From what I understand, they use (VERY LARGE) conventional film, then scan it. However, if you can figure out how to make a precise step size that is a fraction of a pixel in width and height, there is nothing to stop you raster scanning the image directly onto a standard resolution sensor. The pixels you gather are bigger than your step size, but a sufficiently large number of samples would allow you to interpolate what the sub-pixels would need to be in order for each and every overlapping superpixel to match the information collected.
(eg: if you've actual light intensities of 2 4 8 4 2, and your sensor's pixel size is 3 times that - so it overlaps 3 of these light intensities at a simgle shot, then you will get values of 2 6 14 16 14 6 2, assuming you can move the sensor one third of the distance covered by a single pixel and the extreme limits of the sensor allow the last pixel to be 2/3rds covered. Simply subtracting out what you've already collected within the window allows you to infer what the remaining 1/3 of the window MUST be for the values to be correct.)
This use of filtering is by no means your only option. I would also totally disregard all colour sensors and use a colour pinwheel. Scan with a red filter, then scan with a green filter, then scan with a blue filter. The technique was used by a Russian photographer to make colour photographs of very high quality in about 1910 (although by having microsecond filter switches, rather than several minutes of ripping the camera open and swapping plates, you could avoid the distortion caused by motion). It was also used by John Logi Baird, when he invented this strange thing called television, but his colour displays were absurdly primitive even for the time and were dumped in preference to black-and-white rasterised displays. Regardless, the use of a high-resolution monochrome system with a colour wheel to create the illusion of a colour display has a long and illustrious history.
Greyscale CCDs and other light-sensitive devices are often at vastly higher resolutions than colour systems, so by using one such device and a rotating wheel to filter correctly, you would be able to produce a much higher resolution device than you could from a purely photosensitive device alone.
A third method is instead of using just a converging lens, if you added a diverging lens just after the focal point, you could spread the image out. Instead of trying to mimic pixels closer together, simply move the light rays further apart. The effect is logically the same. You could do the same thing by moving the sensor further away, as the density of the light falls off with the square of the distance, but it's hard to get a good photograph with a thirty-foot camera.
Of course, there's no reason you couldn't combine the techniques to produce even greater resolutions yet, or special-purpose photographs that cannot be produced using a conventional system. For example, if you had a four-colour pinwheel, you could produce photographs that people with tetrachromatic vision could correctly see. By using subtraction/bitmasking for pixel interpolation, it may be possible to devise a cheap way to handly high contrast/high dynamic range images that conventional digital cameras are useless at. Mind you, you'd need a camera that supported JPEG2000 or OpenEXR, which most cameras don't, but that's a relatively minor software issue.
All in all, camera technology is advancing for those people who have a particular need for it to, but vendors have no interest in supplying Joe Average with high-tech gizmos. Aside from the fact that Joe Average can barely take photos, the professional guys have large bags of money and are willing to pay. Who, in their right minds, would sell a dirt-cheap gigapixel high-dynamic-range eight-colour motion-cancelling tea-making camera for a hundred bucks to ten thousand wannabes, when they can sell the exact same camera for a hundred thousand to the ten photographers in the world who could afford them? For every dollar of overhead, the second option is nine thousand, nine hundred and ninety times better for the vendor.
You can only prove a winning strategy exists in a full information scenario. No such proof exists when only partial information exists, although you can approximate it pretty well if you get close enough to the full information case. Ergo, there will be cases when giving the "enemy" genuine information is actually the correct thing to do.
There is also the perspective that obscurity is not the same as security. If you have secret A and are trying to prevent B from knowing it, you can NEVER be certain that B does not have that information. If they obtain knowledge of A, and keep that hidden from you, then your obscurity becomes a weapon against you. This is the problem the Germans had once the Enigma ciphers were broken. By relying totally on obscurity, the Germans became extremely vulnerable. Obscurity is a VERY dangerous tool.
By far the best tactic is to assume the enemy could know everything - not necessarily that they will, but that they could. This introduces a degree of fault-tolerence into actions. It does not rely on an assumed weakness that may not exist (and therefore make those carrying out the action the weaker party), but assumes that the opposition is as competent and capable as it chooses to be. As this is often much closer to reality, it is a better assumption to make.
In terms of encryption, for example, using an obscure algorithm puts you at gigantic risk as it can't have had the eyeballs to verify that it is indeed secure. Furthermore, people are more likely to use weak keys, as they won't see the point in taking care, as they're working on the basis that they don't need to. A very stupid practice. The best you could do is make the algorithm public, utterly destroy any delusions of absolute mental superiority, and force people to work damn hard to use the algorithm correctly. If the enemy finds a fault and keeps it secret, they would have done so anyway, so you lose nothing. If Joe Smurf on sci.crypto finds a fault and publishes it, you will have time to fix the bug or switch to another method. Overall, you lose nothing.
Assuming the enemy is an idiot, merely because they're the enemy, is the best way to lose a battle or a war. Either that or acting stupid.
There are exceptions. Linux is an exception, because it doesn't matter what agenda any individual has, everyone else has the right to pervert that agenda to suit themselves, within the limits of the license. Lordi is an exception, because the money-grubbing pundits at the Eurovision Song Contest found it hard to argue with a hoard of demonic creatures - even if they were from Finland.
However, most attempts to create exceptions on any kind of large scale have (so far) been corrupted to the point where they cease to be freedoms and enslave those who embrace them. Indeed, there are good reasons to believe that freedom of the individual enslaves the collective, and freedom of the collective enslaves the group, that you cannot be simultaneously a free individual AND a member of a free society. There are also good reasons - totally independently of money - for believing that both extremes are unstable. As soon as one segment becomes more free than another, or more influential than another, there will be a natural drift of power away from those with less and towards those with more. Eventually, neither the individuals nor the collective will be free.
You will never, ever see an introverted, non-judgemental, empathic, intellectually exploratory, emotionally self-sufficient President of America. Geeks make up about 10% of the population, and socially-conscious geeks probably make about 2-3%. That 3% probably understands the dynamics of the world better than the other 97%, but their total influence is a big, fat zero. It's not even remotely close to even their proportion in society. This is in part because they're far too busy doing things they consider important, but it still means that their "freedoms" are dictated by some other group, which makes it more a permission than a freedom.
Having said that, nobody has yet developed a workable alternative, including the intellectually exploratory and socially conscious, suggesting that freedom will remain a mere delusion for a long time yet. In the example in the parent post, power will naturally drift into the hands of those who would dump toxic waste (it's cheap) or despoil a place to gain an advantage (political or economic expediency). It's mostly about money, just not entirely about money. It's also about what makes a person good at what they do, and why that eliminates a large fraction of the populace from ever having any meaningful say.
Even longer answer: There is a mathematical problem, called "The Byzantine General's Problem", which asks a similarish question: "If there are N people in a group, and M of them corrupt any communication that takes place, what is the smallest ratio of honest people to corrupt people that would allow accurate communication to take place?" This is very closely related to crytography and secure comunications. A variant of this problem is used to describe how you would divide a key into fragments, such that if parts of the key are revealed, the message remains secure, and if parts are destroyed, you can still read the message.
Methods along these lines are used by NASA to make messages from deep space probes "more readable". They don't contain more information, per se, but the message can be damaged in transit by noise and still be perfectly readable. This form of "more readable" certainly therefore exists.
Now, what about other forms of "more readable"? Well, you're stuck with the semantic information in the original. There's nothing you can do about that, as there are no meaningful semantic parsers for computers at this time. You've a few trivial syntax parsers, but that's it. You can't therefore enhance the meaning, but you MIGHT be able to enhance the delivery. However, this would not be by means of encryption (which is a mathematical process) but by the application of standard rules.
What about images? Images and sounds are a special case, because these are processed very directly by sensors that can be described mathematically. Unlike a word, which may be processed by the same senses but is then REprocessed by the linguistic centers of the brain, what you see and what you hear goes through a process that is much easier to understand. I do not know if anybody has actually tried the experiment, but I can see no reason why you could not "uncrypt" such data in a way that provides the brain with less noise and more of the key features it is looking for. This would arguably be enhanced from the perspective of understanding, but without experimental data, I don't think anyone could tell you by how much.
A variant on that theme, which MAY under certain circumstances be possible, would be to both encrypt AND uncrypt. All singal processing is subject to a glitch called aliasing, where two or more distinct inputs produce identical outputs. There is no guarantee that two brains will share an identical set of aliases. In consequence, it should be possible to produce an image comprised entirely of such aliases as exist for the intended recipient. The intended recipient will see the intended image, but nobody who does not share those aliases will. This is encryption, in that we are applying a well-defined reversible mathematical function with a well-defined data set and well-defined key. It merely happens that the key used is biologically stored over a bunch of neural circuitry in the brain. This does have one major strength - duplicating the key is going to be a very very hard problem. It also has one major weakness - your recipient had better not walk into any walls, drink strong liquor or otherwise change their neural key. Make that two weaknesses - nobody has a friggin' clue how to do this kind of stuff.
The problem is that the modern military has forgotten many of the lessons history taught their predecessors. Rommel was highly regarded, not because he followed some textbook solution or blamed the manufacturer if things went wrong, but because he innovated, experimented and improvised. The same is true of many of the "great" commanders in history - Julius Caesar disarmed the Celtic navy by using hooks on giant poles to rip the sails off. Hannibal got ruddy great elephants over the alps and invented whole new forms of combat. The American revolutionaries created the sniper.
Battles are not won or lost by whoever has the best terms and conditions from the manufacturer. If you're losing, you won't be around to complain, and if you're winning, you generally won't care.
Every time a major power (such as the US) has paid more attention to giving kickbacks to corporate sponsors than it has to producing successful products or successful missions, that power has had its arse well and truly kicked. Sometimes the power wins anyway, but it is not because of its unimaginative and self-serving attitude, it is despite it. It's not very hard to win when you have total land, sea and air supremecy, and can do round-the-clock carpet-bombing campaigns. (But even then, failure of imagination is lethal. Operation Market Garden got slaughtered because of such egotism.)
Personally, I dislike military structures. I find the notion of winning an argument by having the winner define what the argument was to be primitive and tribal. However, if we're going to have such organizations, we might as well make sure they're functional and concious, rather than degenerately repeating every mistake history has ever recorded.
...it actually is your fault. 99.99% of the time, phishing scams are detectable easily enough and it is indeed the fault of consumers who act like sheep. However, that isn't guaranteed to be the case, all of the time.
The anti-spam agency, Blue Frog, was essentially defeated because of a Russian spammer knowing enough corrupt/criminal network engineers along the Internet backbone to be able to blackmail them. This means that a phisher could potentially inject scams that even the most discerning of geeks could not distinguish from the genuine article. This is only in part because the Internet depends so heavily on trust. It is also a major failing of the banks and other organizations who only use security for logins (and even then are starting not to). It's usually very bad security, as well.
Where a scam is made possible because the organizations are failing to adhere to any kind of meaningful security standard, the banks should pay not just in full but double - once for their neglect and once for their incompetence.
Going back to the backbone issue, I'd add one further point. Where a scam or other criminal action is made possible because of corruption by network engineers on the backbone, those engineers should never see the light of day again and the network company should be fined to within an inch of its miserable corporate life.
Finally, where a country either passively or actively encourages crime over the Internet, I would prefer at the very least if there was some process to electronically isolate them. Completely cut all wired and satellite links that can be identified with that nation. Zapping a few Internet Cafes with Predator drones would be kinda cool, too.
Any natural number that is a prime can be split into a sum of natural numbers (where the number of elements in the sum is at least 2) for all natural numbers that are primes other than 1. It is always possible to have one of those numbers non-prime (except for the natural numbers 2 and 3). By dividing the non-prime numbers into the prime factors, and then repeating the process, you can split the original number into some expression e, where if we swap the constants for variables (or unknowns) the general form is E. There will be a number of possible values of E, so I'll call the set of all possible expressions for that prime {E}. Ok, now let us say that the set of all possible sets of expressions for all possible primes is {{E}*}. Because I'm allowing any general solution, there will be a finite number of generalized solutions (ie: using undefined constants) but potentially an infinite number of specialized solutions where the constants are all defined.
5, for example, is a solution to the general formula A(x^C)+D, sum(x)+C, Cx+D, and so on. (Because A, C and D are constants, they would need to be the same constants for all primes you apply the expression to. Otherwise, you've not really generalized or simplified anything.) It is easy to show this has an infinite number of special solutions - pick any value of C and for any equation with D in it, simply subtract/add the necessary value to give you the prime you want.
The original question, then, is whether there is any prime number (other than the three special cases listed) for which {E}/\ {{E}*} = {E}. That would be a prime for which no possible expression to derive it, no matter how generalized, can ever be used to derive any other prime number. Intuitively, it seems apparent that no such prime exists, as there will be many primes with an infinite number of solutions that do not overlap. However, there are also an infinite number of primes and I have no obvious way of proving that the infinite set of primes is a perfect subset of the infinite number of derivations.
Let us define something else, which is #({E}/\ {{E}*}) - the number of general solutions for a given prime that are also general solutions for at least one other prime. The larger the prime, the larger the number of possible generalized solutions, because the larger the number of subcomponents you can break it into which can be turned into equations. However, the number of overlaps is not so easily determined. It is very easy to imagine a large prime in which most general forms are the same as only a few other primes, whereas a smaller prime might overlap with a great many primes in a great many ways. It is also possible to imagine a set of primes in which all primes in that set have the same general form, but no member has a general form that is the same as any outside of that set.
(It is easy to prove that there exist two primes which have no general form in common, as that is simply the same as the proof that no general equation for primes exists.)
This leads to interesting possibilities - "islands" of primes that are totally disconnected from all other primes, "peninsulas" where you almost have an island but some perfect subset of the cloud does have a general form in common with other primes, "mountains" where you have a massive number of general forms totally in common with a large number of primes, and so on.
If you were to draw out the interrelationships between primes as a topology, what do you see? A random blob? A sea of islands? Multiple large masses that are otherwise disconnected? If islands, or multiple disconnected continents, exist in prime number space, does this mean that prime numbers aren't a single, definable set of numbers at all, but multiple concepts that should (in general) not be treated as the same at all? Will the map indicate that we can generalize the definition of prime number in a useful way, that the concept can be usefully extended and meaningfully applied outside of the natural numbers?
...I personally would just stick with the standards. However, the display is not great on Firefox, Opera or Konqueror, with the control widget window sometimes covering a full screen or more. It's sporadic. I haven't seen too many other display bugs, that's the main one, but it's one I'd really like fixed. as it almost has to be a code bug, not a browser bug.
There are a myriad of other features I would like. The ability to sort was a key feature for me in the classic display, and this new display could have vastly more power on things like that. (For example, sorting by probable association, so if there are multiple threads discussing the same thing, they are together.) You are limited on server-side processing, as it's time-critical - you're serving 100,000 users (the same circulation as a national newspaper), so powerful server-side features can be a real headache. With that kind of userbase, it's much better for the server to just update an index of probable keywords and deliver that to the client to do any further processing or sorting.
That, to me, is an important tool. Information is useless if it's so scattered that it takes more effort to collate it than to learn it from scratch. The ideal would be to be able to instruct the browser to recursively go through the related articles listed at the top and pull those indexes as well, so that users have the power to view the history and background of a discussion across multiple articles. (It won't prevent rehashing, which is inevitable, but it may encourage people to move further in their thoughts than is otherwise likely to happen.)
What has always made Slashdot exciting is that it has dared to challenge conventional wisdom on how news works - even the UK's Guardian newspaper cites Slashdot as the inspiration behind the blogs attached to articles, and the BBC's user moderation system is very likely a derivative of the system we all love and use. It's experimented with presentation, filtering, tagging, cross-referencing and windsurfing. The new front-end provides a thick-client interface, with all the possibilities that implies. All that power (Power! POWER! And it's even better than an IBM POWER! Bwahahahahahahaha!)
STU III is your friend in these situations. No authorized key, no communication. Which means that if your inter-department phones are STU III-grade, then you technically don't need to know who is on the other end, you only need to know that the keys are solely provided to authorized people. Everything else can be inferred through your normal web of trust.
Why should beta tags matter? Let's say you have a bank of servers, say 3xN of them. N are maintaining live connections that reach the outside world. Another N are cloning the connections and therefore completely in sync, should any of the original servers fall over (ie: zero latency/zero loss standby). The third are on some kind of hot standby status, so they can be brought up to speed within a few seconds. Oh, and they're all using LinuxBIOS, so reboot time is on the order of seconds as well. If a machine fails, the live standby machine takes over seamlessly, the hot standby syncs up by the small increment needed to become a live standby machine, and the original server is rebooted and brought up to hot standby using the latest system snapshots.
You now have a system that may not be 100% perfect but would reduce most outages to nothing a user could detect.
So far, so good. Now, we distribute the machines - scalable reliable multicast is a nice toy for connection sharing, and peer-to-peer systems provide the mechanism for distributing snapshots over a dynamic topology. We're now at the point where we don't rely on the ISP of the service provider, because there is no single service provider, and we don't rely on a single path over the Internet, as the primary and live standby machines are 100% interchangable at all times, and the odds of both paths being blocked for longer than TCP timeout is pretty remote. However, we can cover for that by enabling multipath routing, so that the Internet's routers are capable of tracking multiple paths from A to B. A blocked connection, at that point, would require massive failures at a great many points, at which point you are more likely to be concerned with the asteroid that just struck than about reaching Amazon.com.
Would this work? Yes, if we distinguish server providers from service providers, and if billing was heirarchical and transparent, rather than direct. Then you could treat the entire Internet as one uber-gigantic super-cluster and have applications transparently migrate across it according to network and performance requirements. This would require some major improvements in the field of computer security, but involves no computation not already being done by proxy meshes, OpenMOSIX and peer-to-peer systems. Technically, this is entirely possible, and by having heirarchical billing, it would be possible to sanely handle the costing. Money would go one step upstream, one step downstream or to a neighboring peer. Nobody else.
Then, the software could be as beta as you like, as there are no single points of failure, or even a single pair of points that would allow catastrophic failure.
Would this be a good idea? Well, depends. Programmers could earn a fortune off the IT security contracts that would be needed to provide a dynamically self-compensating service mesh.
"We're all hopelessly paranoid, making trust impossible, and rigidly stratified, making flexible response or communication impossible, oh and so totally self-centered that the mere notion of cooperation causes outbreaks of hives".
Come to think of it, that's what your translation says, too.
In England, it is a legal requirement that all makes over the age of 12 practice with the longbow. This qualifies you to be a debt collector in cases of spammer bankrupcy.
War is for one thing only--the destruction of your enemy.
...that war was about achieving strategic and tactical goals, preferably by causing minimal damage in the process. That destruction - whether of people, installations or anything else - is invariably going to cost so much more in the long-term (and encourage hostility, vengence, etc) that it is vastly cheaper to win without harming anyone or anything than it is to cause widespread devastation. That more wars were lost by slaughtering the enemy and winning the wrong battles than have ever been lost by compassion and consideration. This is the dictate of Sun Tzu and has been the accepted wisdom of military strategists (with any brains) and wargamers for many thousands of years.
You want examples besides Afghanistan and Iraq? Well, try the Battle for Monte Casino, where the allies suffered terrible losses and nearly faced a total defeat purely because they carpet-bombed a monastary that was standing nearby not doing a damn thing. (The Germans realized that the crazily-collapsed massive stone masonry would make excellent cover, so when the allies charged, they were simply mown down in their hundreds, totally unable to get at the enemy. All through their insatiable need for total destruction.) Total extraneous cost: Many tens of thousands of lives and a few billion dollars, for a hilltop that turned out to be quite unnecessary.
Ok, so can I name any battle or war won with virtually no destruction at all? I'd use the battle for the Channel Islands, in the English Channel, as an example. No deaths, no destruction. None. Whatsoever. Either when the Germans invaded, or when the British recaptured them. Not a single shot fired, not a single bomb dropped. Ok, were they defended? Hey, anyone who goes there can see bloody great reinforced concrete gun emplacements, bunkers, trenches and fortresses. Anyone with a metal detector can find unused bullets and unexploded bombs still scattered over the landscape. It's a wonder the population isn't providing the primary funding for the prosthetics industry. Those islands were defended by people quite capable and quite willing to ride out any kind of attack, the German occupiers in particular who included fanatical brigades. The two opposing armies negotiated a peaceful transfer in each direction. Total extraneous cost: Nothing.
...is completely correct. I don't think there's any serious doubt we need more super-colliders. I'd actually quite like it if they had a double-ring super-collider so that instead of striking a stationary target, the particles struck other particles also moving at a relativistic velocity. The velocities are not additive, but the difficulty of accelerating something follows a power law, whereas accelerating twice as much to the same velocity is a purely linear problem. Ok, hitting targets moving at 99.99% of C is damn hard, but we're reaching the limits of what we're capable of building with only one of the masses moving.
Also, since e=mc^2 is a bi-directional function, it should be possible to "fuse" photons into matter. That's what happened in the early universe - matter is far far younger than the universe itself. However, c is very very large, so c^2 is horrifyingly gigantic, which makes creating matter an extremely difficult problem. However, if you rely on atom-smashing, you cannot possibly prove that you have produced every subatomic particle that could exist, as there are starting points that are entirely valid but are not atoms and some of these starting points are just too unstable and short-lived to smash up. There are therefore some particles which will need to be manufactured from simpler components - not an easy task, but obviously possible as it has been done in accelerators in the past. That only works for relatively large synthetic particles, though. You cannot fabricate something that is on the same scale as a higg's particle but different, as the higg's particle is the smallest you can get and still have matter. Beyond that is just energy. If you wanted to see if alternative solutions exist for particles on that scale, you would need to condense it out of energy. You've no choice in the matter, because that's the building block on that scale.
(Also, quantum foam is littered with quantum black holes and quantum worm holes. Your left eyeball contains more quantum-scale black holes than all the micro-scale black holes that every accelerator on Earth will generate from now until the sun runs out of hydrogen fuel. No, you can't use that as an excuse in an eye exam. You can use it as a demonstration of why such phenomena are of absolutely no significance, though.)
It's not obvious what a negative mass would do in a positive gravity. Experiments with the Casmir Effect demonstrate negative energy densities, which would imply we can experiment with negative mass, but nobody has been able to do much beyond cause the aparatus to shatter, which limits things a little. If you can wait a couple of hundred years (provided civilization hasn't fried itself and/or the planet by then), we'll have the technology to answer that question. Until then, any answer you can come up with has equal footing with any other, as we've absolutely no meaningful data to work with, even as far as deciding which solution is the simplest needed but no simpler.
99.9% of everything that has been observed is obvious. Once it has been observed. I would claim that the blatant abuse of the term "innovative" by companies, trade magazines and even those programmers who should (in theory) know better is proof enough that what is blatantly obvious to you and me is NOT obvious to a great many others. Rather, the term has become diluted and polluted to the point where it fails to serve a useful purpose and is merely marketroid mind-mangling.
The purpose of my post was two-fold, of which one was to draw attention to awards. I do not regard it as a grand unified theory to claim that awards (a) be honest, (b) serve a purpose beyond free advertising, and (c) have integrity. I don't know what degrees you have, but I'll bet you worked damn hard for everything you have been awarded, especially any industry commendations or professional society recognition. Yet corporations (be they music, film or computer software) can get a thousand times the recognition purely on the strength of buying a bunch of judges more drinks than everyone else. Whilst you or I can be convicted in court of lying in a resume = and rightly so, when was the last time a marketing manager or CEO did time for even the most blatantly fraudulant claims, when made through a trade journal's "reviews" or "awards"?
The second purpose was to get REAL information on the product in question. Not a list of advertised features, or even a list of actually functional features (which can be much harder to obtain), but a definitive explanation of what innovation actually exists within the product. That's not something the company will tell you. That's not even something you can gather through a code inspection. That's something you learn through cold, hard experience alone. The answer, from what I can gather, is that it does not alter the relative sizes of the time-slices and does not introduce (or eliminate) timing errors or otherwise alter program behaviour. That's a damn hard thing to do, particularly in the cases of hard real-time and parallel applications (which, these days, can mean any multi-threaded program on a multi-core and/or multi-CPU system). I believe Sun's CPUs have 6 or 8 cores, so that would include virtually everything the program is used on that is not strictly serialized. A trivial example: when message passing, a send before a receive will NOT behave the same way as a receive before a send, One will be zero-copy, the other will be buffered. By introducing even the tiniest delay in one thread but not in the other, you can alter both the code used and resource requirements. If a fault exists in only one path, the mere presence of the product can totally skew the results, making it useless for tracking runtime behaviour. It takes a lot of effort to write a program that has no visible footprint whatsoever to the other programs being run.
Damn good run-time inspection software has been around for a long time, as the mainframe experts noted. (Which is a valid point. Mainframe time is expensive and crashing a mainframe is an open invitation to feed deep ocean trench microbes.) The question then becomes one of whether the invisible footprint is an innovation of something genuinely new, or an evolution of a pre-existing technology. Are the mainframe tools precursors or a wholly independent branch? That is an entirely valid question and one I believe is eminently suitable for Slashdot. In fact, I'd argue that it's vastly more suitable for Slashdot than almost any other tech site, because you are going to find people who actually have an answer here. There will be people with the real-life experience and skills to be able to say.
(It always amuses me when I'm modded informative, interesting and flamebait, particularly when some of the replies get in-depth and highly informative, as it shows how different perspectives can totally alter the interpretation of a posting. If there are any psychologists reading Slashdot, this would be superb research material for a paper on the tribal nature of western society at many levels and the mutual uninteligability of those groups outside of their scope. I'll even waive my usual fee.)
That's because Linux geeks are increasingly wary of "Thought Contamination", "Intellectual Property" and the ghosts of AT&T lawsuits past. (If you even SEE proprietary code, you can be accused of Intellectual Property Theft if you write something similar.) SCO is doing an amazingly bad job of going that route, but the laws now prohibiting reverse-engineering and favouring legal solutions to technical problems are strong indicators that we will see more and more such cases. For this reason, Linux coders are wary of even LOOKING at Sun's code. It may be available, but if you re-use an idea seen in that code, you are liable for horrific lawsuits. There is no evidence that Sun's current management would do this, but they won't manage Sun forever and policies shift like the desert sands when things get rough. The risks and liabilities are just too damn great. It is safer to ignore the gift horse entirely than to risk the fate of Troy, even when the technology is desperately needed. It is better to live in desperation than to be sacrified on the alter of the IP Gods. Alive, you can code an alternative.
No, Sun have skewered themselves over their license. As good as the program may be, the DMCA Death Star and Darth IP are not worth it.
Now THAT is a genuinely informative, useful and brilliant reply. Thank you! I don't care how steep the learning curve is, to me that's a non-issue, and if the course is only a week then I don't consider that to be effort. People are in so much of a rush, these days, and learn bugger all. Learning any true craft takes time, patience and practice. A week-long course (40 hours, if it's full-time, which would be unusual) seems awfully short, when a good solid understanding of the theory, how the program relates to that theory, how to make effective use of the program and how to stretch the program to its limits would - for most computer programs - take at least a year and would likely be obsolete within five to ten years after that. (Which is why I feel that it is stupid for America to insist that its population pay for education and not provide any time for that education to occur. It leads to skill shortages, stagnation and brain rot.)
Sun's license is a non-issue iff (no, that's not a typo) it plugs into existing instrumentation or existing patches for such instrumentation. Soft linking is not deemed a GPL issue. The same is true if Sun added a module to the kernel. There are so many hooks (such as LSM), profilers and monitors in the kernel that can be exploited - even, as in the case of LSM, when that was not the purpose of the hook, that I can see no problems with getting all the information Sun could possibly want or need without the need to introduce code that would require the GPL. And if such code were added, ONLY the hooks and inlined data grabbers would need to be GPL. It would allow anyone to write a comparable module, but a standard Linux instrumentation API is badly needed anyway and is therefore inevitable. It would seem better to have Sun set the standard, than be cut out of the market - especially with a quality product. Remember Betamax vs VHS - it's the standard that wins, even when quality is truly important.
No, what I said was "these are the criteria I am claiming it needs to meet to qualify as excellent and innovative respectively, it needs to meet both to qualify as both, and no matter how good it is in just one category, if it does not have the necessary qualifications then it does not deserve an award that claims it does."
DTrace may be excellent. It may be the best program since sliced silicon. But if it's not innovative, then whatever other award it may be entitled to, it is NOT entitled to an award for innovation UNLESS all other entrants also had zero innovation. But if zero is the top score then all other entrants suck with regards to innovation. I don't see how you can possibly avoid such a conclusion. If any other entrant exceeded zero innovation, but DTrace did not, or if any other entrant exceeded DTrace in innovation assuming comparable program quality, then the other entrant was more deserving of the award. I don't see how you can possibly avoid that conclusion, either.
On the other hand, DTrace may be highly innovative, the most innovative program the Universe has ever seen. Alien races from a thousand light-years away may be gasping in awe at the staggering freshness and novelty of the concepts contained within it. However, if the coding is crap then it is NOT entitled to an award of excellence UNLESS all other entrants are also crap. But if the coding sucks that badly all round, then all other entrants are crap with regards to programming. As before, I don't see how you can possibly avoid such a conclusion. If any other entrant exceeded zero program quality, but DTrace did not, or if any other entrant exceeded DTrace in coding standards assuming equal innovation, then the other entrant was more deserving of the award.
Since all of this directly follows from the stipulation that an award for excellence and innovation must offer excellence and innovation (except where the best effort in either category is zero), it seems much simpler to state the stipulations and treat the reader with the courtesy and respect of assuming they can perform simple expansion of the set of invariants that must be true.
It's interesting that it's done in mission-critical systems like ATC - it shows that the dynamic probes used must be of a very high grade. High-profile high-risk systems aren't the kind of thing where you fix the serious bugs later. (Can you imagine what your average 747 pilot would say if they were told to stay in a holding pattern until Patch Tuesday?)
Oh, I worry about bugs throughout the infrastructure. OS bugs, compiler bugs, system library bugs, firmware bugs - all of these can turn even a 100% perfect application (were such a thing to exist) into a smouldering heap of junk. They are unpredictable and almost impossible to trace in those situations where the programmer only has the application to look at. Dynamic instrumentation is, I believe, slightly worse in that non-fatal bugs in a system call, for example, would eventually be inferred by observing that data is mangled after such a call in all places in the system. With dynamic instrumentation that uses embedded operations in the code, the same holds true. Instrumentation that runs in parallel and dips in at intervals is much more of a problem, as there is then almost certainly no correlation between anything in the code and side-effects from the instrumentation. You can eventually deduce that the errors must be external to the code (and all functions linked to it), but in any seriously large application, or if the OS is complex (or, worse, black-box), this can take a hellish long time.
Probably the worst-case scenario is where the side-effects aren't direct. The instrumentation might very occasionally add a delay that, on rare occasions, causes a time-sensitive component of the application to miss a critical deadline. The bug would then not be in the code of either program, but would be in the sequence of operations of successive time-slices. Sequencing bugs are bloody murder, because they are not programming bugs. The code can be 100% clean and still have this class of bug. (Sequencing bugs are much more general than, say, race conditions, and would typically be at a much lower level.)
Debugging programs is extremely difficult and time-consuming to do right, because by the very fact that you are running in a debugging environment, you have changed the characteristics of the environment the program is running in (unless it is ALWAYS running in such a mode). Even disregarding all the above problems, I feel certain that the vast majority of programmers have encountered bugs that cease to exist when debugging information is added, or where the program is placed in a debugger... or, for that matter, ONLY exist when debugging information is added. The last of these is particularly nasty for those in rigid work environments. If there's some bug X that users are seeing that is obscured by bug Y that is introduced by debugging/instrumentation data, there are workplaces where fixing bug Y is not permitted as it is not a user-documented bug and so no time/money has been budgetted for fixing it. That can make fixing X really fun. You can spot projects that are likely a victim of the "no complaint, no fix" attitude - they eventually work just well enough but no better, ran way over on time and are likely to be fragile under unusual conditions.
Surprisingly, this is not a dig at the Usual Target of Slashdot Gripe. Rather, I've seen this attitude when employed within the public sector, which is notorious for producing an amazing amount of crap. Which is ironic, because the less formal and informal projects from the public sector are equal to or better than commercial projects. Sure, there are a lot of crap projects on Freshmeat, but if you look at the really good stuff, you'll see a lot comes from Government research groups, Universities and - occasionally - the US DoD, but they're all projects managed by geeks, not wannabe accountants. (How bad does a person have to be if they can only pretend to be an accountant?)
Quantum foam must be zero-sum, as it is mathematically equivalent to a perfect vaccuum, except on the quantum scale where a perfect vaccuum is impossible as it violates the second law of thermodynamics. Regular anti-matter is just a property of symmetry when applied to ordinary matter, and so does indeed have a positive mass. When you get into the game of virtual particles, where any number of pairs of particles must be creatable and destroyable without changing the state of the system (which would violate the law of conservation of mass and energy), the rules change quite dramatically. Quantum foam is fun stuff, as it cannot exist at a statistically-significant level, it can only exist at levels below that point as fluctuations in both directions from the mean.
First, the sum total of any pair of virtual particles must be zero. We are not talking here about traditional antimatter (which does have positive mass) but about virtual particles whose existance - even under the conditions of inflation theory - CANNOT violate the law of conservation of mass and energy. Quantum foam is mathematically identical to a perfect vaccuum EXCEPT on a sufficiently local level, which is clearly impossible if you go around ascribing positive masses to everything. (It is more common to call the opposite partner to a quantum foam particle an anti-particle, to avoid this confusion.)
As for your second point, I do so love rhetorical questions that confuse axioms with postulates. Especially as your question is easily enough answered (and shows lack of basic research skills). I shall leave it as an exercise to the interested reader (if any exist) to figure out why your point can either be correct about one part OR the other of what I wrote, but not both at the same time. It shouldn't take long. Personally, I prefer debates where it is the point under discussion, not the attributes of one of the discussors, that is debated. Put up, shut up or give up, but please can it with the personal insults.
Which only reinforces my opinion that Sun's software is not innovative. It might be wonderful, it might be a lot of things, but from what I've seen and what you say innovative it is not.
Slashdot Mirror
(eg: if you've actual light intensities of 2 4 8 4 2, and your sensor's pixel size is 3 times that - so it overlaps 3 of these light intensities at a simgle shot, then you will get values of 2 6 14 16 14 6 2, assuming you can move the sensor one third of the distance covered by a single pixel and the extreme limits of the sensor allow the last pixel to be 2/3rds covered. Simply subtracting out what you've already collected within the window allows you to infer what the remaining 1/3 of the window MUST be for the values to be correct.)
This use of filtering is by no means your only option. I would also totally disregard all colour sensors and use a colour pinwheel. Scan with a red filter, then scan with a green filter, then scan with a blue filter. The technique was used by a Russian photographer to make colour photographs of very high quality in about 1910 (although by having microsecond filter switches, rather than several minutes of ripping the camera open and swapping plates, you could avoid the distortion caused by motion). It was also used by John Logi Baird, when he invented this strange thing called television, but his colour displays were absurdly primitive even for the time and were dumped in preference to black-and-white rasterised displays. Regardless, the use of a high-resolution monochrome system with a colour wheel to create the illusion of a colour display has a long and illustrious history.
Greyscale CCDs and other light-sensitive devices are often at vastly higher resolutions than colour systems, so by using one such device and a rotating wheel to filter correctly, you would be able to produce a much higher resolution device than you could from a purely photosensitive device alone.
A third method is instead of using just a converging lens, if you added a diverging lens just after the focal point, you could spread the image out. Instead of trying to mimic pixels closer together, simply move the light rays further apart. The effect is logically the same. You could do the same thing by moving the sensor further away, as the density of the light falls off with the square of the distance, but it's hard to get a good photograph with a thirty-foot camera.
Of course, there's no reason you couldn't combine the techniques to produce even greater resolutions yet, or special-purpose photographs that cannot be produced using a conventional system. For example, if you had a four-colour pinwheel, you could produce photographs that people with tetrachromatic vision could correctly see. By using subtraction/bitmasking for pixel interpolation, it may be possible to devise a cheap way to handly high contrast/high dynamic range images that conventional digital cameras are useless at. Mind you, you'd need a camera that supported JPEG2000 or OpenEXR, which most cameras don't, but that's a relatively minor software issue.
All in all, camera technology is advancing for those people who have a particular need for it to, but vendors have no interest in supplying Joe Average with high-tech gizmos. Aside from the fact that Joe Average can barely take photos, the professional guys have large bags of money and are willing to pay. Who, in their right minds, would sell a dirt-cheap gigapixel high-dynamic-range eight-colour motion-cancelling tea-making camera for a hundred bucks to ten thousand wannabes, when they can sell the exact same camera for a hundred thousand to the ten photographers in the world who could afford them? For every dollar of overhead, the second option is nine thousand, nine hundred and ninety times better for the vendor.
There is also the perspective that obscurity is not the same as security. If you have secret A and are trying to prevent B from knowing it, you can NEVER be certain that B does not have that information. If they obtain knowledge of A, and keep that hidden from you, then your obscurity becomes a weapon against you. This is the problem the Germans had once the Enigma ciphers were broken. By relying totally on obscurity, the Germans became extremely vulnerable. Obscurity is a VERY dangerous tool.
By far the best tactic is to assume the enemy could know everything - not necessarily that they will, but that they could. This introduces a degree of fault-tolerence into actions. It does not rely on an assumed weakness that may not exist (and therefore make those carrying out the action the weaker party), but assumes that the opposition is as competent and capable as it chooses to be. As this is often much closer to reality, it is a better assumption to make.
In terms of encryption, for example, using an obscure algorithm puts you at gigantic risk as it can't have had the eyeballs to verify that it is indeed secure. Furthermore, people are more likely to use weak keys, as they won't see the point in taking care, as they're working on the basis that they don't need to. A very stupid practice. The best you could do is make the algorithm public, utterly destroy any delusions of absolute mental superiority, and force people to work damn hard to use the algorithm correctly. If the enemy finds a fault and keeps it secret, they would have done so anyway, so you lose nothing. If Joe Smurf on sci.crypto finds a fault and publishes it, you will have time to fix the bug or switch to another method. Overall, you lose nothing.
Assuming the enemy is an idiot, merely because they're the enemy, is the best way to lose a battle or a war. Either that or acting stupid.
However, most attempts to create exceptions on any kind of large scale have (so far) been corrupted to the point where they cease to be freedoms and enslave those who embrace them. Indeed, there are good reasons to believe that freedom of the individual enslaves the collective, and freedom of the collective enslaves the group, that you cannot be simultaneously a free individual AND a member of a free society. There are also good reasons - totally independently of money - for believing that both extremes are unstable. As soon as one segment becomes more free than another, or more influential than another, there will be a natural drift of power away from those with less and towards those with more. Eventually, neither the individuals nor the collective will be free.
You will never, ever see an introverted, non-judgemental, empathic, intellectually exploratory, emotionally self-sufficient President of America. Geeks make up about 10% of the population, and socially-conscious geeks probably make about 2-3%. That 3% probably understands the dynamics of the world better than the other 97%, but their total influence is a big, fat zero. It's not even remotely close to even their proportion in society. This is in part because they're far too busy doing things they consider important, but it still means that their "freedoms" are dictated by some other group, which makes it more a permission than a freedom.
Having said that, nobody has yet developed a workable alternative, including the intellectually exploratory and socially conscious, suggesting that freedom will remain a mere delusion for a long time yet. In the example in the parent post, power will naturally drift into the hands of those who would dump toxic waste (it's cheap) or despoil a place to gain an advantage (political or economic expediency). It's mostly about money, just not entirely about money. It's also about what makes a person good at what they do, and why that eliminates a large fraction of the populace from ever having any meaningful say.
Text is purely integer.
Even longer answer: There is a mathematical problem, called "The Byzantine General's Problem", which asks a similarish question: "If there are N people in a group, and M of them corrupt any communication that takes place, what is the smallest ratio of honest people to corrupt people that would allow accurate communication to take place?" This is very closely related to crytography and secure comunications. A variant of this problem is used to describe how you would divide a key into fragments, such that if parts of the key are revealed, the message remains secure, and if parts are destroyed, you can still read the message.
Methods along these lines are used by NASA to make messages from deep space probes "more readable". They don't contain more information, per se, but the message can be damaged in transit by noise and still be perfectly readable. This form of "more readable" certainly therefore exists.
Now, what about other forms of "more readable"? Well, you're stuck with the semantic information in the original. There's nothing you can do about that, as there are no meaningful semantic parsers for computers at this time. You've a few trivial syntax parsers, but that's it. You can't therefore enhance the meaning, but you MIGHT be able to enhance the delivery. However, this would not be by means of encryption (which is a mathematical process) but by the application of standard rules.
What about images? Images and sounds are a special case, because these are processed very directly by sensors that can be described mathematically. Unlike a word, which may be processed by the same senses but is then REprocessed by the linguistic centers of the brain, what you see and what you hear goes through a process that is much easier to understand. I do not know if anybody has actually tried the experiment, but I can see no reason why you could not "uncrypt" such data in a way that provides the brain with less noise and more of the key features it is looking for. This would arguably be enhanced from the perspective of understanding, but without experimental data, I don't think anyone could tell you by how much.
A variant on that theme, which MAY under certain circumstances be possible, would be to both encrypt AND uncrypt. All singal processing is subject to a glitch called aliasing, where two or more distinct inputs produce identical outputs. There is no guarantee that two brains will share an identical set of aliases. In consequence, it should be possible to produce an image comprised entirely of such aliases as exist for the intended recipient. The intended recipient will see the intended image, but nobody who does not share those aliases will. This is encryption, in that we are applying a well-defined reversible mathematical function with a well-defined data set and well-defined key. It merely happens that the key used is biologically stored over a bunch of neural circuitry in the brain. This does have one major strength - duplicating the key is going to be a very very hard problem. It also has one major weakness - your recipient had better not walk into any walls, drink strong liquor or otherwise change their neural key. Make that two weaknesses - nobody has a friggin' clue how to do this kind of stuff.
I sure hope the ions are moving, at least! Otherwise, things'll get fun when the charge causes the resistance of the air to break down.
Battles are not won or lost by whoever has the best terms and conditions from the manufacturer. If you're losing, you won't be around to complain, and if you're winning, you generally won't care.
Every time a major power (such as the US) has paid more attention to giving kickbacks to corporate sponsors than it has to producing successful products or successful missions, that power has had its arse well and truly kicked. Sometimes the power wins anyway, but it is not because of its unimaginative and self-serving attitude, it is despite it. It's not very hard to win when you have total land, sea and air supremecy, and can do round-the-clock carpet-bombing campaigns. (But even then, failure of imagination is lethal. Operation Market Garden got slaughtered because of such egotism.)
Personally, I dislike military structures. I find the notion of winning an argument by having the winner define what the argument was to be primitive and tribal. However, if we're going to have such organizations, we might as well make sure they're functional and concious, rather than degenerately repeating every mistake history has ever recorded.
The anti-spam agency, Blue Frog, was essentially defeated because of a Russian spammer knowing enough corrupt/criminal network engineers along the Internet backbone to be able to blackmail them. This means that a phisher could potentially inject scams that even the most discerning of geeks could not distinguish from the genuine article. This is only in part because the Internet depends so heavily on trust. It is also a major failing of the banks and other organizations who only use security for logins (and even then are starting not to). It's usually very bad security, as well.
Where a scam is made possible because the organizations are failing to adhere to any kind of meaningful security standard, the banks should pay not just in full but double - once for their neglect and once for their incompetence.
Going back to the backbone issue, I'd add one further point. Where a scam or other criminal action is made possible because of corruption by network engineers on the backbone, those engineers should never see the light of day again and the network company should be fined to within an inch of its miserable corporate life.
Finally, where a country either passively or actively encourages crime over the Internet, I would prefer at the very least if there was some process to electronically isolate them. Completely cut all wired and satellite links that can be identified with that nation. Zapping a few Internet Cafes with Predator drones would be kinda cool, too.
5, for example, is a solution to the general formula A(x^C)+D, sum(x)+C, Cx+D, and so on. (Because A, C and D are constants, they would need to be the same constants for all primes you apply the expression to. Otherwise, you've not really generalized or simplified anything.) It is easy to show this has an infinite number of special solutions - pick any value of C and for any equation with D in it, simply subtract/add the necessary value to give you the prime you want.
The original question, then, is whether there is any prime number (other than the three special cases listed) for which {E}
Let us define something else, which is #({E}
(It is easy to prove that there exist two primes which have no general form in common, as that is simply the same as the proof that no general equation for primes exists.)
This leads to interesting possibilities - "islands" of primes that are totally disconnected from all other primes, "peninsulas" where you almost have an island but some perfect subset of the cloud does have a general form in common with other primes, "mountains" where you have a massive number of general forms totally in common with a large number of primes, and so on.
If you were to draw out the interrelationships between primes as a topology, what do you see? A random blob? A sea of islands? Multiple large masses that are otherwise disconnected? If islands, or multiple disconnected continents, exist in prime number space, does this mean that prime numbers aren't a single, definable set of numbers at all, but multiple concepts that should (in general) not be treated as the same at all? Will the map indicate that we can generalize the definition of prime number in a useful way, that the concept can be usefully extended and meaningfully applied outside of the natural numbers?
There are a myriad of other features I would like. The ability to sort was a key feature for me in the classic display, and this new display could have vastly more power on things like that. (For example, sorting by probable association, so if there are multiple threads discussing the same thing, they are together.) You are limited on server-side processing, as it's time-critical - you're serving 100,000 users (the same circulation as a national newspaper), so powerful server-side features can be a real headache. With that kind of userbase, it's much better for the server to just update an index of probable keywords and deliver that to the client to do any further processing or sorting.
That, to me, is an important tool. Information is useless if it's so scattered that it takes more effort to collate it than to learn it from scratch. The ideal would be to be able to instruct the browser to recursively go through the related articles listed at the top and pull those indexes as well, so that users have the power to view the history and background of a discussion across multiple articles. (It won't prevent rehashing, which is inevitable, but it may encourage people to move further in their thoughts than is otherwise likely to happen.)
What has always made Slashdot exciting is that it has dared to challenge conventional wisdom on how news works - even the UK's Guardian newspaper cites Slashdot as the inspiration behind the blogs attached to articles, and the BBC's user moderation system is very likely a derivative of the system we all love and use. It's experimented with presentation, filtering, tagging, cross-referencing and windsurfing. The new front-end provides a thick-client interface, with all the possibilities that implies. All that power (Power! POWER! And it's even better than an IBM POWER! Bwahahahahahahaha!)
You now have a system that may not be 100% perfect but would reduce most outages to nothing a user could detect.
So far, so good. Now, we distribute the machines - scalable reliable multicast is a nice toy for connection sharing, and peer-to-peer systems provide the mechanism for distributing snapshots over a dynamic topology. We're now at the point where we don't rely on the ISP of the service provider, because there is no single service provider, and we don't rely on a single path over the Internet, as the primary and live standby machines are 100% interchangable at all times, and the odds of both paths being blocked for longer than TCP timeout is pretty remote. However, we can cover for that by enabling multipath routing, so that the Internet's routers are capable of tracking multiple paths from A to B. A blocked connection, at that point, would require massive failures at a great many points, at which point you are more likely to be concerned with the asteroid that just struck than about reaching Amazon.com.
Would this work? Yes, if we distinguish server providers from service providers, and if billing was heirarchical and transparent, rather than direct. Then you could treat the entire Internet as one uber-gigantic super-cluster and have applications transparently migrate across it according to network and performance requirements. This would require some major improvements in the field of computer security, but involves no computation not already being done by proxy meshes, OpenMOSIX and peer-to-peer systems. Technically, this is entirely possible, and by having heirarchical billing, it would be possible to sanely handle the costing. Money would go one step upstream, one step downstream or to a neighboring peer. Nobody else.
Then, the software could be as beta as you like, as there are no single points of failure, or even a single pair of points that would allow catastrophic failure.
Would this be a good idea? Well, depends. Programmers could earn a fortune off the IT security contracts that would be needed to provide a dynamically self-compensating service mesh.
Come to think of it, that's what your translation says, too.
In England, it is a legal requirement that all makes over the age of 12 practice with the longbow. This qualifies you to be a debt collector in cases of spammer bankrupcy.
You want examples besides Afghanistan and Iraq? Well, try the Battle for Monte Casino, where the allies suffered terrible losses and nearly faced a total defeat purely because they carpet-bombed a monastary that was standing nearby not doing a damn thing. (The Germans realized that the crazily-collapsed massive stone masonry would make excellent cover, so when the allies charged, they were simply mown down in their hundreds, totally unable to get at the enemy. All through their insatiable need for total destruction.) Total extraneous cost: Many tens of thousands of lives and a few billion dollars, for a hilltop that turned out to be quite unnecessary.
Ok, so can I name any battle or war won with virtually no destruction at all? I'd use the battle for the Channel Islands, in the English Channel, as an example. No deaths, no destruction. None. Whatsoever. Either when the Germans invaded, or when the British recaptured them. Not a single shot fired, not a single bomb dropped. Ok, were they defended? Hey, anyone who goes there can see bloody great reinforced concrete gun emplacements, bunkers, trenches and fortresses. Anyone with a metal detector can find unused bullets and unexploded bombs still scattered over the landscape. It's a wonder the population isn't providing the primary funding for the prosthetics industry. Those islands were defended by people quite capable and quite willing to ride out any kind of attack, the German occupiers in particular who included fanatical brigades. The two opposing armies negotiated a peaceful transfer in each direction. Total extraneous cost: Nothing.
Also, since e=mc^2 is a bi-directional function, it should be possible to "fuse" photons into matter. That's what happened in the early universe - matter is far far younger than the universe itself. However, c is very very large, so c^2 is horrifyingly gigantic, which makes creating matter an extremely difficult problem. However, if you rely on atom-smashing, you cannot possibly prove that you have produced every subatomic particle that could exist, as there are starting points that are entirely valid but are not atoms and some of these starting points are just too unstable and short-lived to smash up. There are therefore some particles which will need to be manufactured from simpler components - not an easy task, but obviously possible as it has been done in accelerators in the past. That only works for relatively large synthetic particles, though. You cannot fabricate something that is on the same scale as a higg's particle but different, as the higg's particle is the smallest you can get and still have matter. Beyond that is just energy. If you wanted to see if alternative solutions exist for particles on that scale, you would need to condense it out of energy. You've no choice in the matter, because that's the building block on that scale.
(Also, quantum foam is littered with quantum black holes and quantum worm holes. Your left eyeball contains more quantum-scale black holes than all the micro-scale black holes that every accelerator on Earth will generate from now until the sun runs out of hydrogen fuel. No, you can't use that as an excuse in an eye exam. You can use it as a demonstration of why such phenomena are of absolutely no significance, though.)
It's not obvious what a negative mass would do in a positive gravity. Experiments with the Casmir Effect demonstrate negative energy densities, which would imply we can experiment with negative mass, but nobody has been able to do much beyond cause the aparatus to shatter, which limits things a little. If you can wait a couple of hundred years (provided civilization hasn't fried itself and/or the planet by then), we'll have the technology to answer that question. Until then, any answer you can come up with has equal footing with any other, as we've absolutely no meaningful data to work with, even as far as deciding which solution is the simplest needed but no simpler.
The purpose of my post was two-fold, of which one was to draw attention to awards. I do not regard it as a grand unified theory to claim that awards (a) be honest, (b) serve a purpose beyond free advertising, and (c) have integrity. I don't know what degrees you have, but I'll bet you worked damn hard for everything you have been awarded, especially any industry commendations or professional society recognition. Yet corporations (be they music, film or computer software) can get a thousand times the recognition purely on the strength of buying a bunch of judges more drinks than everyone else. Whilst you or I can be convicted in court of lying in a resume = and rightly so, when was the last time a marketing manager or CEO did time for even the most blatantly fraudulant claims, when made through a trade journal's "reviews" or "awards"?
The second purpose was to get REAL information on the product in question. Not a list of advertised features, or even a list of actually functional features (which can be much harder to obtain), but a definitive explanation of what innovation actually exists within the product. That's not something the company will tell you. That's not even something you can gather through a code inspection. That's something you learn through cold, hard experience alone. The answer, from what I can gather, is that it does not alter the relative sizes of the time-slices and does not introduce (or eliminate) timing errors or otherwise alter program behaviour. That's a damn hard thing to do, particularly in the cases of hard real-time and parallel applications (which, these days, can mean any multi-threaded program on a multi-core and/or multi-CPU system). I believe Sun's CPUs have 6 or 8 cores, so that would include virtually everything the program is used on that is not strictly serialized. A trivial example: when message passing, a send before a receive will NOT behave the same way as a receive before a send, One will be zero-copy, the other will be buffered. By introducing even the tiniest delay in one thread but not in the other, you can alter both the code used and resource requirements. If a fault exists in only one path, the mere presence of the product can totally skew the results, making it useless for tracking runtime behaviour. It takes a lot of effort to write a program that has no visible footprint whatsoever to the other programs being run.
Damn good run-time inspection software has been around for a long time, as the mainframe experts noted. (Which is a valid point. Mainframe time is expensive and crashing a mainframe is an open invitation to feed deep ocean trench microbes.) The question then becomes one of whether the invisible footprint is an innovation of something genuinely new, or an evolution of a pre-existing technology. Are the mainframe tools precursors or a wholly independent branch? That is an entirely valid question and one I believe is eminently suitable for Slashdot. In fact, I'd argue that it's vastly more suitable for Slashdot than almost any other tech site, because you are going to find people who actually have an answer here. There will be people with the real-life experience and skills to be able to say.
(It always amuses me when I'm modded informative, interesting and flamebait, particularly when some of the replies get in-depth and highly informative, as it shows how different perspectives can totally alter the interpretation of a posting. If there are any psychologists reading Slashdot, this would be superb research material for a paper on the tribal nature of western society at many levels and the mutual uninteligability of those groups outside of their scope. I'll even waive my usual fee.)
No, Sun have skewered themselves over their license. As good as the program may be, the DMCA Death Star and Darth IP are not worth it.
Sun's license is a non-issue iff (no, that's not a typo) it plugs into existing instrumentation or existing patches for such instrumentation. Soft linking is not deemed a GPL issue. The same is true if Sun added a module to the kernel. There are so many hooks (such as LSM), profilers and monitors in the kernel that can be exploited - even, as in the case of LSM, when that was not the purpose of the hook, that I can see no problems with getting all the information Sun could possibly want or need without the need to introduce code that would require the GPL. And if such code were added, ONLY the hooks and inlined data grabbers would need to be GPL. It would allow anyone to write a comparable module, but a standard Linux instrumentation API is badly needed anyway and is therefore inevitable. It would seem better to have Sun set the standard, than be cut out of the market - especially with a quality product. Remember Betamax vs VHS - it's the standard that wins, even when quality is truly important.
DTrace may be excellent. It may be the best program since sliced silicon. But if it's not innovative, then whatever other award it may be entitled to, it is NOT entitled to an award for innovation UNLESS all other entrants also had zero innovation. But if zero is the top score then all other entrants suck with regards to innovation. I don't see how you can possibly avoid such a conclusion. If any other entrant exceeded zero innovation, but DTrace did not, or if any other entrant exceeded DTrace in innovation assuming comparable program quality, then the other entrant was more deserving of the award. I don't see how you can possibly avoid that conclusion, either.
On the other hand, DTrace may be highly innovative, the most innovative program the Universe has ever seen. Alien races from a thousand light-years away may be gasping in awe at the staggering freshness and novelty of the concepts contained within it. However, if the coding is crap then it is NOT entitled to an award of excellence UNLESS all other entrants are also crap. But if the coding sucks that badly all round, then all other entrants are crap with regards to programming. As before, I don't see how you can possibly avoid such a conclusion. If any other entrant exceeded zero program quality, but DTrace did not, or if any other entrant exceeded DTrace in coding standards assuming equal innovation, then the other entrant was more deserving of the award.
Since all of this directly follows from the stipulation that an award for excellence and innovation must offer excellence and innovation (except where the best effort in either category is zero), it seems much simpler to state the stipulations and treat the reader with the courtesy and respect of assuming they can perform simple expansion of the set of invariants that must be true.
Oh, I worry about bugs throughout the infrastructure. OS bugs, compiler bugs, system library bugs, firmware bugs - all of these can turn even a 100% perfect application (were such a thing to exist) into a smouldering heap of junk. They are unpredictable and almost impossible to trace in those situations where the programmer only has the application to look at. Dynamic instrumentation is, I believe, slightly worse in that non-fatal bugs in a system call, for example, would eventually be inferred by observing that data is mangled after such a call in all places in the system. With dynamic instrumentation that uses embedded operations in the code, the same holds true. Instrumentation that runs in parallel and dips in at intervals is much more of a problem, as there is then almost certainly no correlation between anything in the code and side-effects from the instrumentation. You can eventually deduce that the errors must be external to the code (and all functions linked to it), but in any seriously large application, or if the OS is complex (or, worse, black-box), this can take a hellish long time.
Probably the worst-case scenario is where the side-effects aren't direct. The instrumentation might very occasionally add a delay that, on rare occasions, causes a time-sensitive component of the application to miss a critical deadline. The bug would then not be in the code of either program, but would be in the sequence of operations of successive time-slices. Sequencing bugs are bloody murder, because they are not programming bugs. The code can be 100% clean and still have this class of bug. (Sequencing bugs are much more general than, say, race conditions, and would typically be at a much lower level.)
Debugging programs is extremely difficult and time-consuming to do right, because by the very fact that you are running in a debugging environment, you have changed the characteristics of the environment the program is running in (unless it is ALWAYS running in such a mode). Even disregarding all the above problems, I feel certain that the vast majority of programmers have encountered bugs that cease to exist when debugging information is added, or where the program is placed in a debugger... or, for that matter, ONLY exist when debugging information is added. The last of these is particularly nasty for those in rigid work environments. If there's some bug X that users are seeing that is obscured by bug Y that is introduced by debugging/instrumentation data, there are workplaces where fixing bug Y is not permitted as it is not a user-documented bug and so no time/money has been budgetted for fixing it. That can make fixing X really fun. You can spot projects that are likely a victim of the "no complaint, no fix" attitude - they eventually work just well enough but no better, ran way over on time and are likely to be fragile under unusual conditions.
Surprisingly, this is not a dig at the Usual Target of Slashdot Gripe. Rather, I've seen this attitude when employed within the public sector, which is notorious for producing an amazing amount of crap. Which is ironic, because the less formal and informal projects from the public sector are equal to or better than commercial projects. Sure, there are a lot of crap projects on Freshmeat, but if you look at the really good stuff, you'll see a lot comes from Government research groups, Universities and - occasionally - the US DoD, but they're all projects managed by geeks, not wannabe accountants. (How bad does a person have to be if they can only pretend to be an accountant?)
Quantum foam must be zero-sum, as it is mathematically equivalent to a perfect vaccuum, except on the quantum scale where a perfect vaccuum is impossible as it violates the second law of thermodynamics. Regular anti-matter is just a property of symmetry when applied to ordinary matter, and so does indeed have a positive mass. When you get into the game of virtual particles, where any number of pairs of particles must be creatable and destroyable without changing the state of the system (which would violate the law of conservation of mass and energy), the rules change quite dramatically. Quantum foam is fun stuff, as it cannot exist at a statistically-significant level, it can only exist at levels below that point as fluctuations in both directions from the mean.
As for your second point, I do so love rhetorical questions that confuse axioms with postulates. Especially as your question is easily enough answered (and shows lack of basic research skills). I shall leave it as an exercise to the interested reader (if any exist) to figure out why your point can either be correct about one part OR the other of what I wrote, but not both at the same time. It shouldn't take long. Personally, I prefer debates where it is the point under discussion, not the attributes of one of the discussors, that is debated. Put up, shut up or give up, but please can it with the personal insults.
Which only reinforces my opinion that Sun's software is not innovative. It might be wonderful, it might be a lot of things, but from what I've seen and what you say innovative it is not.