
Multicasting the ISPs can't turn off
Mobile IP
Mobile Networks
Extensible Headers
Router Discovery
Automatic Configuration
Per-destination MTU optimization
There are probably a whole bunch of other advantages not listed here. Go to your local USAGI dealership and test drive an IPv6 today.
Originally, IPv6 handled mobile IP by migrating the routing information up through the routers, and by using transitional IP addressing. You kept the same suffix, not the same address, as you moved from network to network. But for some certain length of time, you had both the old address and the new one. This allowed for a totally clean transition and has the same observable effect as source-based routing, but is not subject to this DDoS attack strategy.
IIRC, the main reason the transitional scheme was dropped was because routers would need to track more states. Like they're not going to be tracking gigantic numbers of states in order to have a workable authenticated source-routing system.
However, there is one good thing about this. People might finally realize IPv6 is NOT an addressing scheme, it is a very powerful protocol. (Would you believe I had to correct a senior network engineer on that yesterday?)
The machines these days are digital and have a right to be filled. Thus, by not stocking the machines, they are managing to violate the machine's digital rights. This has to be covered by the DMCA somewhere.
(1) is already done in England, extensively, and is the main reason it's less lawsuit-happy than the US.
The last thing we want is for lawyers to be in prisons. THOUSANDS of potential clients, all with nowhere to run. No, that's just not fair.
Re:So lemme get this straight....
on
The Human Mutation
·
· Score: 4, Interesting
One obvious impact would be to look and see if this gene has undergone any further mutations - however trivial - or whether the associated junk DNA has. Of particular interest would be polyglots or other people with exceptional ability in communicating and understanding. Also of interest would be archaeological DNA where the relevant protein has survived. (It's rare for Y chromosomes to survive hundreds or thousands of years, but every so often it happens. Maybe this gene can also survive.)
I'm assuming here that the mutation is involved in communication, as I know that the wiring in the front of the brain is linked to autism, which impacts the brain's I/O channels, and I/O is a major difference between apes and humans. However, this is an assumption and should be taken as such.
We know that the ability to filter information has changed over time. Some of that has been changes elsewhere in the brain, but there is no advantage in a brain adapting to process information it hasn't got. Whereas, we already know from tetrachromats and synesthetes that there IS a usable advantage in getting information that would not normally be processed. If this gene is responsible for improving I/O bandwidth, then we should see a series of minor mutations over time that correspond to known I/O improvements within the brain.
Could this be useful in some other way? Well, provided (a) it is involved in I/O enhancements, and (b) we can understand the relationship between changes within it and those enhancements, it should be possible to induce mutations that can improve the brain further, provided the change did not exceed the brain's ability to adapt.
Slashdot Mirror
Is the whole field of living systems infinite? Quite probably yes. That does not mean each and every subcategory is also infinite, although it can be.
Of course, there are some questions. For example, is human anatomy so well understood that it can be called a "dead" subject? (Dead in the sense of it is stagnant, it won't evolve or grow, in the same way that Latin is a dead subject, even though it has a lot of value in some disciplines.) Maybe, maybe not. Depends on how detailed something can be and still be anatomy and not some other branch of human biology. We've not mapped out all the regions of the brain, and we know from studies of tetrachromats that the brain can form in multiple ways, reassigning portions of itself. The anatomy of the brain is therefore not absolute, but is much more amorphous. Unless we know every possible condition that can physically reshape the brain, that is definitely not a dead area of research.
The brain seems to have mechanisms for which we have no identified pathway. For example, aluminium in the brain will cause the formation of tau protein knots - apparently some archaic form of protective mechanism against the metal that is normally disabled. Nobody, as far as I know, has identified what part of the brain is responsible for this protection, what re-enables it or when in the evolutionary trail it would have had any value. (In modern humans, it causes the brain to crush itself from the inside.)
Anything else? The stomach still causes problems - and, no, antacid won't fix them. Going by current structural knowledge, it is impossible for a prion to traverse the stomach wall. Current research on new-variant CJD requires that they can. Is our structural understanding incomplete or wrong?
Then there is the issue of stem cells. The body produces a lot of adult stem cells, but stem cells aren't specialized and it is unclear to me that they should be considered part of a structure that they have not yet joined.
I'm inclined to say that anatomy is close to being a dead subject, but that there are significant areas of uncertainty that need to be cleared up before research should be considered truly finished.
America has no plans to liberate the space/time continuum, but if Cronos won't step down, there will be a regime change.
These charges are clearly designed to intimidate non-technical people who lack the knowledge to understand the claims or the time to do the research. I seriously doubt the DOJ could be talked into investigating under racketeering laws, but with things as they stand, it might be possible for EU Linux organizations to file a "friend of the court" on Microsoft's appeals. It's awfully hard for Microsoft to claim a willingness to work with others on interoperability if anything interoperable is automatically claimed as Microsoft property.
(It might - just might - persuade the appeals court that Microsoft can have no defense against illegal monopolistic practices and anticompetitiveness, if they insist on claiming that they actually own their only meaningful competitor.)
And yes, it's blatantly absurd to imagine Microsoft can have any claims to IP on framebuffer support within the OS. That's what makes it a good illustration of their malpractices. (The framebuffer and KGI are the only two "Linux GUIs" - X11 isn't limited to running on Linux, predates Windows and is merely packaged with the kernel, it is not part of Linux itself.) Since they clearly and expressly state IP on something that is quite impossible for them to have IP on, and since their legal department is hardly stupid, it is clear that whatever it is that they are referring to has nothing to do with whatever they are saying.
Now, false accusations for the purpose of extorting money is a very serious matter. If they are deliberately confusing issues that have nothing to do with Linux with that OS, that would be grounds for a counter-suit from hell. The same would be true if Microsoft can be shown to have made a fictional claim for the purpose of intimidating customers. I imagine the OLPC project can't be happy with this - remember, customers won't be sued according to Microsoft, but the OLPC group isn't a customer. Dell can't be happy, either, and that is presumably why they're paying up the protection money.
The thing Dell has to remember, and what the English learned over one and a half thousand years ago, was that anyone who extorts successfully once will continue extorting. (The English call the money involved "Dane Gold", in honor of the fact that the Danes were very good at parking warships off the English coast and demanding gold in return for not attacking. The Danes got so good at it that they almost never needed to actually attack. They made far better money from threatening.)
What about pile-of-pc supercomputers? Each node HAS to be tightly coded - you just can't get away with high per-node overheads on such systems (which is one reason it is very hard to make such systems work well). Fast graphics (where the libraries, code, GUI and kitchen sink are all loaded into the GPU) is not trivial by any means, either.
Yes, many people write in C. Some maths people still use Fortran, because the compilers have evolved for longer and so have better-understood optimizations. But if you want fast, you work in assembly or raw machine code. Compilers have improved enormously over the years and so have processors, but amazingly many programs are no faster now than they were in the 80s - or are even slower. Bloat has far surpassed the combined improvements in software and hardware.
Even if you don't code in raw machine code, having the skill is what makes the difference between a coder that writes code people crave versus a coder that writes code people merely use.
(For those who want to know what old-time machines were like to program, there are simulators for the Manchester Mk. 1 - a 40-bit system from the 1940s - and other early computers on various historic computer sites. Some of the Mk. 1 programs are amazing in their sophistication, given that there were a total of 7 instructions in the entire instruction set - and no add operator.)
It is perfectly true that X11 is not a part of Linux - although most of it is old enough to be considered prior art with respect to any patents Microsoft may have, so it wouldn't even matter if it was. About the only "new" GUI is Berlin and that is no longer under development.
An alternative view is that when you talk about a "probability" in quantum mechanics, you are really talking about the fraction of the wave function that describes something that exists in the state you are considering the probability for. In that case, all possible outcomes occur, but you only see the dominant one. (This is slightly better than the "many worlds" theory, in that it doesn't require the spontaneous creation of universes.) Here, there are not really "different outcomes", just different ratios, which may or may not be deterministic. This interpretation allows for "randomness", but in a different sense than is normally considered, but could also be treated as a chaotic system.
A third view is the "many worlds" theory. Since every possibility is mapped out into a gigantic n-ary tree of possibilities, it is entirely valid to say that every given path is fixed and that the tree is static. What varies is the path you take through it. Is that a fixed future or not? The path you take isn't fixed, but the result of whatever path you do take is.
Personally, I dislike the "many worlds" theory, it smacks too much of Fred Hoyle's "October the First Is Too Late". Ugh. Ghastly. Since the second option can be treated as chaos theory, it's really just a minor variant of the first. In either case, the future would indeed be entirely predetermined but it would also be unknowable.
Is that likely? Well, does it matter? If the future is not knowable, under the constraints of chaos, it is utterly unimportant for any practical purposes as to whether it is fixed. It may be interesting to know, but it's really not useful - either experimentally or experientially. The only value would be that some probabilities in QM may be replaceable by the corresponding set of Strange Attractors and chaotic systems. However, even if you could do that, unless this led to a simpler model, there really wouldn't be any point. If probabilities are a simpler way to describe things and produce the same results, then whether it is mechanically correct or not, it is the best way to describe things.
There is a case for the existence of imaginary components to dimensions (it's how you avoid a singularity at the big bang), but it is unclear to me if an object at 0K would remain a point if viewed from a different perspective. This is important. For all fractals, not only does the size depend on the scale you are measuring with, but so does the number of apparent dimensions. In cases where the size is an abstract concept, it can be easier to measure the apparent number of dimensions from different perspectives and see if it varies.
The usual example Mandelbrot gives is a ball of string. As seen from far enough away, it is a point. Zero dimensions. Move closer, and it is a disk. Two dimensions are needed to describe everything you can define at this distance. Closer still, and it becomes a ball. Three dimensions are now required. Yet closer and you can make out the line of the thread. A line is one dimensional. There is no need to define anything other than the position on the line, in order to define the entire object, so the object now has one apparent dimension. Closer still and the line acquires a width. Back to two dimensions. As you zoom in, the thread becomes fully three dimensional.
You know what? Somehow, I don't see these as being the people who will get targeted. Odd, that, given that very few thieves are going to give a damn about a show-trial of some insignificant individual, and that said individual might be responsible for maybe a couple of dozen minor infractions, usually on a personal level. When a megacorp is involved, start thinking not in tens, but tens of millions. If some future version of Microsoft Office contains pirated software, stop to consider that 98% of all desktop users could be installing that.
Think also of the deterrent value. A twelve-year-old gets arrested. Probably won't even deter their friends. Might even encourage them to stick it to "the man". A multi-billion-dollar company gets skewered - especially if they've "friends" in high places - and people will notice.
So why is this effort more likely to target the twelve-year-olds? Easy. This isn't about IP. Never has been. If it was, half the Hollywood studios would have filed for bankruptcy years ago. No, this is about creating an atmosphere of fear and/or paranoia. Why? To control people? No. The Government is paranoid, and paranoids are incapable of controlling themselves, much less anyone else. However, paranoids are very good at spreading that paranoia around. That's all that's happening here. It's nothing more than a contagious mental disease being coughed around.
Zen's performance issues were fixed by Avon, under Orac's guidance.
I think there was a character he forgot to add, except his readers would then be asking where you could buy the X-rated Windows from.
There are eddies in the wash of the space/time continuum.
...it was the rocket scientists who didn't know, so that means you ARE a rocket scientist!
There are therefore 0.001% of Scientologists who are NOT perfectly rational, law-abiding folk, who now have access to information on that reporter's every move, and who would have every excuse possible to deal with the heresy problem.
I dislike Scientology, but Scientology is actually not the issue here. The issue here is that a situation has been created that is extremely dangerous to all parties. It wouldn't even take a crazy inside the organization. All it would take is for someone - anyone - who could gain access to the information to do so and act on it. It would make almost perfect cover for a criminal - they wouldn't be the ones on the suspect list. By the time the police had figured out it wasn't those spying on him, the criminal would be long gone.
This is why fear and intimidation make extremely bad weapons. They can be used against the wielder as easily as they can against the target. No matter what the rights and wrongs of the situation, such weapons MUST be removed from the arena.
There are a number of cancers which leave free-floating cancerous clumps of cells in the bloodstream. Patients with such cancers often get extra chemotherapy injected into the spine to stop it reaching there. A free-floating cancer clump would seem to be easier to filter with this sort of sponge than an individual virus.
Would it make more sense for these folks to use the product on a market that actually exists right now, so that they can refine and develop the idea further for viruses who have not yet evolved to be transmissible between humans and therefore whose lethal form is as yet unknowable?
(It sounds a great idea, but great ideas need to be researched thoroughly, which isn't cheap. Free-floating cancers could be a potential source of revenue between now and when it's needed for a viral epidemic.)
Oh, one small correction. We'd be able to get continuous maps. In other words, you'd be able to see temperature (and atmosphere composition) changes over time.
(At that point, something like SETI would get potentially much less of an intragalactic lottery and be much more interesting.)
I honestly don't know what resolution the NASA folks are working at, but the image presented is almost certainly some form of interpolation from available data, as even a super-Jovian planet is far too small to get more than a pixel or two resolution at any distance. I don't quite know what they're calculating, rather than directly observing, but there is simply no way they're getting that kind of resolution with direct measurement. Not of a planet.
There's nothing wrong with mathematical techniques, and if they're as good as they seem to be, then obviously the square kilometer array will be able to resolve Earth-sized planets at greater resolution than initially expected. Which is good, so long as the methods applied are valid.
(Let's face it - if anyone quibbles with these results, who do you think NASA would send over there to personally check? The person getting fame and glory for the organisation, or the person who is spoiling the party?)
Since heavy stars should not have formed that early on, it is clearly an illegal immigrant from another universe and therefore not eligible for Social Security. Indeed, reports are that a team of astrophysicists has been deployed to deport it to an acceptable parallel realm of existence.
That's the best you could find? Then consider what MSN's store is selling. NOW you know why nobody talks about it... ...if they know what's good for them.... Bwahahahahahahaha!
I always considered the one destination thing to be rather crude. But let's say that we keep that (for now), but modify the semantics. Instead of there being one physical endpoint, have one group endpoint in which exactly one member of the group will respond. (Then you can use all the conventional reliability mechanisms and all that carp.)
How would this work? The standard "unicast" packet would be transported a-la multicast to all members of the group. In all cases, the group will have exactly one physical member - your machine. Your machine gets the packet, responds as it would to any other inbound packet, and nothing changes. Everything that currently works would continue to work.
Your machine migrates to another network and rejoins the group. The packet temporarily gets sent to both destinations, but there is still only one recipient - your machine. Hence, there is still only one response to each inbound packet. The old connection gets pruned and ceases to operate.
This would absolutely positively require that the host machine is strongly authenticated, or you'd get people sniffing traffic simply by joining other people's groups. However, strong authentication is certainly doable. I don't see any serious problem there, IPSec, SSL and TLS can provide authenticated connections. Actually, these simplify things a little because then it really wouldn't matter who could wiretap. The strongest encryption ciphers and modes available for these have no (publicly) known flaws.
The next change would be that you would be using unicast semantics and unicast protocols but routing via multicast to ensure that the traffic got to your new endpoint. That one is trickier but I don't see any fundamental objection. Routing is routing. So you duplicate the traffic at times. Big deal.
Finally, you would need to own the group address. Your physical address could be assigned by the local network and you wouldn't care. Multicast routing doesn't need to transport endpoints, so so long as your LAN can identify your machine as the group owner, your physical address can remain totally invisible to the rest of the network.
Are things like this done? Sure. It's very similar to the self-optimizing routing scheme originally proposed for mobile IP, and has a lot in common with anycasting in that you are targeting a group in order to reach a single machine. It's also very similar to wireless routing techniques where the transmission has to be damn-near guaranteed to get to the endpoint where (for one reason or another) you can't be sure what path gets to that endpoint.
I seriously doubt anyone will implement this, not because of any obvious flaws (although there are probably some not-so-obvious ones), but because if you thought IPv6 was taking its time getting deployed, that would be a cinch compared to telling the Internet backbone to dump unicast routing entirely and add the compute power needed to handle dynamic groups at this kind of level. (Best guess is that it's going to double CPU requirements on average.)
You still type IP addresses? Most machines have names and if the name's too long, you can always add a bookmark to /etc/hosts.
What's more, IPv4+NAT (as standard) doesn't give you half the features of IPv6. I've listed them before, I'll list them again here. Sure, not many use them NOW, but most of these are major areas of growth and Internet-aware devices will (sooner or later) have to use IPv6 to get the support they need.
There are probably a whole bunch of other advantages not listed here. Go to your local USAGI dealership and test drive an IPv6 today.
IIRC, the main reason the transitional scheme was dropped was because routers would need to track more states. Like they're not going to be tracking gigantic numbers of states in order to have a workable authenticated source-routing system.
However, there is one good thing about this. People might finally realize IPv6 is NOT an addressing scheme, it is a very powerful protocol. (Would you believe I had to correct a senior network engineer on that yesterday?)
The machines these days are digital and have a right to be filled. Thus, by not stocking the machines, they are managing to violate the machine's digital rights. This has to be covered by the DMCA somewhere.
The last thing we want is for lawyers to be in prisons. THOUSANDS of potential clients, all with nowhere to run. No, that's just not fair.
I'm assuming here that the mutation is involved in communication, as I know that the wiring in the front of the brain is linked to autism, which impacts the brain's I/O channels, and I/O is a major difference between apes and humans. However, this is an assumption and should be taken as such.
We know that the ability to filter information has changed over time. Some of that has been changes elsewhere in the brain, but there is no advantage in a brain adapting to process information it hasn't got. Whereas, we already know from tetrachromats and synesthetes that there IS a usable advantage in getting information that would not normally be processed. If this gene is responsible for improving I/O bandwidth, then we should see a series of minor mutations over time that correspond to known I/O improvements within the brain.
Could this be useful in some other way? Well, provided (a) it is involved in I/O enhancements, and (b) we can understand the relationship between changes within it and those enhancements, it should be possible to induce mutations that can improve the brain further, provided the change did not exceed the brain's ability to adapt.
You're right. There's no way Professor X will let that many Slashdotters hang out in his house.