First, once upon a time, there were no real alternatives to RSA. These days, I know of at least five other major categories of public-key encryption and there are undoubtedly more. This makes the idea of dropping RSA - at least in the really sensitive stuff where any known vulnerability is really bad news - a definite possibility.
Second, these are timing-based attacks that perform branch prediction. This requires no changes to OpenSSL or any other source to completely mask. You first mix the optimization methods when compiling, and then use the real-time scheduler, to ensure that as many branches as possible (preferably all) take the same length of time within the margin of error of any observation. You can't choose between that which you cannot distinguish.
Finally, this is specific to an implementation. If OpenSSL had multiple branches for the same algorithm (just a different implementation which it could select at random for any block of data), or if you had multiple OpenSSL installations which differed sufficiently in implementation that the computer could again select at random, you'd make it a lot harder for an attacker.
There was an article on Slashdot a while back on a new crater discovered in Antarctica. It was a couple hundred miles across and was believed associated in some way to the Great Extinction. Well, there's a neat website that lets you calculate the size of crater and damage done for a given size of asteroid. It took a while to find one that would produce the crater observed that would have a combination of speed and size that would leave anything left alive at all, let alone 5%-10% of the biomass.
My crude reverse-engineering of the asteroid suggests that it would have to have been moving very very slowly compared to the Earth, and be about 50 miles across. Even so, the calculator predicted that anything within the horizon of such an impact would be instantly vaporized and that the entire hemisphere would be subject to earth tremors of magnitude 11.2 or above. That was about the smallest-scale devastation I could find that would produce the right-sized crater.
(Faster asteroids would be smaller, for the same-sized crater, but end up releasing much more energy, as energy goes up with the square of the velocity.)
Now, turning an asteroid (or comet) is plausible, but it has to be done early. You say you can only achieve a meter or so, but in reality that doesn't mean anything. You change the trajectory, and the change of displacement is then the distance the asteroid travels divided by the tangent of the angle between the original path and the new path. (The tangent is equal to the opposite over the adjacent - SOH CAH TOA. You make the adjacent the line it would originally have followed and the opposite becomes the displacement.) Objects travelling along a curved trajectory need to be mapped into a linear system first, which is usually a very simple transform.
So how does this help? Well, since you are changing an angle, the implication is that if you increase the distance away you make this change, you will increase the displacement from the original position. If the change in displacement exceeds the Earth's radius plus the safety margin needed to prevent the Earth's gravity from causing the collision to occur anyway, then it makes bugger all difference if you can make one degree of change or one billionth of a second of a degree. All that matters is that the cumulative change places the body outside the danger zone.
What does this mean in practice? In practice, it means that if it's just about to collide, there is nothing you can do to stop it and there are few structures in the world capable of withstanding 11.2 magnitude tremors. Evacuating the hemisphere and placing everyone on a geologically-sound plateau would be far cheaper and would have a much better chance of success. Near-zero, as opposed to absolutely zero.
If the body is unlikely to collide for a couple of orbits and a few hundred years, then you can talk about serious landscaping the solar system. That's the kind of distance where even a small angle will make a large difference. Better yet, gravity is vastly more powerful than any explosion - if you can shift the orbit just enough to place the body close to a large planet, the total deflection will vastly exceed anything explosives can achieve. Gravity is a significant force on these scales.
This all assumes that the body is solid, of course. The Japanese robot probe that landed on an asteroid not too long ago found a nearby asteroid whose density was unimaginably low - it is likely to be nothing more than space grit held together with collective gravitational attraction where the packing is no better - and probably worse - than coarse-grain sand. It could be said that its structure is best described as sheer damn luck. You fire off a nuke on something like that and there's no telling what will happen, other than most of the energy will go straight through it. At this point, we simply don't have anything like a large enough catalog of asteroids, nor in anything like sufficient detail, to know if this is a freak accident or the norm. Until you know enough of the basics, you can't know anything about the complexities.
It makes no difference if Novell adds Microsoft code or not, in that since we can't see Microsoft's code, nobody can know and therefore Novell's submissions MUST be rejected.
This is further supported by AT&T's argument that BSD was contaminated through exposure to AT&T's intellectual property. That lawsuit was settled and not won, if I recall correctly. Novell's employees are now subject to exposure to information Microsoft provably regards as intellectual property - otherwise there'd be no point in a sustained relationship. If code can be contaminated because of exposure to someone else's code, even if that code is never actually used in the final product, then anything Novell releases may be contaminated.
(If the "contamination" argument sounds absurd, then logically I'd have to agree. However, this is the main reason that Open Source developers are careful to "clean room" such stuff, where the person who sees the original code or reverse-engineers the original specs does NOT write the code but ONLY documents the inputs, outputs and relationships. The code-writer is then much safer from claims of contamination, as they are merely writing code to an abstract description of what the code does, they are not writing it based on anything they themselves have seen in other code elsewhere.)
Is there infringing software? Well, that depends on whether anyone was careless. If Mono, SMB, CIFS, Samba, Samba-NG, NTFS, VFAT or any other potentially patentable concepts were developed without rigorous clean-rooming, then there is a risk of contamination. A risk does not mean contamination actually occurred, and contamination is not a proven legal notion as far as I know, as the AT&T fiasco is not much of a legal precedent to work from.
That's for copyright, though, this is patent law. There, if there's an infringement, it need only be within the confines of the patent, it need not be the result of copying or contamination. A completely independent development would violate a patent, as a patent is on a process, not an instance. I believe this is true even if the independent development was parallel to - or even preceded - the patented work. Prior art is not automatically safe - many of the victims of Thomas Eddison were prior inventors and I believe case law (any prior ruling, even if insane, stupid or blatantly absurd) dictates what a judge can do.
This is actually one very major weakness of ANY Open Source license - because prior art isn't safe, corrupt corporations can take an Open Source product, patent the ideas within it and sue the original author. In theory, this shouldn't be possible. In practice, there are good reasons for believing it is not only possible but a very reasonable possibility.
(My counter to that is that exploitable bugs in the law should be fixed and that workarounds are little more than exploits of bugs or weaknesses within the exploits and are therefore at best a temporary reprieve. I'd quite like to see changes that ensured that if prior art was sued for patent infringement that the patent and rights were transferred to the defendant on the grounds that they're more likely the real inventor. I'd also like to see software patents filed in a radiation-proof concrete-lined, lead-lined corrosion-resistant, nuke-proof steel drum before being dropped into a bore-hole that reached the inner limits of the Earth's crust - that being a far safer place for them than in the hands of lawyers. Well, unless the software patent lawyers were also filed in said drum, in which case it wouldn't matter what their hands held.)
...that particular clause is related to the fact that there is no nominal transaction taking place to convey the authority to use from the licensing person or body to the licensee. IANAL, but I believe this is why you get a lot of really strange transactions for a dollar or for a pound, because there needs to be some nominal token of exchange to be a legal exchange. Now, in this case, the license IS a token of exchange, because it can be transferred - and in fact IS transferred whenever someone obtains a copy of the code in source or binary form.
Because you have a transferable token, then it would appear - to my untrained eye - to be a transaction that has involved the payment of that token.
Now, as I understand it, the reason general "free distribution" cases wouldn't be covered is that there is no such token payment and therefore no instrument by which copyright can be carried. This would explain why the general case is NOT covered, but would also mean that this case unquestionably is.
There is another aspect to this. A "Gentlemen's Agreement" requires an agreement between two individuals to be witnessed by an independent third individual. If there is a Gentlemen's Agreement to honour the license as a statement of copyright, then I believe there would be a case for claiming that copyright would apply whether or not it would normally have done so.
Ok, so we have the two parties. Is there an independent witness to an agreement? That one might be interesting to argue. I'd claim that Richard Stallman knows neither party but DOES know what the GPL means and can therefore be an independent witness to the effect that by accepting the GPL, the licensee has accepted the Gentlemen's Agreement. Not sure how well that would stick, though. My suspicion, though, is that if the judge reads and approves of the GPL as a legal instrument, but is 50-50 as to which way to go, the Gentlemen's Agreement argument might be enough to swing it, if such agreements are recognized in that State.
The last possibility is that if the licensee claims that they're not bound by copyright, and since copyright is the sole instrument to allow the licensee to use ANY GPL product whatsoever, Stallman may just declare the licensee Accursed and have them prohibited from using or obtaining ANY GPL product whatsoever, for all of eternity. (Under the license, that IS the standard penalty for blaspheming the GPL in this way.) This would not impact the current lawsuit as much as it opens the licensee up to a lawsuit for EVERY GPL product they use, distribute or possess. Quite simply, the costs involved in defending a massive distributed denial-of-sourcecode attack could cause some serious damage, and because it is an explicitly stated curse, not a single one of those lawsuits would likely be thrown as frivolous.
What you're not adding is that the monster rolled a 20 8 times in a row. On average it was just fine...
(Actually, I've seen weird rolls. When GMing a Rolemaster game, I had the unfortunate luck to see a Black Reaver get slaughtered by a first level character... Ok, ok, there are those who could argue that greater undead/greater demon combos shouldn't be put against first level characters anyway, but they didn't HAVE to pick up the bright, sparkly, foot diameter gemstone...)
It's interesting to note which country the researchers came from... Being an axial dipole means exactly nothing, unless you encounter a horseshoe-shaped planet. The magnetic axis wobbles all over the place and even reverses, but I've never heard even the remotest suggestion that it has ever been anything other than a very simple axis. The Earth's core generates a magnetic field as a result of (a) being molten, and (b) rotating, so is probably produced by circulatory currents within that core. The only possible arrangement for those currents is to be in distinct hemispheres - they're simply not going to overlap.
The upshot of all this is that the researchers hopefully(?) have better evidence of their claims than a few buzzwords which don't really amount to a whole lot. 600 million years is not a long time, geologically speaking - or even evolutionarily-speaking - and I'm not convinced that every necessary process to get from Iceworld to habitable planet could occur in such a short space of time. I could be wrong, but I would need some VERY hard evidence.
What sort of evidence? Well, certainly volcanoes existed back then, and the eventual form lava takes depends on how rapidly it cools. Find me a lava bed in a land mass that would have been tropical at that time (based on current theories of plate movement) and which would not have been connected to open water, but where the rock structure shows clear evidence of cooling of the kind geologists associate with plunging molten lava into ice, AND where that rock also shows clear evidence of prolonged frost cracking. This would not be "solid proof", but it should be very persuasive.
Any other possibilities? Sure. Ice sheets and glaciers form very distinct U-shape valleys. Very very distinct from the V-shape of river valleys. If, indeed, no rivers existed 600 million years ago, only ice sheets, then it follows that any valley of that age or older MUST show the characteristics associated with such ice. Sure, a river may have cut through after, but that can't affect the sides of the valley above it. It can only affect the ground that it was cutting through at the time. The ice sheets, however, will have NOT been level with the ground but will have risen above it. Thus, the ice WOULD have reshaped any river valleys.
Thus, if a single valley, anywhere, shows clear and distinct evidence of cutting by water prior to 600 million years ago, OR if any valley of that time-frame does NOT show evidence of cutting by ice, the theory is falsified. Either will do.
If their claims stand up to the igneous rock test AND the valley test, then I'd be inclined to take the theory more seriously. As it stands, I simply don't see that they've enough to base their claims on. You know when academia is going south when Art Bell's old show could claim greater evidence and a more rigorous analysis. Hey, I happen to think the show has a lot of merit, both in examining controversial and unorthodox thinking that would not otherwise get the hearing it deserves and also as excellent entertainment, but we're in DEEP trouble when the fringiest of the fringe sound more careful, more exact and more scientific than highly-skilled, highly-trained, highly-paid experts.
I dunno. I could very easily imagine the richer suspects paying the cops $4,000 to brain-scan the poorer ones, especially if the rich suspects are innocent. (Time is money, reputation is even more money, and at the current rates, this might easily be a sound investment to some people.)
Heh! No offence taken. You added to what I know - well, I knew the scientific method already, but I did not know that people could train their brain's activity levels to the point of decieving fMRI. Yes, I knew people could train themselves to set up certain brainwaves via EEG, but fMRI is a different ballgame - and not many have one in their homes to do personal training with.:)
Ultimately, your post was exactly what I would call the perfect response - calling me on those things that were either not right or not clear, adding the necessary information (emphasised because this is such a vital point), and doing so in a way that is perfectly friendly. Good communication skills are rare in any technical field - you'd have to look in the top 1% of IT to find any who are in the top 50% of even the average population - so it's always good to see people who are good communicators. Besides which, the only way I can be right all the time next week is to learn what the errors are in my thinking this week!
For the polygraph tests, I was working on the basis that a polygraph test is a known quantity. The confidence limits on accuracy, repeatability and process are well-established and well-defined. They also already have the equiptment, experience in using it, etc. fMRI lie detection is a relatively unknown quantity, the confidence limits are - ergo - much broader. Starting from the statement "at most it admits the polygraph is less accurate", it is not obvious how to compare the two. At what point does something become less accurate, when you're dealing with distributions and not point values?
There are many ways to do this, but here's two. The easiest approach, if you have enough data, is to find the ranges that give you the same confidence limits - preferably 1% or 0.5%, as anything above that is absurd. You pick the worst end of the range for the fMRI and the best end of the range for polygraph tests. In order to absolutely, firmly, definitively say that fMRI actually is better than polygraph, the polygraph's upper limit has to be strictly below the lower limit for fMRI - otherwise, there is a non-zero possibility that polygraph tests are superior to fMRI. (Not just in some specific case, but in general.) Because courts presume innocence and because lies are more harmful than truth is helpful in investigative work, it's better to assume the least accuracy than the greatest accuracy. That means we now pick the lower limit for polygraphs as being the effective merit of the results. Given the absurdly wide margins there must be for fMRI, if we assume polygraph's limits are below that, the lower limit for polygraph has to be very close to zero.
(If the limits overlap, even to the smallest degree, then replacing all polygraph systems with fMRI would be stupid, as it would be impossible to be sure if that was making things better or worse, at vast cost. Nobody, not even some of the Government agencies I've dealt with in the past, could possibly be so stupid. So, if they made such a move, despite the gigantic spread of bounds for fMRI and assuming the choice was rational, it would seem that we can deduce the worst-case for polygraph would mathematically need to be very close to zero.)
The second method, which assumes some data but nowhere near as much, is to take the average expected reliability of the test and place the lower bound at two standard deviations below that. We don't even need to know the values for fMRI for this, we just need to know them for polygraph tests. Two standard deviations would be enough to cover the majority of meaningful cases and lop off all of the tail-end extreme or unusual situations. My guess is that the lower bound for polygraph tests, using this rather crude method, would again be perilously close to zero. This is easier with very little data (you need the mean and either the variance or the standard deviation), but it's harder when working with raw data because at no point do you know if you've calculate
...there's a lot of interest in work being done to use fMRI for lie detection. There are specific areas of the brain that light up when you lie, even if you aren't conciously aware that it is a lie, from what I understand. However, nobody has the foggiest what the accuracy level is (it's too new of an approach), fMRI is vastly more expensive than a polygraph system, only those who did the one study are even remotely qualified to conduct such a test, the psychological aspect is completely unknown (as opposed to a polygraph test where it's all there is), switching to it would essentially admit that every polygraph test they've ever performed is 100% bogus, and it simply doesn't have nearly the same BDSM coolness to it as a chair with straps.
There are a hundred HTTP-derived protocols - SHTTP is not the same as HTTPS, for example.
There are a thousand markup languages, covering everything from geometric data to maths equations to typesetting to fonts. (BLOBs seem to defeat the whole purpose of an ASCII-driven markup system, IMHO.)
There are a million capabilities that couldn't be done as extensions (multicast mosaic was a definite curiosity and only touched the outer fringes of what's possible).
There are a billion ways that any browser could be tweaked to tune it dynamically for not just the system but also the network weather of the time.
The sheer volume of possibilities would overwhelm anyone, and many browsers have met with unfortunate ends as a result. (I can't even remember the name of Sun's pre-Java web browser.) However, those that don't venture out have all eventually stagnated and died.
My question, then, is how do you choose? By what black magic do you discern what will kill and what will cure?
It would be much more interesting to know what would happen if you modified what was provided to the bee/wasp larvae on hatching. Presumably, there would be some change, as diet certainly alters how the brain develops in other animals. However, with such tiny brains and with much of it hardwired, it is unclear what's left to change. However, it can't be all hardwired, as bee dances can be "learned", so some dynamic structures must exist. How dynamic, though?
It also makes me wonder about african bees. These are much more wasp-like in their passion for mindless violence, and are far inferior to regular honey bees when it comes to processing pollen, so must presumably be midway on the evolutionary scale.
Then, one must also consider that there are MANY species and sub-species that fall into the category of honey bee. Since the genetic code is relatively short, it should be moderately easy to understand how they interrelate. I would assume it would also be possible to genetically engineer honey bees that have far greater efficiency. (The "killer bees" are the results of a catastrophic experiment to mix african bees and American honey bees, together with allowing them to escape into the wild. All-round, excellent proof that gross stupidity can be bred into researchers. What I'm thinking of is having a totally sealed environment, similar to biosphere 2, and a program that's based more on science than air warfare grants.)
...of what the bad guys know is to tell them and mark it off on the list. Anything else is down to chance.
The chance of them knowing is the probability of them finding the information multiplied by the probability of knowing the value multiplied by the probability of producing a workable exploit.
The chance of you knowing if they know is the probability of them knowing multiplied by the probability of you knowing who the bad guys even are, multiplied by the probability of obtaining real information (they can jam anyone monitoring them by flooding the information space with junk information), multiplied by the probability of you knowing you even have real information, multiplied by the probability of being able to determine what the information actually means.
Counterintelligence is an exceptionally difficult field with a painfully poor track record. Most published successes have been by a series of sheer fluke events and staggering luck. Most published failures were unlikely to be anything else. We don't know about the unpublished stuff, but percentagewise, are we more likely to see bragging over achievements or failures, if both can be equally hidden?
I'm not saying that everything should be published, merely that it should not be assumed that not publishing is the same as others not knowing.
Now, can a case ever be made for publishing everything? Yes. Game Theory requires that all "full information scenarios" have a strategy for one side and one side only that will ALWAYS result in the winning conditions being met, no matter what the other side does. It is possible to imagine situations, particularly in computing where there is essentially no randomness and a "full information scenario" is possible, where the outcome can be guaranteed, if you want it to be.
No matter what anybody else might say, it is not the job of an enemy to make your life easy, so we shouldn't expect them to. We should expect them to do the researcxh, the legwork, the analysis to figure everything out. They might indeed just wait until someone tells them, but that should be a bonus. It should not be your modus operandi. In computer security, you must assume that there are opponents out there who could have all of the industry-standard backdoor passwords, a complete printout of every Operating System and network device QA test that failed and got overlooked, and a copy of the highest-end vulnerability scanner the commercial sector has going for it.
Hell, we know that a Russian spammer got a tier-1 backbone provider to turn off Blue Frog's Internet connectivity. Turning off a link like that is very traceable, but appears to have been regarded as mere amusement for the backbone provider. The same provider is hardly likely to show scruples when it comes to handing out internal or commercially-sensitive data, software or anything else. Given the repeatedly low scores on security for many US government departments and the almost routine mishandling of classified data, there are probably those in the information black markets who know more national secrets than the entire White House combined. If one backbone provider is riddled with corruption and pwned by organized crime, then we must assume that such people are unlikely to be avoiding big money out of a sense of decency and moral fortitude.
But if the most dangerous people have the most dangerous information already - and that includes whatever terrorists might actually exist - then most of the obscurity only serves to increase the value of what has already been stolen. This makes the thieves rich, the criminals dangerous, and the politicians popular for appearing to do something, but it doesn't make anyone else - users, vendors, bystanders - any better off at all. Illusions are fun on the stage, but they should be left there.
It is unclear what an IPv4 compatibility address would mean, in an IPv6 browser. Do you push it to the IPv6 stack and let it handle it, or convert it to an IPv4 address when such an address is available (on the basis that this is likely what the user wants)?
Also, if you have a name that can resolve to both A records (IPv4) and AAAA records (IPv6), there is a big problem of prioritization. This is important, as there is no guarantee that the two sites would be the same and the way that URLs are currently handled does not make it easy to choose one over the other.
Transitory addresses can also pose a problem. 99% of the work is handled by the OS, but any cookie that uses an IP address must be updated to the new address if it is to remain useful. The same is true of IPv4, but IPv4 users haven't used mobile IP much and generally use a fail-over or load-balancing system that presents a single address to the outside world.
These are all corner cases, we're just going to see a lot more corner cases with IPv6 than we did with IPv4, and I'm not convinced any implementation is taking that into account.
...Napoleon's troops apparently did shoot the nose and some of the face off the Sphinx with a cannon. It is not completely clear if there was further damage done (either directly or as a consequence of sending shock waves through sandstone). This will likely have shortened the lifespan of the monument and has contributed to the modern, sometimes bitter, feuds over the nature and origins of the Sphinx.
The destruction of the two standing Buddhas was a terrible disaster for archaeologists and historians. Partly because, as we know from the vicious infighting in these disciplines over whether the Seven Wonders ever existed, there is a tendancy to ignore documented evidence if there's no visible evidence. If the Taliban find the Sleeping Buddha before the archaeologists do, they may be able to eliminate any proof it ever existed as we only have the vaguest of hints of a third giant statue near the location of the first two. We have no idea where it is or what other archaeological finds might be located around it, but being buried means that it is possible it is in far better condition - maybe with the original paints intact - than the two standing statues were.
Importance level: High. Enough of the paints survived on the two standing statues to make educated guesses as to the composition, but it's hard to be certain when you're dealing with microscopic flakes that have been exposed to any number of chemicals in the air and rain (or in the rock itself), where you can't be sure how many layers there were, or even how well the paint was mixed, and where photosensitive chemicals might have done almost anything (except tap-dance) in that sort of timeframe.
(Ok, so the importance is only high for people with an interest in Afghan history, Buddhist history, Silk Road history, monumental carvings, archaeology in general, ancient technology, or who would quite fancy being able to only paint the outside of their house once every five hundred years and still have it look good.)
It is also important to consider that people are less careful with those things that they feel no ownership of, have no understanding of, and are devoid of any connection towards. That's to be expected and should surprise nobody. But, by implication, negligence or destructive tendancies towards historic artifacts is most likely when the individual or society are totally disconnected with history. Making the ownership explicit and universal can only help to empower communities.
I do not believe the world can continue with totally isolated nation-states. The European Union is probably the clearest example, as that is a collection of member nations who have voluntarily sacrificed some of their authority to an overarching power. France, England, and so on, do NOT have "sovereign and complete rights" over their geographic areas. They have extensive rights, but they are far from absolute. However, those they have assigned rights to do not have absolute power either. Rather, there is a muddled but deliberate separation of powers.
The International Court of Justice, the African Union, and even the United States, are other examples of where members have voluntarily sacrificed absolute authority in favour of having separation of powers and outside agencies to which certain responsibilities can be delegated.
I see no contradiction, then, in doing this for world heritage. It is merely an extension of an arrangement that has generally worked rather well. (It is only when powers cease to be seperate, and absolute authority forms, that serious problems arise. Total and irrevocable seperation of powers won't eliminate war, but it would likely eliminate a great many wars.)
Your example of oil is an interesting one, for many reasons. Britain's North Sea oil reserves were badly mismanaged, through absolutism. America did just as badly but has escaped the worst of the suffering for now. In many poor countries where oil is to be found, you see exploitation and environmental destruction on a horrendous scale, usually with the consent of the Government there. The same is true for gold mining, where miners are exposed to a range of utterly lethal chemicals (often in the drinking water) with the consent of those in authority. I can't be certain of it, as I've not been there to ask those living in such conditions, but I'd be willing to hazard a guess that if they could earn the same or more AND keep their arms, legs and brain intact, they'd not give a damn who "owned" these resources.
Absolute ownership of the land is indeed a part of the "Nation-State" concept, but that concept is slowly being torn down as worn-out and decayed, susceptable to corruption and inherently economically unstable. America was the first of the modern nations to look for alternatives. States have some powers, the Federal Government has others. The judges, legislators and civil servants are all totally distinct - well, in theory. That's a six-way split of authority with minimal overlap. It's by no means perfect and it could probably do with being split further, but the US - by outlasting all other superpowers - has shown that the Nation-State is inferior by design to a more collaborative system. It just has to figure out how to get a collaborative system that's any good.
Any stolen object that can be returned without damaging it further (sawing a stone column into three pieces isn't exactly quality care) AND which will be kept with an equal or greater standard of care should be returned otherwise unconditionally.
"But it belongs to (country)! Why should (some other country) keep it?" In the end, none of this "belongs" to a country. History cuurered everywhere at the same time. (Duh!) For the most part, the political boundaries that marked these countries no longer exist, the political entities have vanished into oblivion, no living direct descendents who could claim even a moral ownership are known to survive, so for the most part the only meaningful designation is "world heritage" (which I believe to not be used nearly enough and most definitely not recognized nearly enough).
So, if object X is being, or would very likely be, damaged by being in country Y, I believe country Y has lost all right to the ownership of object X. I don't like the fact that Britain has the Elgin Marbles, but I like even less the fact that they'd be destroyed by pollution if they were ever returned. The Greece of back then no longer exists, any more than the Egypt of the Pharaohes exists today. In some cases, there simply isn't a country in which an object is truly safe. In that case, you document every last facet like crazy and hope. (You can't move the Great Pyramid and you certainly can't hide it, though reducing pollution might cut down on the deterioration.)
But what makes something "world heritage"? The object itself? Usually no. Except in some rare cases, the object has no value in and of itself. For inorganic objects, it is the information the object posesses - from the chemical structure through to any symbols or writings on it, and the information associated with it - where it was made, when, how and why, where it was found, the nature of the site, other items found there and their respective characteristics and associations, and so on. These are the things that have any lasting meaning. Once you know the object - totally - you can always make another using exactly the same materials, tools and methods.
For organic objects, it's tougher. If a bone is damaged or destroyed, there is next to nothing you can do. And time is rarely kind to anything of organic nature. Tutankamun is in very bad shape now and the remains will probably not survive a whole lot longer. Part of that is due to Carter's team, but part is due to Egypt having very high levels of acidic pollution and acid rain. You can't expect much to survive under such brutal conditions.
The other problem with organics is that there's much less information you can obtain. With luck, you can extract mtDNA, maybe even use modelling to produce an impression of what the person looked like. Bodies found in peat bogs and ice fields give slightly more information, perhaps yielding clues of fashion, food and culture that artifacts alone can't. We learned a lot from "Pete Marsh" and the iron-age traveller murdered in the alps, but such finds are almost never in any kind of context, so there is very little you can do to connect them with what was happening at that time. "Pete Marsh" - Lindow Man - might date anywhere from prior to the Roman invasion to a hundred years after the Boudicca Rebellion, making it very hard to know what sort of context is involved.
Getting back to thieves vs. archaeologists - IMHO, it's not a binary thing. I would argue that the "absolute" thief is one who destroys information in search of money, even if that involves destroying the thing they're trying to find. (When archaeologists started paying money per fragment of Dead Sea Scroll recovered, some of the locals cut fragments up so that they could get more money.) I would argue that the "absolute" archaeologist obtains all information, even if that means never reaching the object. (We now have GPR scans of Edward the Confessor's tomb, but reaching it would destroy countless artifacts and could potentia
...it will have to end eventually. As with any system, it is limited to the ability to communicate. In this case, Intel can't communicate with their chip designers and AMD won't communicate with anyone who isn't stinking rich. (I've yet to get them to do so much as reply to an e-mail, answer the phone or even return a call.)
I remember hearing once about a prominant scout leader who was a homosexual. What was his name... oh, that's it! Lord Baden Powell, founder of the Scout and Boy Scout movements. (He was also a British spy, which would imply a certain disregard for the opinions of those whose information he stole.)
There have been many attempts to distribute databases, but there are enormous problems with the approaches typically used (point-to-point makes for very slow synchronization, nobody seems entirely sure what/where/how to parallelize, there are usually single points of failure, etc).
Another area that has been worked on has been to allow database developers to embed more and more complex SQL-based scriptlets and "helper" functions into the database. However, it is a truism that interpreters (even of bytecode) are painfully slow and offer nothing that a module plugin mechanism wouldn't have (provided you could install modules via SQL statements) and most developers would be better leaving data processing to the clients rather than the server anyway. (The closest I've seen was Informix' blade technology, and that was horribly unreliable and tedious.)
The bottom line is this: There have been a lot of database innovations over the years. Some - I'd say most - have been great ideas but the popular interpretation and implementation has been terrible. This is not the fault of any of the ideas, but I would take it as suggestive that programmers are not being as imaginative or creative as they could be.
My question? How can a database company:
Combine the artistry vital to any creative effort with the architectural precision needed to make the art something special?
Keep things recognizable enough to be usable, without repeating everyone else's mistakes?
Carbon fibre is good, and there's a chance that there will be cheap (as in aluminium-cheap) ways to extract titanium, which would massively reduce the costs of lightweight materials.
Size isn't necessarily given up. If you use materials that have half the mass per unit volume, you can have twice the volume. Car shapes tend to waste a lot of material (cuboids have a large surface area to unit volume) and a lot of internal space to aerodynamics (you still want aerodynamics, you just don't want to waste as much space). Frontal area (as far as the air is concerned) is not necessarily the same as the frontal area as far as the occupants are concerned - there was an experiment in building double-skin racing yaughts, where the outer skin was porus and the inner skin had grooves. The idea was to build a hull that broke up turbulance within the hull.
You also need to consider that the engine isn't terribly efficient. 80% of what goes in the air intake is nitrogen. Nitrogen "burns" to form various oxides, but consumes far more energy than it releases. It would be possible to electrostatically enrich the oxygen intake - you can place a charge on oxygen far more easily than on nitrogen, then use the charge to seperate the two) but it works out that because you've only a very short time to do this, it's extremely hard to get any kind of net gain at all. I don't know of any other method of de-nitrifying the air intake, but the more you can exclude, the more energy you will extract for the same amount of fuel and oxygen. (You might be able to get the same effect by extracting the energy faster or cooling the engine better, as endothermic reactions require there to be more than some critical level of energy. I've not really pursued that line of reasoning much, as cooling an engine massively is not a trivial thing to do. Electrostatic seperation requires a y-shaped plastic tube, two metal grids and some wires.)
Improving engine efficiency is one option, but not the only one. The storage of energy is another. Simply tune the engine to work optimally under a very narrow range of conditions (because then you get the best performance) and convert the energy extracted into a more usable form. Regenerative braking can also conserve some energy. It's significant but not vast. Of course, getting the energy through far more efficient means is only one part. The other part is converting the energy into momentum for the car. The process (involving transmission, gears, traction, air resistance, etc) is highly non-linear. It would be better to reduce the gear ratios, have more gears, and have the automatic transmission pick a gear based on all related conditions and not just engine revs (which we have now eliminated anyway).
Air resistance is by far the biggest killer, as cars are horribly shaped with gigantic vertical slopes at the back, a huge hole below the engine cavity and absolutely no sense of aesthetics in the plumbing underneath.
Affluent people don't become rich by throwing away money. Kids are expensive and earn you nothing.
Also, athletic bodies are often damaged or disfigured with massive hormone imbalances and other severe (and occasionally fatal) problems. Gymnasts, for example, do not mature correctly and often suffer from muscle and bone disorders. Body builders, weight-lifters, etc, can disfigure their hearts - I would not expect life-expectency to be nearly so high. Rugby players - well, I can see them evolving into a whole new species that has less to do with class and more to do with causing sheer terror when barreling down the playing field. Soccer players can suffer damage to hearing or their pre-frontal lobes, from a mixture of heading and smashing into the ground at high speed. It's usually not lethal, but if you look at the various team managers for the England squad, it's clearly harmful to thought processes.
I can't see how the Swedes would turn into troglodytes. Oh. That's the British. Ah. I always thought there was something odd with the people in Milton Keynes.
Slashdot Mirror
Yo geeky ladies around the world
Got a wired box to show you
To telnet the boys and girls
Shell your brother, your sister and your momma too
Windows is going down
And you know what just to do
Wave your RAM in the air like you don't care
Run Glide with your games as hackers stop and stare
DVD and DVD and DVD then boot
Come on Linux tell me what's the word
Word op! Everybody say
When you hear the system call your drive will be getting underway.
Second, these are timing-based attacks that perform branch prediction. This requires no changes to OpenSSL or any other source to completely mask. You first mix the optimization methods when compiling, and then use the real-time scheduler, to ensure that as many branches as possible (preferably all) take the same length of time within the margin of error of any observation. You can't choose between that which you cannot distinguish.
Finally, this is specific to an implementation. If OpenSSL had multiple branches for the same algorithm (just a different implementation which it could select at random for any block of data), or if you had multiple OpenSSL installations which differed sufficiently in implementation that the computer could again select at random, you'd make it a lot harder for an attacker.
My crude reverse-engineering of the asteroid suggests that it would have to have been moving very very slowly compared to the Earth, and be about 50 miles across. Even so, the calculator predicted that anything within the horizon of such an impact would be instantly vaporized and that the entire hemisphere would be subject to earth tremors of magnitude 11.2 or above. That was about the smallest-scale devastation I could find that would produce the right-sized crater.
(Faster asteroids would be smaller, for the same-sized crater, but end up releasing much more energy, as energy goes up with the square of the velocity.)
Now, turning an asteroid (or comet) is plausible, but it has to be done early. You say you can only achieve a meter or so, but in reality that doesn't mean anything. You change the trajectory, and the change of displacement is then the distance the asteroid travels divided by the tangent of the angle between the original path and the new path. (The tangent is equal to the opposite over the adjacent - SOH CAH TOA. You make the adjacent the line it would originally have followed and the opposite becomes the displacement.) Objects travelling along a curved trajectory need to be mapped into a linear system first, which is usually a very simple transform.
So how does this help? Well, since you are changing an angle, the implication is that if you increase the distance away you make this change, you will increase the displacement from the original position. If the change in displacement exceeds the Earth's radius plus the safety margin needed to prevent the Earth's gravity from causing the collision to occur anyway, then it makes bugger all difference if you can make one degree of change or one billionth of a second of a degree. All that matters is that the cumulative change places the body outside the danger zone.
What does this mean in practice? In practice, it means that if it's just about to collide, there is nothing you can do to stop it and there are few structures in the world capable of withstanding 11.2 magnitude tremors. Evacuating the hemisphere and placing everyone on a geologically-sound plateau would be far cheaper and would have a much better chance of success. Near-zero, as opposed to absolutely zero.
If the body is unlikely to collide for a couple of orbits and a few hundred years, then you can talk about serious landscaping the solar system. That's the kind of distance where even a small angle will make a large difference. Better yet, gravity is vastly more powerful than any explosion - if you can shift the orbit just enough to place the body close to a large planet, the total deflection will vastly exceed anything explosives can achieve. Gravity is a significant force on these scales.
This all assumes that the body is solid, of course. The Japanese robot probe that landed on an asteroid not too long ago found a nearby asteroid whose density was unimaginably low - it is likely to be nothing more than space grit held together with collective gravitational attraction where the packing is no better - and probably worse - than coarse-grain sand. It could be said that its structure is best described as sheer damn luck. You fire off a nuke on something like that and there's no telling what will happen, other than most of the energy will go straight through it. At this point, we simply don't have anything like a large enough catalog of asteroids, nor in anything like sufficient detail, to know if this is a freak accident or the norm. Until you know enough of the basics, you can't know anything about the complexities.
This is further supported by AT&T's argument that BSD was contaminated through exposure to AT&T's intellectual property. That lawsuit was settled and not won, if I recall correctly. Novell's employees are now subject to exposure to information Microsoft provably regards as intellectual property - otherwise there'd be no point in a sustained relationship. If code can be contaminated because of exposure to someone else's code, even if that code is never actually used in the final product, then anything Novell releases may be contaminated.
(If the "contamination" argument sounds absurd, then logically I'd have to agree. However, this is the main reason that Open Source developers are careful to "clean room" such stuff, where the person who sees the original code or reverse-engineers the original specs does NOT write the code but ONLY documents the inputs, outputs and relationships. The code-writer is then much safer from claims of contamination, as they are merely writing code to an abstract description of what the code does, they are not writing it based on anything they themselves have seen in other code elsewhere.)
Is there infringing software? Well, that depends on whether anyone was careless. If Mono, SMB, CIFS, Samba, Samba-NG, NTFS, VFAT or any other potentially patentable concepts were developed without rigorous clean-rooming, then there is a risk of contamination. A risk does not mean contamination actually occurred, and contamination is not a proven legal notion as far as I know, as the AT&T fiasco is not much of a legal precedent to work from.
That's for copyright, though, this is patent law. There, if there's an infringement, it need only be within the confines of the patent, it need not be the result of copying or contamination. A completely independent development would violate a patent, as a patent is on a process, not an instance. I believe this is true even if the independent development was parallel to - or even preceded - the patented work. Prior art is not automatically safe - many of the victims of Thomas Eddison were prior inventors and I believe case law (any prior ruling, even if insane, stupid or blatantly absurd) dictates what a judge can do.
This is actually one very major weakness of ANY Open Source license - because prior art isn't safe, corrupt corporations can take an Open Source product, patent the ideas within it and sue the original author. In theory, this shouldn't be possible. In practice, there are good reasons for believing it is not only possible but a very reasonable possibility.
(My counter to that is that exploitable bugs in the law should be fixed and that workarounds are little more than exploits of bugs or weaknesses within the exploits and are therefore at best a temporary reprieve. I'd quite like to see changes that ensured that if prior art was sued for patent infringement that the patent and rights were transferred to the defendant on the grounds that they're more likely the real inventor. I'd also like to see software patents filed in a radiation-proof concrete-lined, lead-lined corrosion-resistant, nuke-proof steel drum before being dropped into a bore-hole that reached the inner limits of the Earth's crust - that being a far safer place for them than in the hands of lawyers. Well, unless the software patent lawyers were also filed in said drum, in which case it wouldn't matter what their hands held.)
Because you have a transferable token, then it would appear - to my untrained eye - to be a transaction that has involved the payment of that token.
Now, as I understand it, the reason general "free distribution" cases wouldn't be covered is that there is no such token payment and therefore no instrument by which copyright can be carried. This would explain why the general case is NOT covered, but would also mean that this case unquestionably is.
There is another aspect to this. A "Gentlemen's Agreement" requires an agreement between two individuals to be witnessed by an independent third individual. If there is a Gentlemen's Agreement to honour the license as a statement of copyright, then I believe there would be a case for claiming that copyright would apply whether or not it would normally have done so.
Ok, so we have the two parties. Is there an independent witness to an agreement? That one might be interesting to argue. I'd claim that Richard Stallman knows neither party but DOES know what the GPL means and can therefore be an independent witness to the effect that by accepting the GPL, the licensee has accepted the Gentlemen's Agreement. Not sure how well that would stick, though. My suspicion, though, is that if the judge reads and approves of the GPL as a legal instrument, but is 50-50 as to which way to go, the Gentlemen's Agreement argument might be enough to swing it, if such agreements are recognized in that State.
The last possibility is that if the licensee claims that they're not bound by copyright, and since copyright is the sole instrument to allow the licensee to use ANY GPL product whatsoever, Stallman may just declare the licensee Accursed and have them prohibited from using or obtaining ANY GPL product whatsoever, for all of eternity. (Under the license, that IS the standard penalty for blaspheming the GPL in this way.) This would not impact the current lawsuit as much as it opens the licensee up to a lawsuit for EVERY GPL product they use, distribute or possess. Quite simply, the costs involved in defending a massive distributed denial-of-sourcecode attack could cause some serious damage, and because it is an explicitly stated curse, not a single one of those lawsuits would likely be thrown as frivolous.
(Actually, I've seen weird rolls. When GMing a Rolemaster game, I had the unfortunate luck to see a Black Reaver get slaughtered by a first level character... Ok, ok, there are those who could argue that greater undead/greater demon combos shouldn't be put against first level characters anyway, but they didn't HAVE to pick up the bright, sparkly, foot diameter gemstone...)
The upshot of all this is that the researchers hopefully(?) have better evidence of their claims than a few buzzwords which don't really amount to a whole lot. 600 million years is not a long time, geologically speaking - or even evolutionarily-speaking - and I'm not convinced that every necessary process to get from Iceworld to habitable planet could occur in such a short space of time. I could be wrong, but I would need some VERY hard evidence.
What sort of evidence? Well, certainly volcanoes existed back then, and the eventual form lava takes depends on how rapidly it cools. Find me a lava bed in a land mass that would have been tropical at that time (based on current theories of plate movement) and which would not have been connected to open water, but where the rock structure shows clear evidence of cooling of the kind geologists associate with plunging molten lava into ice, AND where that rock also shows clear evidence of prolonged frost cracking. This would not be "solid proof", but it should be very persuasive.
Any other possibilities? Sure. Ice sheets and glaciers form very distinct U-shape valleys. Very very distinct from the V-shape of river valleys. If, indeed, no rivers existed 600 million years ago, only ice sheets, then it follows that any valley of that age or older MUST show the characteristics associated with such ice. Sure, a river may have cut through after, but that can't affect the sides of the valley above it. It can only affect the ground that it was cutting through at the time. The ice sheets, however, will have NOT been level with the ground but will have risen above it. Thus, the ice WOULD have reshaped any river valleys.
Thus, if a single valley, anywhere, shows clear and distinct evidence of cutting by water prior to 600 million years ago, OR if any valley of that time-frame does NOT show evidence of cutting by ice, the theory is falsified. Either will do.
If their claims stand up to the igneous rock test AND the valley test, then I'd be inclined to take the theory more seriously. As it stands, I simply don't see that they've enough to base their claims on. You know when academia is going south when Art Bell's old show could claim greater evidence and a more rigorous analysis. Hey, I happen to think the show has a lot of merit, both in examining controversial and unorthodox thinking that would not otherwise get the hearing it deserves and also as excellent entertainment, but we're in DEEP trouble when the fringiest of the fringe sound more careful, more exact and more scientific than highly-skilled, highly-trained, highly-paid experts.
I dunno. I could very easily imagine the richer suspects paying the cops $4,000 to brain-scan the poorer ones, especially if the rich suspects are innocent. (Time is money, reputation is even more money, and at the current rates, this might easily be a sound investment to some people.)
Ultimately, your post was exactly what I would call the perfect response - calling me on those things that were either not right or not clear, adding the necessary information (emphasised because this is such a vital point), and doing so in a way that is perfectly friendly. Good communication skills are rare in any technical field - you'd have to look in the top 1% of IT to find any who are in the top 50% of even the average population - so it's always good to see people who are good communicators. Besides which, the only way I can be right all the time next week is to learn what the errors are in my thinking this week!
For the polygraph tests, I was working on the basis that a polygraph test is a known quantity. The confidence limits on accuracy, repeatability and process are well-established and well-defined. They also already have the equiptment, experience in using it, etc. fMRI lie detection is a relatively unknown quantity, the confidence limits are - ergo - much broader. Starting from the statement "at most it admits the polygraph is less accurate", it is not obvious how to compare the two. At what point does something become less accurate, when you're dealing with distributions and not point values?
There are many ways to do this, but here's two. The easiest approach, if you have enough data, is to find the ranges that give you the same confidence limits - preferably 1% or 0.5%, as anything above that is absurd. You pick the worst end of the range for the fMRI and the best end of the range for polygraph tests. In order to absolutely, firmly, definitively say that fMRI actually is better than polygraph, the polygraph's upper limit has to be strictly below the lower limit for fMRI - otherwise, there is a non-zero possibility that polygraph tests are superior to fMRI. (Not just in some specific case, but in general.) Because courts presume innocence and because lies are more harmful than truth is helpful in investigative work, it's better to assume the least accuracy than the greatest accuracy. That means we now pick the lower limit for polygraphs as being the effective merit of the results. Given the absurdly wide margins there must be for fMRI, if we assume polygraph's limits are below that, the lower limit for polygraph has to be very close to zero.
(If the limits overlap, even to the smallest degree, then replacing all polygraph systems with fMRI would be stupid, as it would be impossible to be sure if that was making things better or worse, at vast cost. Nobody, not even some of the Government agencies I've dealt with in the past, could possibly be so stupid. So, if they made such a move, despite the gigantic spread of bounds for fMRI and assuming the choice was rational, it would seem that we can deduce the worst-case for polygraph would mathematically need to be very close to zero.)
The second method, which assumes some data but nowhere near as much, is to take the average expected reliability of the test and place the lower bound at two standard deviations below that. We don't even need to know the values for fMRI for this, we just need to know them for polygraph tests. Two standard deviations would be enough to cover the majority of meaningful cases and lop off all of the tail-end extreme or unusual situations. My guess is that the lower bound for polygraph tests, using this rather crude method, would again be perilously close to zero. This is easier with very little data (you need the mean and either the variance or the standard deviation), but it's harder when working with raw data because at no point do you know if you've calculate
...there's a lot of interest in work being done to use fMRI for lie detection. There are specific areas of the brain that light up when you lie, even if you aren't conciously aware that it is a lie, from what I understand. However, nobody has the foggiest what the accuracy level is (it's too new of an approach), fMRI is vastly more expensive than a polygraph system, only those who did the one study are even remotely qualified to conduct such a test, the psychological aspect is completely unknown (as opposed to a polygraph test where it's all there is), switching to it would essentially admit that every polygraph test they've ever performed is 100% bogus, and it simply doesn't have nearly the same BDSM coolness to it as a chair with straps.
There are a thousand markup languages, covering everything from geometric data to maths equations to typesetting to fonts. (BLOBs seem to defeat the whole purpose of an ASCII-driven markup system, IMHO.)
There are a million capabilities that couldn't be done as extensions (multicast mosaic was a definite curiosity and only touched the outer fringes of what's possible).
There are a billion ways that any browser could be tweaked to tune it dynamically for not just the system but also the network weather of the time.
The sheer volume of possibilities would overwhelm anyone, and many browsers have met with unfortunate ends as a result. (I can't even remember the name of Sun's pre-Java web browser.) However, those that don't venture out have all eventually stagnated and died.
My question, then, is how do you choose? By what black magic do you discern what will kill and what will cure?
It also makes me wonder about african bees. These are much more wasp-like in their passion for mindless violence, and are far inferior to regular honey bees when it comes to processing pollen, so must presumably be midway on the evolutionary scale.
Then, one must also consider that there are MANY species and sub-species that fall into the category of honey bee. Since the genetic code is relatively short, it should be moderately easy to understand how they interrelate. I would assume it would also be possible to genetically engineer honey bees that have far greater efficiency. (The "killer bees" are the results of a catastrophic experiment to mix african bees and American honey bees, together with allowing them to escape into the wild. All-round, excellent proof that gross stupidity can be bred into researchers. What I'm thinking of is having a totally sealed environment, similar to biosphere 2, and a program that's based more on science than air warfare grants.)
Buzzwords, buggy, stickysituation, beeosaurus, waspiewasbee
The chance of them knowing is the probability of them finding the information multiplied by the probability of knowing the value multiplied by the probability of producing a workable exploit.
The chance of you knowing if they know is the probability of them knowing multiplied by the probability of you knowing who the bad guys even are, multiplied by the probability of obtaining real information (they can jam anyone monitoring them by flooding the information space with junk information), multiplied by the probability of you knowing you even have real information, multiplied by the probability of being able to determine what the information actually means.
Counterintelligence is an exceptionally difficult field with a painfully poor track record. Most published successes have been by a series of sheer fluke events and staggering luck. Most published failures were unlikely to be anything else. We don't know about the unpublished stuff, but percentagewise, are we more likely to see bragging over achievements or failures, if both can be equally hidden?
I'm not saying that everything should be published, merely that it should not be assumed that not publishing is the same as others not knowing.
Now, can a case ever be made for publishing everything? Yes. Game Theory requires that all "full information scenarios" have a strategy for one side and one side only that will ALWAYS result in the winning conditions being met, no matter what the other side does. It is possible to imagine situations, particularly in computing where there is essentially no randomness and a "full information scenario" is possible, where the outcome can be guaranteed, if you want it to be.
No matter what anybody else might say, it is not the job of an enemy to make your life easy, so we shouldn't expect them to. We should expect them to do the researcxh, the legwork, the analysis to figure everything out. They might indeed just wait until someone tells them, but that should be a bonus. It should not be your modus operandi. In computer security, you must assume that there are opponents out there who could have all of the industry-standard backdoor passwords, a complete printout of every Operating System and network device QA test that failed and got overlooked, and a copy of the highest-end vulnerability scanner the commercial sector has going for it.
Hell, we know that a Russian spammer got a tier-1 backbone provider to turn off Blue Frog's Internet connectivity. Turning off a link like that is very traceable, but appears to have been regarded as mere amusement for the backbone provider. The same provider is hardly likely to show scruples when it comes to handing out internal or commercially-sensitive data, software or anything else. Given the repeatedly low scores on security for many US government departments and the almost routine mishandling of classified data, there are probably those in the information black markets who know more national secrets than the entire White House combined. If one backbone provider is riddled with corruption and pwned by organized crime, then we must assume that such people are unlikely to be avoiding big money out of a sense of decency and moral fortitude.
But if the most dangerous people have the most dangerous information already - and that includes whatever terrorists might actually exist - then most of the obscurity only serves to increase the value of what has already been stolen. This makes the thieves rich, the criminals dangerous, and the politicians popular for appearing to do something, but it doesn't make anyone else - users, vendors, bystanders - any better off at all. Illusions are fun on the stage, but they should be left there.
Also, if you have a name that can resolve to both A records (IPv4) and AAAA records (IPv6), there is a big problem of prioritization. This is important, as there is no guarantee that the two sites would be the same and the way that URLs are currently handled does not make it easy to choose one over the other.
Transitory addresses can also pose a problem. 99% of the work is handled by the OS, but any cookie that uses an IP address must be updated to the new address if it is to remain useful. The same is true of IPv4, but IPv4 users haven't used mobile IP much and generally use a fail-over or load-balancing system that presents a single address to the outside world.
These are all corner cases, we're just going to see a lot more corner cases with IPv6 than we did with IPv4, and I'm not convinced any implementation is taking that into account.
The destruction of the two standing Buddhas was a terrible disaster for archaeologists and historians. Partly because, as we know from the vicious infighting in these disciplines over whether the Seven Wonders ever existed, there is a tendancy to ignore documented evidence if there's no visible evidence. If the Taliban find the Sleeping Buddha before the archaeologists do, they may be able to eliminate any proof it ever existed as we only have the vaguest of hints of a third giant statue near the location of the first two. We have no idea where it is or what other archaeological finds might be located around it, but being buried means that it is possible it is in far better condition - maybe with the original paints intact - than the two standing statues were.
Importance level: High. Enough of the paints survived on the two standing statues to make educated guesses as to the composition, but it's hard to be certain when you're dealing with microscopic flakes that have been exposed to any number of chemicals in the air and rain (or in the rock itself), where you can't be sure how many layers there were, or even how well the paint was mixed, and where photosensitive chemicals might have done almost anything (except tap-dance) in that sort of timeframe.
(Ok, so the importance is only high for people with an interest in Afghan history, Buddhist history, Silk Road history, monumental carvings, archaeology in general, ancient technology, or who would quite fancy being able to only paint the outside of their house once every five hundred years and still have it look good.)
It is also important to consider that people are less careful with those things that they feel no ownership of, have no understanding of, and are devoid of any connection towards. That's to be expected and should surprise nobody. But, by implication, negligence or destructive tendancies towards historic artifacts is most likely when the individual or society are totally disconnected with history. Making the ownership explicit and universal can only help to empower communities.
The International Court of Justice, the African Union, and even the United States, are other examples of where members have voluntarily sacrificed absolute authority in favour of having separation of powers and outside agencies to which certain responsibilities can be delegated.
I see no contradiction, then, in doing this for world heritage. It is merely an extension of an arrangement that has generally worked rather well. (It is only when powers cease to be seperate, and absolute authority forms, that serious problems arise. Total and irrevocable seperation of powers won't eliminate war, but it would likely eliminate a great many wars.)
Your example of oil is an interesting one, for many reasons. Britain's North Sea oil reserves were badly mismanaged, through absolutism. America did just as badly but has escaped the worst of the suffering for now. In many poor countries where oil is to be found, you see exploitation and environmental destruction on a horrendous scale, usually with the consent of the Government there. The same is true for gold mining, where miners are exposed to a range of utterly lethal chemicals (often in the drinking water) with the consent of those in authority. I can't be certain of it, as I've not been there to ask those living in such conditions, but I'd be willing to hazard a guess that if they could earn the same or more AND keep their arms, legs and brain intact, they'd not give a damn who "owned" these resources.
Absolute ownership of the land is indeed a part of the "Nation-State" concept, but that concept is slowly being torn down as worn-out and decayed, susceptable to corruption and inherently economically unstable. America was the first of the modern nations to look for alternatives. States have some powers, the Federal Government has others. The judges, legislators and civil servants are all totally distinct - well, in theory. That's a six-way split of authority with minimal overlap. It's by no means perfect and it could probably do with being split further, but the US - by outlasting all other superpowers - has shown that the Nation-State is inferior by design to a more collaborative system. It just has to figure out how to get a collaborative system that's any good.
"But it belongs to (country)! Why should (some other country) keep it?" In the end, none of this "belongs" to a country. History cuurered everywhere at the same time. (Duh!) For the most part, the political boundaries that marked these countries no longer exist, the political entities have vanished into oblivion, no living direct descendents who could claim even a moral ownership are known to survive, so for the most part the only meaningful designation is "world heritage" (which I believe to not be used nearly enough and most definitely not recognized nearly enough).
So, if object X is being, or would very likely be, damaged by being in country Y, I believe country Y has lost all right to the ownership of object X. I don't like the fact that Britain has the Elgin Marbles, but I like even less the fact that they'd be destroyed by pollution if they were ever returned. The Greece of back then no longer exists, any more than the Egypt of the Pharaohes exists today. In some cases, there simply isn't a country in which an object is truly safe. In that case, you document every last facet like crazy and hope. (You can't move the Great Pyramid and you certainly can't hide it, though reducing pollution might cut down on the deterioration.)
But what makes something "world heritage"? The object itself? Usually no. Except in some rare cases, the object has no value in and of itself. For inorganic objects, it is the information the object posesses - from the chemical structure through to any symbols or writings on it, and the information associated with it - where it was made, when, how and why, where it was found, the nature of the site, other items found there and their respective characteristics and associations, and so on. These are the things that have any lasting meaning. Once you know the object - totally - you can always make another using exactly the same materials, tools and methods.
For organic objects, it's tougher. If a bone is damaged or destroyed, there is next to nothing you can do. And time is rarely kind to anything of organic nature. Tutankamun is in very bad shape now and the remains will probably not survive a whole lot longer. Part of that is due to Carter's team, but part is due to Egypt having very high levels of acidic pollution and acid rain. You can't expect much to survive under such brutal conditions.
The other problem with organics is that there's much less information you can obtain. With luck, you can extract mtDNA, maybe even use modelling to produce an impression of what the person looked like. Bodies found in peat bogs and ice fields give slightly more information, perhaps yielding clues of fashion, food and culture that artifacts alone can't. We learned a lot from "Pete Marsh" and the iron-age traveller murdered in the alps, but such finds are almost never in any kind of context, so there is very little you can do to connect them with what was happening at that time. "Pete Marsh" - Lindow Man - might date anywhere from prior to the Roman invasion to a hundred years after the Boudicca Rebellion, making it very hard to know what sort of context is involved.
Getting back to thieves vs. archaeologists - IMHO, it's not a binary thing. I would argue that the "absolute" thief is one who destroys information in search of money, even if that involves destroying the thing they're trying to find. (When archaeologists started paying money per fragment of Dead Sea Scroll recovered, some of the locals cut fragments up so that they could get more money.) I would argue that the "absolute" archaeologist obtains all information, even if that means never reaching the object. (We now have GPR scans of Edward the Confessor's tomb, but reaching it would destroy countless artifacts and could potentia
...it will have to end eventually. As with any system, it is limited to the ability to communicate. In this case, Intel can't communicate with their chip designers and AMD won't communicate with anyone who isn't stinking rich. (I've yet to get them to do so much as reply to an e-mail, answer the phone or even return a call.)
I remember hearing once about a prominant scout leader who was a homosexual. What was his name... oh, that's it! Lord Baden Powell, founder of the Scout and Boy Scout movements. (He was also a British spy, which would imply a certain disregard for the opinions of those whose information he stole.)
Another area that has been worked on has been to allow database developers to embed more and more complex SQL-based scriptlets and "helper" functions into the database. However, it is a truism that interpreters (even of bytecode) are painfully slow and offer nothing that a module plugin mechanism wouldn't have (provided you could install modules via SQL statements) and most developers would be better leaving data processing to the clients rather than the server anyway. (The closest I've seen was Informix' blade technology, and that was horribly unreliable and tedious.)
The bottom line is this: There have been a lot of database innovations over the years. Some - I'd say most - have been great ideas but the popular interpretation and implementation has been terrible. This is not the fault of any of the ideas, but I would take it as suggestive that programmers are not being as imaginative or creative as they could be.
My question? How can a database company:
In which case, the article is worth three full houses and a horizontal line.
Size isn't necessarily given up. If you use materials that have half the mass per unit volume, you can have twice the volume. Car shapes tend to waste a lot of material (cuboids have a large surface area to unit volume) and a lot of internal space to aerodynamics (you still want aerodynamics, you just don't want to waste as much space). Frontal area (as far as the air is concerned) is not necessarily the same as the frontal area as far as the occupants are concerned - there was an experiment in building double-skin racing yaughts, where the outer skin was porus and the inner skin had grooves. The idea was to build a hull that broke up turbulance within the hull.
You also need to consider that the engine isn't terribly efficient. 80% of what goes in the air intake is nitrogen. Nitrogen "burns" to form various oxides, but consumes far more energy than it releases. It would be possible to electrostatically enrich the oxygen intake - you can place a charge on oxygen far more easily than on nitrogen, then use the charge to seperate the two) but it works out that because you've only a very short time to do this, it's extremely hard to get any kind of net gain at all. I don't know of any other method of de-nitrifying the air intake, but the more you can exclude, the more energy you will extract for the same amount of fuel and oxygen. (You might be able to get the same effect by extracting the energy faster or cooling the engine better, as endothermic reactions require there to be more than some critical level of energy. I've not really pursued that line of reasoning much, as cooling an engine massively is not a trivial thing to do. Electrostatic seperation requires a y-shaped plastic tube, two metal grids and some wires.)
Improving engine efficiency is one option, but not the only one. The storage of energy is another. Simply tune the engine to work optimally under a very narrow range of conditions (because then you get the best performance) and convert the energy extracted into a more usable form. Regenerative braking can also conserve some energy. It's significant but not vast. Of course, getting the energy through far more efficient means is only one part. The other part is converting the energy into momentum for the car. The process (involving transmission, gears, traction, air resistance, etc) is highly non-linear. It would be better to reduce the gear ratios, have more gears, and have the automatic transmission pick a gear based on all related conditions and not just engine revs (which we have now eliminated anyway).
Air resistance is by far the biggest killer, as cars are horribly shaped with gigantic vertical slopes at the back, a huge hole below the engine cavity and absolutely no sense of aesthetics in the plumbing underneath.
Also, athletic bodies are often damaged or disfigured with massive hormone imbalances and other severe (and occasionally fatal) problems. Gymnasts, for example, do not mature correctly and often suffer from muscle and bone disorders. Body builders, weight-lifters, etc, can disfigure their hearts - I would not expect life-expectency to be nearly so high. Rugby players - well, I can see them evolving into a whole new species that has less to do with class and more to do with causing sheer terror when barreling down the playing field. Soccer players can suffer damage to hearing or their pre-frontal lobes, from a mixture of heading and smashing into the ground at high speed. It's usually not lethal, but if you look at the various team managers for the England squad, it's clearly harmful to thought processes.
I can't see how the Swedes would turn into troglodytes. Oh. That's the British. Ah. I always thought there was something odd with the people in Milton Keynes.