Matter on the way into a black hole gets compressed by the immense gravity to tremendous pressures, forming an accretion disk, which radiates mostly x-rays and similar forms of high-energy radiation. It isn't really the black holes that are radiating, but the matter falling in that does. Tossing matter into a black hole is probably a far more efficient way of converting matter into energy than even nuclear fusion.
Hawking radiation probably isn't what's going on here. The temperature of a typical stellar mass black hole is about 10^-8 kelvin, way colder than the cosmic background radiation at 2.7 K, so it actually gains more energy from the background than it loses from Hawking radiation. Larger black holes are colder still. It's the really really small black holes (such as the hypothesized primordial black holes might have formed by small irregularities in the early universe that collapsed in on themselves; typical one squeezes the mass of a large mountain into a space a quarter the size of a proton) which are supposed to radiate significant quantities of Hawking radiation.
For example, why hasn't someone made an email encryption program where you press "encrypt" and it automatically checks public key databases, locates public keys for the recipient and automatically encrypts the email? And I mean a solution that does NOT require purchasing any kind of certificate in order for it to work. (something that uses PGP/GPG as the underlying encryption would be good)
Four words: monkey in the middle. If you automatically receive public keys from public key databases, how would you know for sure that the key actually belonged to the recipient you intended? Maybe it's a key for someone pretending to be your intended recipient to you, and you could wind up communicating with your intended recipient through that monkey in the middle, who is pretending to be your contact to you, and pretending to be you to your contact, all the while reading everything you guys say and perhaps, subtly changing the content of your communication. Without some form of authentication of public keys, accomplished either by the web of trust model used by PGP/GPG (which you don't seem to fully understand, by presenting such a scheme, which would break the web of trust if used as you intend), or a centralized public key infrastructure based on root certification authorities, these sorts of attacks are possible. If you cannot have some reasonable idea of who it is you're really talking to, why should you bother encrypting your messages? You could be talking to the NSA anyway, and they wouldn't even need to break any cryptography, just pretend to be someone you know to you. In almost all real-world situations, authentication is almost always more important than encryption.
I hope that remains true. The sobering reality is that these two cherished boxes are gradually diminishing in their effectiveness with what's going on these days. These warrantless wiretaps and other infringements of Constitutional guarantees, unreliable electronic voting machines, and so forth are all conspiring to weaken the effectiveness of the soap and ballot boxes in your great country... Soon enough, only the ammo box will help, and God help us all if it ever comes to that.
Right, now that I've looked it up, but the issues are still the same it would seem. Burning these types of boards to get at the metal is still something that you really don't want to do. They typically use brominated flame retardants on these boards, which while they are generally inert and non-poisonous in their normal state, when burned they also produce deadly carcinogenic fumes, especially in the presence of copper. That makes that ingenious technique described in the article all the more useful: it separates the metallic and non-metallic components without burning.
The plastic used for most PCB's is polyvinyl chloride, which in itself isn't particularly toxic under most circumstances. It's in common use these days for plumbing fixtures and the like, as well as for PCB's. However, if you burn PVC plastic, it gets converted into some particularly nasty dioxins and furans which are dangerously carcinogenic.
There are not enough police, there are not enough soldiers, not enough courts, not enough prisons in the United States--or in any country in the world for that matter--to enforce a law that the public doesn't agree with. The art of governing is largely the art of convincing the governed that they are better off following the laws than not, especially if it is a law that touches almost everyone as these changes to copyright do. Thus far, these stalwarts have only succeeded in convincing the public only in a limited sense, and these heavy handed tactics that the RIAA and its ilk are using are, in a way, having the opposite effect.
The power of government really always is with the people, whether the form of government recognizes it or not.
Well, if it really is router firepower that's lacking, then perhaps it's way past time to seriously make the painful switch to IPv6. Contrary to many peoples' impression IPv6 isn't just about a larger IP address space, but also simpler routing. The core routers running BGP4 these days have about 200,000 entries, and that number is again growing exponentially. These kinds of monster routing tables are unnecessary with IPv6.
Public key encryption is, in practice, used pretty much the same way as well. Public key algorithms are generally used as part of a secure key exchange protocol rather than encrypting a message as directly.
Good point, but I think that part of the problem is that wireless bandwidth is a lot, lot less than land-based bandwidth. With the best WiMAX technology you can get at most 70 mbps: compare that with even an OC-3 fiber which is more than double that at 155 mbps, and if you lay an OC-192 you can get more than a hundred times the bandwidth of WiMAX.
Right on. The DRM problem on a general-purpose computer is, from a security standpoint, completelyimpossible. If I have absolute control over my entire computer, and this is still possible today because systems like TCPA haven't been forced down everyone's throats, then any attempt by anyone to restrict what bits I can and cannot copy is doomed to failure. And once I have done it, I can publish my break to the world if I so desire. These people might as well go on rolling a huge boulder up a hill, only to have its weight defeat them just as they are approaching the top. The task of DRM is equally futile. The only way that has a snowball's chance in hell of success would be to ban the general-purpose computer entirely (e.g. a TCPA-based media console), and it is doubtful that they will ever succeed in making such a thing happen, especially as the ban would have to be international in scope and there are some places (e.g. China) that don't care a whit about these matters.
You sound a lot like Ptolemy I (Soter) when he spoke to Euclid and asked if there was some kind of short cut to an understanding of geometry. Euclid's famous reply was: "There is no royal road to geometry." Computer science doesn't just have a lot to do with math: it IS math at its most fundamental. If you recall where the very idea for the automatic computer came from, it grew out of the quest for resolving the most profound issues at the foundations of mathematics (google for the 'Entscheidungsproblem' and Alan Turing). Just as there are no royal roads to geometry, there are no royal roads to computer science either.
If too many people in your educational system think that math is too hard, then maybe your educational system, especially at the lower levels, ought to be rethinking itself if it has as one of its priorities the training of more people who can really make use of computers and computer technology properly. I can't really say this with smug satisfaction as the case is the same here where I come from, but frankly the last place to go about fixing the problem is to water down the mathematical aspects of computer science as you're suggesting might be good. To do that would be to water down the essence of what computer science truly is.
Very, very clever. If I had mod points I'd give them! If Microsoft is really serious about doing this, then they will be doing the very antithesis of what they have been doing since, well, ever. Proprietary file formats anyone? Secret protocols? DRM? All of these things which they've been doing and promoting from the very beginning are precisely the sorts of things that will frustrate future digital archaeologists to no end. Consider the simple fact that we can still read Galileo's technical writings from the 1560's, but not Marvin Minsky's technical writings from the 1960's, thanks to proprietary storage hardware. Stuff is basically written on the wind these days, and Microsoft has done more than any single organization (largely because of their market monopoly) to make information as evanescent as it is now.
If it's non-acedemic (sic) to crack an MD5 hash, please tell me the plaintext for this: f6540dee6b248c863bb90fcaa784fef9
The term "plaintext" has no meaning for cryptographic hashes. The point of the attack is, if you had some text that hashed to say, f6540dee6b248c863bb90fcaa784fef9, I could, with even less work than I initially gathered, generate another set of data, saying something completely different, which also hashes to the same value. This has severe implications in the field of digital signatures, as the way most digital signature algorithms work is as follows:
Generate the hash of the document to be signed.
Encrypt the hash using the signer's private key.
The encrypted hash is the document's digital signature.
Now, if I wanted to verify the signed document:
Generate the hash of the document to be signed.
Decrypt the digital signature using the signer's public key.
Compare the computed hash with the hash recovered by decrypting the digital signature. If they match, the document is supposed to be authentic.
See the central role that hash functions play in this scheme? Now, if I'm able to generate collisions for the hash with a feasible amount of work, then given any digital signature you make using MD5 as the base hash I can create another document that might say something quite different, and attach the digital signature of the initial document to it, and it will look to anyone who cares as though you signed it. Of course, the document might look slightly strange to a person looking at it, but if the signature were used as part of a more complex cryptographic protocol, where only a computer ever really sees it, or if it's a document in a complicated file format (e.g. PostScript) that provides some leeway to add arbitrary data that is ignored by the viewer, then we might be in a spot of serious trouble. These collisions for MD5 are not just harmless, theoretical curiosities as you seem to think, as the page on X.509 certs and the PostScript examples illustrate.
The fact that collisions can be computed also destroys the non-repudiation characteristic of digital signatures. If this is important for your application, continued use of a weak hash function like MD5 is certainly out of the question, as the ability to generate collisions gives the alleged sender of a signed document a valid excuse to plausibly deny that she ever sent it. This of course makes MD5 digital signatures worthless from a legal standpoint.
Fortunately most people and began phasing out the use of MD5 for these and similar applications over the past several years, as it was already long suspected to harbor significant weaknesses. Now, we're hearing similar news about SHA-1 as what we heard about MD5 since roughly 1996, when cryptographers began recommending that alternative algorithms be used. Now, we're hearing the same thing again about SHA-1, and I think it would be wise to heed the advice again and start migrating to better hash functions as soon as possible.
How you got modded insightful is beyond me. This shit is real, very real, not just some propaganda from the Chinese. The attack on MD5 has been demonstrated by generating a couple of forged X.509 certificates based on the MD5 hash. It has long been suspected that MD5 harbored significant weaknesses, but it was confirmed in 2005 when Wang and her team demonstrated in a 2005 paper (warning PDF link) that it is possible to generate MD5 collisions with only about 2^39 hash computations (approx. 500 billion), a level of computational work which is doable in a matter of a few days even on the computer which I am using to type this post, and a very long way from the 2^64 computations required by a brute force attack. MD5 is well and truly broken, and not just in the academic sense, and anyone who says that the break doesn't affect the vast majority of its uses is either hopelessly uninformed or willfully ignorant. Checksums and digital signatures based on MD5 are now all suspect, and the only major application of the algorithm that remains unaffected is its use as a message authentication code, and the fact that the algorithm shown significant weakness in so many other areas should make anyone think twice before using it even for that. The biggest names in cryptography have been watching her work and that of her team with the keenest of interest, and there was an announcement (also here) that SHA-1 collisions could be found in 2^63 operations, which, while not feasible on my humble little PC, is within the realm of feasibility of today's fastest supercomputers and distributed computation clusters. Meaning that the NSA could probably generate SHA-1 collisions if they wanted to. Her most recent peer-reviewed paper on the subject gave a work factor of 2^69 for generating collisions, which while quite high, is quite a ways from the 2^80 required by true brute force, and that would make any serious cryptographer worried about using the algorithm.
This reminds me of an old Dr. Dobb's Journal article that I read more than a decade ago entitled "Personal Supercomputing" (I believe it was back in 1992 or thereabouts) where the author found a good use for a 486+i860 (remember that chip?) combo that involved making the i860 a computation engine and the 486 sort of like an I/O processor, and IIRC it was called PORT. The compiler set for this system didn't generate native i860 or x86 code, but instead compiled C or FORTRAN programs into a type of fixed-length instruction set tailored to the source language. The i860 would interpret this instruction set using a very efficiently hand-optimized interpreter that could fit almost entirely within the on-chip cache, reasoning that the frequent cache misses that come from executing RISC code directly are much more expensive than the interpretation overhead, and it seems that this observation was correct, at least in that case. Essentially, instead of using the i860 as a native RISC processor, the author used it as what could be considered a CISC processor with what amounted to programmable microcode inside its on-chip cache! The author even went so far as to say that this is the way RISC processors should really be used.
I wonder why no one has tried to use this same approach with more modern RISC architectures. I can see that the approach doesn't lend itself well to multitasking (the article was concerned with building supercomputers, to which multitasking is the very antithesis), but it is similar in principle to how VM's such as Java's and the.NET CLR work. It should also be noted that the instruction sets that the compilers that the PORT system used were memory transfer instructions designed in such a way that most individual statements in C or FORTRAN would compile to at most one instruction, rather than stack-based instruction sets like those used by Java and.NET.
But unfortunately, not all of the world has access to such wonderful libraries, and specialized research is somewhat difficult, even if your city is one that is blessed with a nice public library. Boy, I loved it when I discovered sites like this, and this, and this, collections to truly warm the heart of a math geek like me. Good luck finding even a tenth of the books and journals in those three collections in your local public library.
And that guy in the White House actually says he believes in this. How ironic that he should be the one to make it come to pass:
And he causeth all, both small and great, rich and poor, free and bond, to receive a mark in their right hand, or in their foreheads: And that no man might buy or sell, save he that had the mark, or the name of the beast, or the number of his name.
Very well said. Further extensions of copyright, with the way copyright works these days, is tantamount to stealing music from future generations. If the purpose of copyright is to encourage artists to continue creating more works, then the last thing we should do is allow them to rest on their laurels! The length of the copyright term is actually a disincentive to artists to make new works. It also hinders artists in the future from building on top of those works.
By the way, unbreakable DRM really is impossible: it's not even possible to make the problem mathematically infeasible. There is always a way around it provided you have control of all the hardware and software used to play back the protected media. As Bruce Schneier famously put it, making bits on a general-purpose computer uncopyable is like making water not wet. The only way to make DRM work absolutely would be to make true general-purpose computers illegal, or make it illegal to interface protected media to a general-purpose computer. As long as there is no law preventing people from circumventing DRM, someone will do it.
Have you ever tried contributing non-trivial portions of code to something under the GNU project or otherwise copyright the FSF? If you have, then you would know that the FSF is strict about copyright assignment, unlike Linus Torvalds, hence I don't believe that this will ever be a problem for any project of theirs. If some key developer were to break away, the code they try to start the fork from would not belong to them, but to the FSF. The forkers could maintain the fork's GPLv2 status, but that would not allow them to change the licensing conditions and eliminate the 'later version' clause: only the FSF would have the authority to do that.
The entire GNU toolchain will move to GPLv3, as well as all of the other software owned by the Free Software Foundation. That, you can definitely count on. Other software that depends on this vast base of software, much of which is fundamental to the operation of any GNU/Linux system, will probably be encouraged to switch with time. Uninformed projects probably are also uninformed enough to use the FSF's original statement of the GPLv2, which contains the 'later version' clause that the Linux kernel lacks, and thus GPLv3 terms may also apply to them if that is desired.
Microsoft appears to be very good at what the Japanese call kaizen. This is, in a nutshell, gradual or continuous improvement of an existing thing. All of these examples are of that type. The really big innovation that creates a market or produces the Next Big Thing(tm) it seems they have always ever left for other people to do, and then they try to make gradual improvements: most of the time these improvements are nothing other than getting part of it to run less than spectacularly on their own platform. They'll occasionally get useful improvements out the door too (true kaizen), but these are generally rare.
Slashdot Mirror
"Growth for growth's sake is the ideology of the cancer cell." I believe this quote is attributed to the environmentalist Edward Abbey.
Matter on the way into a black hole gets compressed by the immense gravity to tremendous pressures, forming an accretion disk, which radiates mostly x-rays and similar forms of high-energy radiation. It isn't really the black holes that are radiating, but the matter falling in that does. Tossing matter into a black hole is probably a far more efficient way of converting matter into energy than even nuclear fusion.
Hawking radiation probably isn't what's going on here. The temperature of a typical stellar mass black hole is about 10^-8 kelvin, way colder than the cosmic background radiation at 2.7 K, so it actually gains more energy from the background than it loses from Hawking radiation. Larger black holes are colder still. It's the really really small black holes (such as the hypothesized primordial black holes might have formed by small irregularities in the early universe that collapsed in on themselves; typical one squeezes the mass of a large mountain into a space a quarter the size of a proton) which are supposed to radiate significant quantities of Hawking radiation.
Four words: monkey in the middle. If you automatically receive public keys from public key databases, how would you know for sure that the key actually belonged to the recipient you intended? Maybe it's a key for someone pretending to be your intended recipient to you, and you could wind up communicating with your intended recipient through that monkey in the middle, who is pretending to be your contact to you, and pretending to be you to your contact, all the while reading everything you guys say and perhaps, subtly changing the content of your communication. Without some form of authentication of public keys, accomplished either by the web of trust model used by PGP/GPG (which you don't seem to fully understand, by presenting such a scheme, which would break the web of trust if used as you intend), or a centralized public key infrastructure based on root certification authorities, these sorts of attacks are possible. If you cannot have some reasonable idea of who it is you're really talking to, why should you bother encrypting your messages? You could be talking to the NSA anyway, and they wouldn't even need to break any cryptography, just pretend to be someone you know to you. In almost all real-world situations, authentication is almost always more important than encryption.
And may Dimebag rest in peace.
I hope that remains true. The sobering reality is that these two cherished boxes are gradually diminishing in their effectiveness with what's going on these days. These warrantless wiretaps and other infringements of Constitutional guarantees, unreliable electronic voting machines, and so forth are all conspiring to weaken the effectiveness of the soap and ballot boxes in your great country... Soon enough, only the ammo box will help, and God help us all if it ever comes to that.
Right, now that I've looked it up, but the issues are still the same it would seem. Burning these types of boards to get at the metal is still something that you really don't want to do. They typically use brominated flame retardants on these boards, which while they are generally inert and non-poisonous in their normal state, when burned they also produce deadly carcinogenic fumes, especially in the presence of copper. That makes that ingenious technique described in the article all the more useful: it separates the metallic and non-metallic components without burning.
The plastic used for most PCB's is polyvinyl chloride, which in itself isn't particularly toxic under most circumstances. It's in common use these days for plumbing fixtures and the like, as well as for PCB's. However, if you burn PVC plastic, it gets converted into some particularly nasty dioxins and furans which are dangerously carcinogenic.
There are not enough police, there are not enough soldiers, not enough courts, not enough prisons in the United States--or in any country in the world for that matter--to enforce a law that the public doesn't agree with. The art of governing is largely the art of convincing the governed that they are better off following the laws than not, especially if it is a law that touches almost everyone as these changes to copyright do. Thus far, these stalwarts have only succeeded in convincing the public only in a limited sense, and these heavy handed tactics that the RIAA and its ilk are using are, in a way, having the opposite effect.
The power of government really always is with the people, whether the form of government recognizes it or not.
Well, if it really is router firepower that's lacking, then perhaps it's way past time to seriously make the painful switch to IPv6. Contrary to many peoples' impression IPv6 isn't just about a larger IP address space, but also simpler routing. The core routers running BGP4 these days have about 200,000 entries, and that number is again growing exponentially. These kinds of monster routing tables are unnecessary with IPv6.
Public key encryption is, in practice, used pretty much the same way as well. Public key algorithms are generally used as part of a secure key exchange protocol rather than encrypting a message as directly.
Good point, but I think that part of the problem is that wireless bandwidth is a lot, lot less than land-based bandwidth. With the best WiMAX technology you can get at most 70 mbps: compare that with even an OC-3 fiber which is more than double that at 155 mbps, and if you lay an OC-192 you can get more than a hundred times the bandwidth of WiMAX.
Right on. The DRM problem on a general-purpose computer is, from a security standpoint, completely impossible. If I have absolute control over my entire computer, and this is still possible today because systems like TCPA haven't been forced down everyone's throats, then any attempt by anyone to restrict what bits I can and cannot copy is doomed to failure. And once I have done it, I can publish my break to the world if I so desire. These people might as well go on rolling a huge boulder up a hill, only to have its weight defeat them just as they are approaching the top. The task of DRM is equally futile. The only way that has a snowball's chance in hell of success would be to ban the general-purpose computer entirely (e.g. a TCPA-based media console), and it is doubtful that they will ever succeed in making such a thing happen, especially as the ban would have to be international in scope and there are some places (e.g. China) that don't care a whit about these matters.
You sound a lot like Ptolemy I (Soter) when he spoke to Euclid and asked if there was some kind of short cut to an understanding of geometry. Euclid's famous reply was: "There is no royal road to geometry." Computer science doesn't just have a lot to do with math: it IS math at its most fundamental. If you recall where the very idea for the automatic computer came from, it grew out of the quest for resolving the most profound issues at the foundations of mathematics (google for the 'Entscheidungsproblem' and Alan Turing). Just as there are no royal roads to geometry, there are no royal roads to computer science either.
If too many people in your educational system think that math is too hard, then maybe your educational system, especially at the lower levels, ought to be rethinking itself if it has as one of its priorities the training of more people who can really make use of computers and computer technology properly. I can't really say this with smug satisfaction as the case is the same here where I come from, but frankly the last place to go about fixing the problem is to water down the mathematical aspects of computer science as you're suggesting might be good. To do that would be to water down the essence of what computer science truly is.
Very, very clever. If I had mod points I'd give them! If Microsoft is really serious about doing this, then they will be doing the very antithesis of what they have been doing since, well, ever. Proprietary file formats anyone? Secret protocols? DRM? All of these things which they've been doing and promoting from the very beginning are precisely the sorts of things that will frustrate future digital archaeologists to no end. Consider the simple fact that we can still read Galileo's technical writings from the 1560's, but not Marvin Minsky's technical writings from the 1960's, thanks to proprietary storage hardware. Stuff is basically written on the wind these days, and Microsoft has done more than any single organization (largely because of their market monopoly) to make information as evanescent as it is now.
The term "plaintext" has no meaning for cryptographic hashes. The point of the attack is, if you had some text that hashed to say, f6540dee6b248c863bb90fcaa784fef9, I could, with even less work than I initially gathered, generate another set of data, saying something completely different, which also hashes to the same value. This has severe implications in the field of digital signatures, as the way most digital signature algorithms work is as follows:
Now, if I wanted to verify the signed document:
See the central role that hash functions play in this scheme? Now, if I'm able to generate collisions for the hash with a feasible amount of work, then given any digital signature you make using MD5 as the base hash I can create another document that might say something quite different, and attach the digital signature of the initial document to it, and it will look to anyone who cares as though you signed it. Of course, the document might look slightly strange to a person looking at it, but if the signature were used as part of a more complex cryptographic protocol, where only a computer ever really sees it, or if it's a document in a complicated file format (e.g. PostScript) that provides some leeway to add arbitrary data that is ignored by the viewer, then we might be in a spot of serious trouble. These collisions for MD5 are not just harmless, theoretical curiosities as you seem to think, as the page on X.509 certs and the PostScript examples illustrate.
The fact that collisions can be computed also destroys the non-repudiation characteristic of digital signatures. If this is important for your application, continued use of a weak hash function like MD5 is certainly out of the question, as the ability to generate collisions gives the alleged sender of a signed document a valid excuse to plausibly deny that she ever sent it. This of course makes MD5 digital signatures worthless from a legal standpoint.
Fortunately most people and began phasing out the use of MD5 for these and similar applications over the past several years, as it was already long suspected to harbor significant weaknesses. Now, we're hearing similar news about SHA-1 as what we heard about MD5 since roughly 1996, when cryptographers began recommending that alternative algorithms be used. Now, we're hearing the same thing again about SHA-1, and I think it would be wise to heed the advice again and start migrating to better hash functions as soon as possible.
How you got modded insightful is beyond me. This shit is real, very real, not just some propaganda from the Chinese. The attack on MD5 has been demonstrated by generating a couple of forged X.509 certificates based on the MD5 hash. It has long been suspected that MD5 harbored significant weaknesses, but it was confirmed in 2005 when Wang and her team demonstrated in a 2005 paper (warning PDF link) that it is possible to generate MD5 collisions with only about 2^39 hash computations (approx. 500 billion), a level of computational work which is doable in a matter of a few days even on the computer which I am using to type this post, and a very long way from the 2^64 computations required by a brute force attack. MD5 is well and truly broken, and not just in the academic sense, and anyone who says that the break doesn't affect the vast majority of its uses is either hopelessly uninformed or willfully ignorant. Checksums and digital signatures based on MD5 are now all suspect, and the only major application of the algorithm that remains unaffected is its use as a message authentication code, and the fact that the algorithm shown significant weakness in so many other areas should make anyone think twice before using it even for that. The biggest names in cryptography have been watching her work and that of her team with the keenest of interest, and there was an announcement (also here) that SHA-1 collisions could be found in 2^63 operations, which, while not feasible on my humble little PC, is within the realm of feasibility of today's fastest supercomputers and distributed computation clusters. Meaning that the NSA could probably generate SHA-1 collisions if they wanted to. Her most recent peer-reviewed paper on the subject gave a work factor of 2^69 for generating collisions, which while quite high, is quite a ways from the 2^80 required by true brute force, and that would make any serious cryptographer worried about using the algorithm.
Funny, but CERN itself makes that same misspelling of 'hadron' here. "This is the underground tunnel of the Large Hardon (sic) Collider (LHC)..."
This reminds me of an old Dr. Dobb's Journal article that I read more than a decade ago entitled "Personal Supercomputing" (I believe it was back in 1992 or thereabouts) where the author found a good use for a 486+i860 (remember that chip?) combo that involved making the i860 a computation engine and the 486 sort of like an I/O processor, and IIRC it was called PORT. The compiler set for this system didn't generate native i860 or x86 code, but instead compiled C or FORTRAN programs into a type of fixed-length instruction set tailored to the source language. The i860 would interpret this instruction set using a very efficiently hand-optimized interpreter that could fit almost entirely within the on-chip cache, reasoning that the frequent cache misses that come from executing RISC code directly are much more expensive than the interpretation overhead, and it seems that this observation was correct, at least in that case. Essentially, instead of using the i860 as a native RISC processor, the author used it as what could be considered a CISC processor with what amounted to programmable microcode inside its on-chip cache! The author even went so far as to say that this is the way RISC processors should really be used.
I wonder why no one has tried to use this same approach with more modern RISC architectures. I can see that the approach doesn't lend itself well to multitasking (the article was concerned with building supercomputers, to which multitasking is the very antithesis), but it is similar in principle to how VM's such as Java's and the .NET CLR work. It should also be noted that the instruction sets that the compilers that the PORT system used were memory transfer instructions designed in such a way that most individual statements in C or FORTRAN would compile to at most one instruction, rather than stack-based instruction sets like those used by Java and .NET.
But unfortunately, not all of the world has access to such wonderful libraries, and specialized research is somewhat difficult, even if your city is one that is blessed with a nice public library. Boy, I loved it when I discovered sites like this, and this, and this, collections to truly warm the heart of a math geek like me. Good luck finding even a tenth of the books and journals in those three collections in your local public library.
And that guy in the White House actually says he believes in this. How ironic that he should be the one to make it come to pass:
Very well said. Further extensions of copyright, with the way copyright works these days, is tantamount to stealing music from future generations. If the purpose of copyright is to encourage artists to continue creating more works, then the last thing we should do is allow them to rest on their laurels! The length of the copyright term is actually a disincentive to artists to make new works. It also hinders artists in the future from building on top of those works.
By the way, unbreakable DRM really is impossible: it's not even possible to make the problem mathematically infeasible. There is always a way around it provided you have control of all the hardware and software used to play back the protected media. As Bruce Schneier famously put it, making bits on a general-purpose computer uncopyable is like making water not wet. The only way to make DRM work absolutely would be to make true general-purpose computers illegal, or make it illegal to interface protected media to a general-purpose computer. As long as there is no law preventing people from circumventing DRM, someone will do it.
Have you ever tried contributing non-trivial portions of code to something under the GNU project or otherwise copyright the FSF? If you have, then you would know that the FSF is strict about copyright assignment, unlike Linus Torvalds, hence I don't believe that this will ever be a problem for any project of theirs. If some key developer were to break away, the code they try to start the fork from would not belong to them, but to the FSF. The forkers could maintain the fork's GPLv2 status, but that would not allow them to change the licensing conditions and eliminate the 'later version' clause: only the FSF would have the authority to do that.
The entire GNU toolchain will move to GPLv3, as well as all of the other software owned by the Free Software Foundation. That, you can definitely count on. Other software that depends on this vast base of software, much of which is fundamental to the operation of any GNU/Linux system, will probably be encouraged to switch with time. Uninformed projects probably are also uninformed enough to use the FSF's original statement of the GPLv2, which contains the 'later version' clause that the Linux kernel lacks, and thus GPLv3 terms may also apply to them if that is desired.
Microsoft appears to be very good at what the Japanese call kaizen. This is, in a nutshell, gradual or continuous improvement of an existing thing. All of these examples are of that type. The really big innovation that creates a market or produces the Next Big Thing(tm) it seems they have always ever left for other people to do, and then they try to make gradual improvements: most of the time these improvements are nothing other than getting part of it to run less than spectacularly on their own platform. They'll occasionally get useful improvements out the door too (true kaizen), but these are generally rare.
And penned by no less than the great Greek legislator himself!