Err, well, no. They can take out features, they probably have in the past and they certainly will in the future. Linux is no different. Those who wrote code assuming Intermezzo or the kernel-based devfs would always be there are, well, victims of their own folly.
The same applies to hash functions. You have some mechanism for identifying which function you are using and then you use it. Hard-coding is for wimps and fools, and is almost always the true cause of backwards incompatibility. Correctly-engineered software either uses defines or, if it's really good, negotiates. Bad software, and pandering to it, is why Windows has accumulated more crud than a farmyard without the benefit of being able to use it to fertilize.
(Despite conventional wisdom, Windows is not popular because it supports bad software. It's popular because it's pre-installed and has enough good software that people can ignore all the bad crap. If Linux were to insist on supporting all of the bad crap written for it - do YOU use a.out binaries still? - for the indefinite future, it would not improve its popularity. If anything, it would risk the popularity it has gained and its forward momentum, all achieved by willingly sacrificing the bad from the past (with the exception of X11) in favour of developing the good for the future.
True, a good dose of salt will improve the hashes. (But it may lead to higher blood pressure.) My concern is that the vast majority of "hackers"/crackers are either skript kiddies or macho wannabes. They can exploit known weaknesses, they can use known techniques, they can download known black hat toolkits, but they will never discover or write anything worth a damn themselves.
If 99% of the risk comes from people with 1% of a functioning brain, it makes no sense to not take simple precautions that might (only might) marginally increase the risks with the remaining 1% for the short timespan it takes for you to plug in a different PAM module and re-enter the passwords.
You are entirely correct in saying that heavy salting will have the same effect as using a different algorithm, at least as far as pre-existing cracking tools are concerned and probably as far as any future cracking tools are concerned, provided there are no major function weaknesses found and the hash lengths are the same.
However, your salting method then becomes just as much of an unknown as the alternative algorithm, and all you do is rename the problem of using an untested algorithm without actually changing it. In fact, a sufficiently complex salting method will transform any hash function into any other hash function, provided they have the same basis and produce the same length of hash.
In the end, given that both proprietary and open source software have been subject to numerous zero-day vulnerabilities over the years, given that both have retroactively removed features, functions and capabilities with both regular and emergency patches, and that both expressly deny any "fitness for use" guarantee, the excuse of "but it might not work" sounds, well, feeble and ever so amusing. It's a "who dares, wins" world and those who do nothing will go nowhere.
You cannot "lead" from behind. Ergo, you cannot both be a leader AND and follower at the same time. If you are waiting until someone else tells you it's ok, you're a follower.
In a race, the ones who look behind them fall over. Those who look ahead at least finish (a key requirement in winning).
Skein was hardly designed by Joe Punk. Neither was MD6. Both have sample code. So who is doing all this DIY stuff you speak of?
You can either know you are safe or you can be ahead of the game. You can't have both, it's a choice. (You can in fact be safe, when ahead of the game, you just can't know it.)
Unless you suspect an attack on your home computer stash of downloaded music by the NSA, would you care to name any skript kiddies who have the maths noggin to break a system they can't just download the hack for?
Cryptography professors, as with all professors, are not generally that smart. Some are and they do well enough in research or are hired by three-letter agencies or really big companies involved in crypto. The majority teach because they lack the brains or the guts (it takes both) to innovate.
Oh, and I would remind you that Torvalds ignored HIS professors when it came to kernel design and opted to write his own code rather than use the publicly-available BSD4.3 or MACH tapes. I wouldn't claim to be his equal, but I would claim to have infinitely more parallels that some gutless 6-digit UIDer who, if they'd been the inventor of Linux, would have based it on TOPS-20 and never ported it to platforms never "tested" by experts for a true multitasking OS. Assuming they ever ported it from the "safe" minicomputer of their choice at all.
According to one expert in such matters, "If you want to be the best, if you want to beat the rest, dedication is what you need." Cautious types need not apply.
Depends what they "choose cancer" over. (yeah, yeah, doesn't apply in this case, but eventually will, so I'm going to consider the general case where it does.) We know that the nucleic DNA coding isn't the only thing that creates specific proteins. Junk DNA alters the interpretation, as do some of the other molecules that hang around the DNA.
If you were to edit that segment of code, you cannot be certain (with today's knowledge) what impact that will have. The retrovirus method of inserting DNA caused a rare form of leukaemia in some people as I recall. It is possible that unfortunate side-effect was due to accidentally altering the metadata or some other aspect of the coding in unexpected ways, or perhaps having the defective (but switched-off) gene still present caused problems as its own metadata would also still be present but would likely still have been active.
Therefore, altering DNA to remove the gene for one form of cancer (which can be treated) may increase your chances of getting a much less treatable cancer, or one that can't be treated at all. It might have all kinds of other nasty effects that we've yet to discover. Genetic science is still very much in its infancy. Bear in mind that if you start counting from when DNA was first discovered, about as much time has passed as had when the Apple IIe and Commodore PET were the latest in home computing. If you start counting from when the human genome was first transcribed, we're closer to the era of Colossus and ENIAC.
That's not an unfair comparison - look at deCODEme's lists of identifiable genetic diseases and genetic traits. Not very long, are they? And they don't require much in the way of hard evidence to add something to those lists, either - two studies will do it. We can assume anything not on those lists has evidence too weak even to pass such minimal muster, which is just about everything, and they don't even get into such things as DNA metadata and other coding abnormalities. That isn't too distant from the best knowledge the top-of-their-field experts had in computer science at Bletchley Park and certainly sounds very much like the state of homebrew computing in the late 70s - bits and pieces of wisdom with an awareness of lots of blank spots.
(Science moves a lot faster these days, but genetics is a lot more complex, so it all balances out. Twenty, maybe thirty, years down the road, genetic editing might be as safe and reliable as Open Source is today. Mind you, Firefox keeps crashing on me and I can never keep Bygfoot going past three seasons. Hmmm. Better make that forty from now, then, to be on the safe side.)
Using Linux as an example, most of the "mega patchsets" could be considered forks from the vanilla kernel but end up testing the component patches extensively, allowing for a better choice of what goes mainstream. In other words, the "forks" have accelerated Linux kernel development. The distro-specific patches - well, not too sure about those, but more than a few have also made it mainstream - again likely for the same reason. Better testing by more people. (This is not to fault the Linux development model, quite the opposite. It's praise for the number of imaginative ways that patches have been brought to people's attention and praise for the handling of the logistics of so many patchsets - pseudo-forks, I guess you could call them - that the Linux kernel developers manage superbly. Sun would be wise to look to these people for inspiration on how to combine the best from the forks and how to know what the best even is.)
Ignoring the attempt at humour for a moment, I wouldn't agree with the statement. I would, however, argue that neither action nor punishment fail to deter and in some cases the combination of action and punishment have created an incentive. This is true of copyright, illicit substances and probably a whole string of other offenses. This doesn't mean they should all automatically be legalized. The opposite of authoritarianism isn't anarchism. What it does mean is that such offenses should be treated very differently and much less punitively. In some cases, yes, abolishing the offense probably would make sense, but it's not wise to assume that is automatically the best choice. If Prohibition was America's biggest mistake, abolishing Prohibition was the next biggest. You don't fix one extreme by going to the other. You certainly don't fix it by trying to pretend it never happened.
England didn't ban heroin until the 50s (and only under heavy pressure from the US). There is little evidence it was seriously abused before then. But if they removed the ban overnight, it could be a total disaster. The shift in perceptions, the change in attitudes and the secrecy of the underground subcultures that abuse such stuff, not to mention the big pockets and bigger firearms of those wanting to protect their profits, would make an instant shift an instant disaster. If you're in sixth gear, flat out, you do NOT put the car into reverse. This does not make reverse gear a bad thing, it just means you need to take the intermediate steps first. The state of the system is utterly wrong for what you're attempting.
The problem with the drugs culture is that we don't really know what the right state is. The level of neurological research on how drugs affect the brain is minimal and knowledge of the effect on the rest of the body is virtually non-existent. Sure, this could be fixed. All you need is a battery of PET, MRI and fMRI scanners, drugs containing radioactive tracer isotopes and a bunch of volunteers stupid enough to have their brain glowing with positrons for dangerously long periods of time. You'll soon find where the uptake is, the effect the byproducts have, how the brain structure changes and how brain activity is altered. Because the changes and any build-up are gradual over an uncertain length of time, you're likely to kill a large percentage of the volunteers with brain tumours from continually pumping radioactive material into soft tissue, hence the level of stupidity required. Without the neurological data, though, it is quite impossible to form a scientifically sane policy. At the moment, there's a lot of superstition and religious nonsense by both sides, but there simply isn't any science worthy of the name. Without that, how can you know when and how to liberalize?
SIMD - Single Instruction, Multiple Data. Useful with highly parallelizable problems.
MISD - Multiple Instructions, Single Data. Radiosity in parallel with Raytracing would be an example of that.
MIMD - Multiple Instructions, Multiple Data. Very common in high-end supercomputing.
RDMA - Remote Direct Memory Access. Access another machine's memory as if it were your own. Bypasses the kernel and is very lightweight on the CPU(s), mostly done on the NIC. (NICs capable of RDMA are sometimes called RNICs.)
PIM - Processor In Memory. Advanced functions (like message passing) embedded into the RAM chips themselves, bypassing the CPU(s) and the Operating System entirely. Very useful when you're shunting data around, rather than crunching it.
Consider this. A multi-core processor will typically have one cache which is shared between cores. SMP systems, on the other hand, have one cache per CPU. For SIMD or MISD problems, a single cache is not a big penalty. For MIMD problems, it will kill you stone dead. Most "interesting" problems (the problems that eat Crays for breakfast, and enjoy a leisurely lunch of Blue Genes sautee'd with Information Silos in a brisk red hat) are MIMD. It's these problems that have people talking in hypercubes, Processor-in-Memory, MPI-enabled RAM, RDMA, and any other trick they can possibly come up with to shunt data around faster. When these guys can (and do) play with 60 gigabits per second bandwidth in both directions, they need CPU technology that won't take one look at that kind of hardware abuse and run screaming through the night.
Which is why the "maintainable" is such an important element. It's also why the full RASDIT cycle requires that on completion, you go back to the requirements gathering. The theory is that even though this method is slow, the software will converge on the users' needs as the amount of fresh code required will drop with each full cycle. It's entirely possible this is the methodology used in mission-critical systems, such as power stations, as you don't want to patch often and the requirements will change very slowly.
As I understand it, the waterfall method is not a full repeat of the entire cycle, but a partial implementation, partial test, followed by another partial implementation, partial test, cascading down through a whole series of these. There are a lot of very very short cycles in this method, at least as I was taught it.
RASDIT doesn't cascade down like that. Implementation is bottom-up, testing is top-down, they are very distinct phases that don't interact beyond the transfer of the code. There are no cycles and one phase does not start until the prior phase has completely stopped. I've seen Slinkies used to describe this methodology, but never waterfalls.
Now, I'll freely admit that the terminology I learned is (a) old, and (b) quite possibly localized, so we can probably account for 99% of the disagreement by means of such external considerations. If what you're describing is the current terminology, then that is the terminology I should be using. Nuff said. Ergo, I shall do so and thank you for bringing it to my attention.
The old slow method (RASDIT - Requirements, Analysis, Specification, Design, Implementation, Testing) is hardly the waterfall method. Waterfalling is a relatively rapid cycle between implementing and testing in small stages, each stage feeding into the next, as a cascading series of waterfalls. The RASDIT software life-cycle states that after testing comes deployment, after deployment you go right back to the start and re-assess requirements. There is no direct feed between one cycle and the next (you re-use from the last life-cycle, but that was the last life of the product not this one) and start-to-finish can take years to decades. If the waterfall method evokes thoughts of a gentle series of cascading waterfalls, step-by-step from start to finish, the RASDIT approach is a thousand-foot plunge. You're damn sure what you have is solid, or you're landscape. RASDIT is not used much these days, but that is the only approach that can be called "old" or "slow". Molasses is fast in comparison. (Tastier, too.)
RASDIT is the rite of passage of all serious software engineers. It is possible to write software this way and you will have a very high probability of producing robust, reliable, efficient code. It probably won't be maintainable, but it will very likely work and work damn well. All methods other than developing the code to the tests will produce inferior results, but they will produce inferior results far faster and in a form that is easier to debug.
I think the difference is that a lot of open source projects operate on a release-early-release-often philosophy - what software engineers refer to as the waterfall model. In such a scenario, bugs will appear in earlier releases and decline in number sharply over time. Closed source projects tend to operate on the more classical release schedule, which tends to be a lot slower with a lot more expansive software lifecycle, but SHOULD produce far far fewer bugs in each release.
MySQL, with the current Oracle owners, seems to have moved to more of the classical mindset. That's fine and I personally don't think there's anything wrong with it if it's done right. But to be done right, you need fewer bugs in a release than the rapid-cycling of the RERO approach, and should aim for code of a quality comparable to RERO over the same timeframe.
It is clear in this case that MySQL has failed the test. It is a good product, but the new methodology is fixing less than the old methodology. That needs to be when the developers stop and think about whether they want to try switching to something else or go back to what had worked well in the past.
You're right. The Rolling Stones painted it all black.
Seriously, yes, many scientists leave out data that doesn't suit their theories. It's still bad science and they're still bad scientists. The number of good scientists is extremely small. I don't see the scientific world in shades of grey, simply because either it's a valid, falsifiable, theory or it isn't; either the data was collected and presented or it wasn't. There isn't any room for shades of grey in science. Data selection is why Copernicus failed to see what Kepler observed - orbits aren't circles. This put science back centuries because of religious and metaphysical bigotry. Sorry, that ain't science, that's stupidity and as close to evil as science gets. I have no respect for such corruption. It's claimed Pythagoras had people murdered in order to keep mathematical proofs that contradicted his cult teachings from leaking out.
If you add up all the bigotry and corruption over the millennia, you start talking serious timeframes. I doubt medieval Europe could have put a man on the moon (see: The Tomorrow People, Rift in Time) but I could easily see astronomy, some areas of physics, mathematics and possibly chemistry being many hundreds of years further advanced than they are today, had people chosen to put scientific integrity over and above all else. Beyond Star Trek, but achieved yesterday, in those fields that are less sensitive to things that were probably not going to happen much sooner, like microtechnology and nanotechnology. To me, the incalculable damage done to humanity through such actions is unacceptable and beyond any classification of "greyness". There is no moral relativism in whether or not the square root of two is a fraction or an irrational, there is only a fact of whether it is or it isn't. It's extremely quantized.
(Relativism is valid in essentially every aspect of culture, philosophy, politics, and so on, but 0 != 1 no matter what. It's not dependent on the observer.)
This is clearly a new branch of Homeland Security for Cthulhu. Tired of people carrying around copies of the Necronomicon to dispell him, he has invested in an army of multi-jointed drifters to act as a lynch mob should any deep sea fish try to exchange knowledge of his whereabouts for a reduction in EU tuna quota.
The data doesn't care, but oil drillers are unlikely to give information that harms their potential to drill, and can afford to be "selective" on what they provide. They also have more than enough technical equipment and expertise to "improve" the data, if it is in their interests to do so. That is why it is generally bad science to get information from those who have a vested interest in your conclusions being what they want them to be. It has nothing to do with the camera and everything to do with the eyes far behind it.
Depends on the level you're talking about and the interface used. I'd consider many MMORGs, MUDs and "interactive fiction" games to make it extremely hard to differentiate between the character and the player. I'd consider games like Elite, Virus 2000 and TORCS to likewise blur the boundaries substantially. If you add in the current work on CAVE systems, neural interfaces and other such gizmos, you can definitely see the possibilities of a total collapse between virtual environments and reality. Mind you, I'm not so sure I'd be opposed to that. You'll notice that most (if not all) of the types of game, or examples of games, I gave have a very dedicated core "cult" following bordering on addiction (although apparently not everyone agrees, according to other stories on the front page). The "addictive" element, in my opinion, is exactly what is being talked about - the total collapse of the fourth wall.
Now, on the flip-side, if you look at games that are perhaps "popular" for a while but aren't really considered that addictive and are soon forgotten when something else comes along, you will see that they rigidly differentiate between player and character. It is my belief that it is this element that makes games exciting for people, that if you actually try to keep players at a distance from their characters, you will destroy the magic of the game entirely.
Is this just true of computer games? I'll argue no. RPGs that keep the player at a distance die off. Those that immerse the player survive and thrive. Is it only true of games? I'll say no to that as well. Books, movies and songs that draw the audience into the "sub-creation" (as Tolkien called it) are longer-lasting that those that don't.
Indeed, Tolkien's call for MORE sub-creation and GREATER destruction of the "fourth wall" is IMHO a better guide for how popular culture should develop. We should not fear our own creations, only our fear of them.
Anyone treated in hospital is likely to fill out a form asking if their DNA can be used for medical research. Anyone who has used a genealogical DNA service (and there are many) or one of the genetic disease detection services (there are rather fewer of those, deCODEme being one of the better-known) has their DNA on file on the service's database. Now, whilst there is some protection from unscrupulous departments abusing these kinds of socially-provided services, "law enforcement" gets a LOT of leeway when it comes to bending/breaking the rules. Also, bear in mind that the FBI has been caught using anti-terror legislation to cart off all of the computers in libraries and other communal services. All you need is one "potential suspect" on one of these massive databases and suddenly the powers-that-be have authorization to confiscate/duplicate the entire system, protections be damned. This is not the fault of such services, and they can have all the good intentions they like when it comes to privacy protection, but none of your records have been safe for many many years now. If there is a "fault" in this, it is the fault of those Americans who encouraged (through votes, money or even simply not telling their senators and representatives they were being stupid - inaction is just as bad as incorrect action) Congress to pass such paranoid legislation.
The British have been just as foolish, trusting (despite all evidence to the contrary) that it was safe to provide Government with such sweeping powers. Iceland has no doubt passed on it's thanks. I'm not a proponent of small Government by any means. Size doesn't matter, it's what you do with it that counts. And what's been done is the encouraging of the paranoid delusions of a few, to the detriment of many.
But... but... I want things to just work when I tweak things! And I want the latest of everything! And I want all the tools that I've bookmarked on Freshmeat that the distros don't carry! And... and... and I want a pony!
People are exposed to 2.5 tesla magnetic fields in conventional MRIs, and there are MRIs that go up to 9.3 tesla (even though that's a bit beyond what the FDA approves). There are scanners for animals that go up to 12 tesla. Even the least of these is well beyond anything the Earth generates. Clearly, however, magnetic fields are not as dangerous as has been assumed. At least, within reasonable limits.
This is a localized magnetic field, however. If you were to set up a Faraday cage, so that you were enclosed in a uniform magnetic field, you shouldn't ever detect it. A totally uniform field has zero effect on anything inside that region of uniformity. This would offer some interesting possibilities, as it would be a great deal cheaper to rig up a Faraday cage of some given internal volume than a totally pressurized dome of equal internal volume. The rest of the air would be breathable, and short exposure to the localized fields outside should be no more hazardous than short exposure to the magnetic fields in an MRI, and simpler forms of life (plants, for example) should be safe enough, allowing you to place the terraforming mechanism outside the inhabitable zone.
There is another option, however. It is radioactive material that keeps the Earth's core as hot as it is. We have plenty of such material we desperately want to get rid of. Provided you could find a method of getting the nuclear waste into the Martial core, there should be more than enough by now to strengthen and stabilize the natural magnetic field to usable levels for longer than we'd ever care about. Getting it there would be a problem and a half - mechanical drilling would be impossible and relying on maintaining a sub-critical mass to achieve the "China Syndrome" long enough to nuclear drill through would at best risk landing you with a supervolcano in your lap. However, I'm going to call that an "implementation detail" and not worry about it.
It would be more correct to say we lack evidence for viable alternatives, assuming the current models used, for which we now lack evidence unless evidence has been lacking on the existence of dark matter. Which may be great for grant checks, but it's lousy science.
Technically speaking, there is no such thing as a "rest mass". In fact, at the most fundamental level (that of energy that contains information), there is no such thing as mass at all. Ignoring that, for a moment, there can be no invariants except from the perspective of an observer who is outside of space/time. Trying to Newtonianize space/time may simplify things, but I'm not convinced it's terribly useful. It seems like it would be as effective as using Plato's model of the solar system, adjusted for Kepler's orbits and relativistic motion, by adding more wheels. If you add enough, it will work. Well, up until the point where you need to split the atom to get the wheels to fit.
There is a value in moving out of primitive Cartesian space/time, as the equations for relativistic properties are horrible in that notation, but this doesn't remove the need for relativistic properties. It simply allows you to codify them in simpler ways. But it's no wonder that the top minds have gone off their rockers, round the bend, and completely up the spout, on mad quests for undetectable, invisible matter, gravity waves that you could no more sense than the ether even if they did exist, and other insanities that contradict the very science from which their existence is supposedly derived, if they are trying to cram 20th/21st century physics into 17th century concepts. What are they planning on doing next week - reinventing the LHC using flint implements?
The problem with scientists trying to "make things easy" is that science is inherently hard. sin(x) != x, except where x=0. The fundamental rule, attributed to a great many scientists - Einstein included, is that you should indeed make things as simple as possible, BUT NO SIMPLER. Those last three words are absolutely critical. If a rest mass has no meaning in the system you are using, then substituting the relativistic mass for the rest mass will break the system. Do Not Want. If that means they need to define macros on their wordprocessors to spell everything out, then by all means do so. If that means we have to live with 12 dimensional silly string, I am far more willing to tolerate that than trying to divide by zero.
Slashdot Mirror
The same applies to hash functions. You have some mechanism for identifying which function you are using and then you use it. Hard-coding is for wimps and fools, and is almost always the true cause of backwards incompatibility. Correctly-engineered software either uses defines or, if it's really good, negotiates. Bad software, and pandering to it, is why Windows has accumulated more crud than a farmyard without the benefit of being able to use it to fertilize.
(Despite conventional wisdom, Windows is not popular because it supports bad software. It's popular because it's pre-installed and has enough good software that people can ignore all the bad crap. If Linux were to insist on supporting all of the bad crap written for it - do YOU use a.out binaries still? - for the indefinite future, it would not improve its popularity. If anything, it would risk the popularity it has gained and its forward momentum, all achieved by willingly sacrificing the bad from the past (with the exception of X11) in favour of developing the good for the future.
If 99% of the risk comes from people with 1% of a functioning brain, it makes no sense to not take simple precautions that might (only might) marginally increase the risks with the remaining 1% for the short timespan it takes for you to plug in a different PAM module and re-enter the passwords.
You are entirely correct in saying that heavy salting will have the same effect as using a different algorithm, at least as far as pre-existing cracking tools are concerned and probably as far as any future cracking tools are concerned, provided there are no major function weaknesses found and the hash lengths are the same.
However, your salting method then becomes just as much of an unknown as the alternative algorithm, and all you do is rename the problem of using an untested algorithm without actually changing it. In fact, a sufficiently complex salting method will transform any hash function into any other hash function, provided they have the same basis and produce the same length of hash.
In the end, given that both proprietary and open source software have been subject to numerous zero-day vulnerabilities over the years, given that both have retroactively removed features, functions and capabilities with both regular and emergency patches, and that both expressly deny any "fitness for use" guarantee, the excuse of "but it might not work" sounds, well, feeble and ever so amusing. It's a "who dares, wins" world and those who do nothing will go nowhere.
Oh, and I would remind you that Torvalds ignored HIS professors when it came to kernel design and opted to write his own code rather than use the publicly-available BSD4.3 or MACH tapes. I wouldn't claim to be his equal, but I would claim to have infinitely more parallels that some gutless 6-digit UIDer who, if they'd been the inventor of Linux, would have based it on TOPS-20 and never ported it to platforms never "tested" by experts for a true multitasking OS. Assuming they ever ported it from the "safe" minicomputer of their choice at all.
According to one expert in such matters, "If you want to be the best, if you want to beat the rest, dedication is what you need." Cautious types need not apply.
If you were to edit that segment of code, you cannot be certain (with today's knowledge) what impact that will have. The retrovirus method of inserting DNA caused a rare form of leukaemia in some people as I recall. It is possible that unfortunate side-effect was due to accidentally altering the metadata or some other aspect of the coding in unexpected ways, or perhaps having the defective (but switched-off) gene still present caused problems as its own metadata would also still be present but would likely still have been active.
Therefore, altering DNA to remove the gene for one form of cancer (which can be treated) may increase your chances of getting a much less treatable cancer, or one that can't be treated at all. It might have all kinds of other nasty effects that we've yet to discover. Genetic science is still very much in its infancy. Bear in mind that if you start counting from when DNA was first discovered, about as much time has passed as had when the Apple IIe and Commodore PET were the latest in home computing. If you start counting from when the human genome was first transcribed, we're closer to the era of Colossus and ENIAC.
That's not an unfair comparison - look at deCODEme's lists of identifiable genetic diseases and genetic traits. Not very long, are they? And they don't require much in the way of hard evidence to add something to those lists, either - two studies will do it. We can assume anything not on those lists has evidence too weak even to pass such minimal muster, which is just about everything, and they don't even get into such things as DNA metadata and other coding abnormalities. That isn't too distant from the best knowledge the top-of-their-field experts had in computer science at Bletchley Park and certainly sounds very much like the state of homebrew computing in the late 70s - bits and pieces of wisdom with an awareness of lots of blank spots.
(Science moves a lot faster these days, but genetics is a lot more complex, so it all balances out. Twenty, maybe thirty, years down the road, genetic editing might be as safe and reliable as Open Source is today. Mind you, Firefox keeps crashing on me and I can never keep Bygfoot going past three seasons. Hmmm. Better make that forty from now, then, to be on the safe side.)
Using Linux as an example, most of the "mega patchsets" could be considered forks from the vanilla kernel but end up testing the component patches extensively, allowing for a better choice of what goes mainstream. In other words, the "forks" have accelerated Linux kernel development. The distro-specific patches - well, not too sure about those, but more than a few have also made it mainstream - again likely for the same reason. Better testing by more people. (This is not to fault the Linux development model, quite the opposite. It's praise for the number of imaginative ways that patches have been brought to people's attention and praise for the handling of the logistics of so many patchsets - pseudo-forks, I guess you could call them - that the Linux kernel developers manage superbly. Sun would be wise to look to these people for inspiration on how to combine the best from the forks and how to know what the best even is.)
Ignoring the attempt at humour for a moment, I wouldn't agree with the statement. I would, however, argue that neither action nor punishment fail to deter and in some cases the combination of action and punishment have created an incentive. This is true of copyright, illicit substances and probably a whole string of other offenses. This doesn't mean they should all automatically be legalized. The opposite of authoritarianism isn't anarchism. What it does mean is that such offenses should be treated very differently and much less punitively. In some cases, yes, abolishing the offense probably would make sense, but it's not wise to assume that is automatically the best choice. If Prohibition was America's biggest mistake, abolishing Prohibition was the next biggest. You don't fix one extreme by going to the other. You certainly don't fix it by trying to pretend it never happened.
England didn't ban heroin until the 50s (and only under heavy pressure from the US). There is little evidence it was seriously abused before then. But if they removed the ban overnight, it could be a total disaster. The shift in perceptions, the change in attitudes and the secrecy of the underground subcultures that abuse such stuff, not to mention the big pockets and bigger firearms of those wanting to protect their profits, would make an instant shift an instant disaster. If you're in sixth gear, flat out, you do NOT put the car into reverse. This does not make reverse gear a bad thing, it just means you need to take the intermediate steps first. The state of the system is utterly wrong for what you're attempting.
The problem with the drugs culture is that we don't really know what the right state is. The level of neurological research on how drugs affect the brain is minimal and knowledge of the effect on the rest of the body is virtually non-existent. Sure, this could be fixed. All you need is a battery of PET, MRI and fMRI scanners, drugs containing radioactive tracer isotopes and a bunch of volunteers stupid enough to have their brain glowing with positrons for dangerously long periods of time. You'll soon find where the uptake is, the effect the byproducts have, how the brain structure changes and how brain activity is altered. Because the changes and any build-up are gradual over an uncertain length of time, you're likely to kill a large percentage of the volunteers with brain tumours from continually pumping radioactive material into soft tissue, hence the level of stupidity required. Without the neurological data, though, it is quite impossible to form a scientifically sane policy. At the moment, there's a lot of superstition and religious nonsense by both sides, but there simply isn't any science worthy of the name. Without that, how can you know when and how to liberalize?
Consider this. A multi-core processor will typically have one cache which is shared between cores. SMP systems, on the other hand, have one cache per CPU. For SIMD or MISD problems, a single cache is not a big penalty. For MIMD problems, it will kill you stone dead. Most "interesting" problems (the problems that eat Crays for breakfast, and enjoy a leisurely lunch of Blue Genes sautee'd with Information Silos in a brisk red hat) are MIMD. It's these problems that have people talking in hypercubes, Processor-in-Memory, MPI-enabled RAM, RDMA, and any other trick they can possibly come up with to shunt data around faster. When these guys can (and do) play with 60 gigabits per second bandwidth in both directions, they need CPU technology that won't take one look at that kind of hardware abuse and run screaming through the night.
Which is why the "maintainable" is such an important element. It's also why the full RASDIT cycle requires that on completion, you go back to the requirements gathering. The theory is that even though this method is slow, the software will converge on the users' needs as the amount of fresh code required will drop with each full cycle. It's entirely possible this is the methodology used in mission-critical systems, such as power stations, as you don't want to patch often and the requirements will change very slowly.
As I understand it, the waterfall method is not a full repeat of the entire cycle, but a partial implementation, partial test, followed by another partial implementation, partial test, cascading down through a whole series of these. There are a lot of very very short cycles in this method, at least as I was taught it.
RASDIT doesn't cascade down like that. Implementation is bottom-up, testing is top-down, they are very distinct phases that don't interact beyond the transfer of the code. There are no cycles and one phase does not start until the prior phase has completely stopped. I've seen Slinkies used to describe this methodology, but never waterfalls.
Now, I'll freely admit that the terminology I learned is (a) old, and (b) quite possibly localized, so we can probably account for 99% of the disagreement by means of such external considerations. If what you're describing is the current terminology, then that is the terminology I should be using. Nuff said. Ergo, I shall do so and thank you for bringing it to my attention.
The old slow method (RASDIT - Requirements, Analysis, Specification, Design, Implementation, Testing) is hardly the waterfall method. Waterfalling is a relatively rapid cycle between implementing and testing in small stages, each stage feeding into the next, as a cascading series of waterfalls. The RASDIT software life-cycle states that after testing comes deployment, after deployment you go right back to the start and re-assess requirements. There is no direct feed between one cycle and the next (you re-use from the last life-cycle, but that was the last life of the product not this one) and start-to-finish can take years to decades. If the waterfall method evokes thoughts of a gentle series of cascading waterfalls, step-by-step from start to finish, the RASDIT approach is a thousand-foot plunge. You're damn sure what you have is solid, or you're landscape. RASDIT is not used much these days, but that is the only approach that can be called "old" or "slow". Molasses is fast in comparison. (Tastier, too.)
RASDIT is the rite of passage of all serious software engineers. It is possible to write software this way and you will have a very high probability of producing robust, reliable, efficient code. It probably won't be maintainable, but it will very likely work and work damn well. All methods other than developing the code to the tests will produce inferior results, but they will produce inferior results far faster and in a form that is easier to debug.
I think the difference is that a lot of open source projects operate on a release-early-release-often philosophy - what software engineers refer to as the waterfall model. In such a scenario, bugs will appear in earlier releases and decline in number sharply over time. Closed source projects tend to operate on the more classical release schedule, which tends to be a lot slower with a lot more expansive software lifecycle, but SHOULD produce far far fewer bugs in each release.
MySQL, with the current Oracle owners, seems to have moved to more of the classical mindset. That's fine and I personally don't think there's anything wrong with it if it's done right. But to be done right, you need fewer bugs in a release than the rapid-cycling of the RERO approach, and should aim for code of a quality comparable to RERO over the same timeframe.
It is clear in this case that MySQL has failed the test. It is a good product, but the new methodology is fixing less than the old methodology. That needs to be when the developers stop and think about whether they want to try switching to something else or go back to what had worked well in the past.
You're right. The Rolling Stones painted it all black.
Seriously, yes, many scientists leave out data that doesn't suit their theories. It's still bad science and they're still bad scientists. The number of good scientists is extremely small. I don't see the scientific world in shades of grey, simply because either it's a valid, falsifiable, theory or it isn't; either the data was collected and presented or it wasn't. There isn't any room for shades of grey in science. Data selection is why Copernicus failed to see what Kepler observed - orbits aren't circles. This put science back centuries because of religious and metaphysical bigotry. Sorry, that ain't science, that's stupidity and as close to evil as science gets. I have no respect for such corruption. It's claimed Pythagoras had people murdered in order to keep mathematical proofs that contradicted his cult teachings from leaking out.
If you add up all the bigotry and corruption over the millennia, you start talking serious timeframes. I doubt medieval Europe could have put a man on the moon (see: The Tomorrow People, Rift in Time) but I could easily see astronomy, some areas of physics, mathematics and possibly chemistry being many hundreds of years further advanced than they are today, had people chosen to put scientific integrity over and above all else. Beyond Star Trek, but achieved yesterday, in those fields that are less sensitive to things that were probably not going to happen much sooner, like microtechnology and nanotechnology. To me, the incalculable damage done to humanity through such actions is unacceptable and beyond any classification of "greyness". There is no moral relativism in whether or not the square root of two is a fraction or an irrational, there is only a fact of whether it is or it isn't. It's extremely quantized.
(Relativism is valid in essentially every aspect of culture, philosophy, politics, and so on, but 0 != 1 no matter what. It's not dependent on the observer.)
This is clearly a new branch of Homeland Security for Cthulhu. Tired of people carrying around copies of the Necronomicon to dispell him, he has invested in an army of multi-jointed drifters to act as a lynch mob should any deep sea fish try to exchange knowledge of his whereabouts for a reduction in EU tuna quota.
The data doesn't care, but oil drillers are unlikely to give information that harms their potential to drill, and can afford to be "selective" on what they provide. They also have more than enough technical equipment and expertise to "improve" the data, if it is in their interests to do so. That is why it is generally bad science to get information from those who have a vested interest in your conclusions being what they want them to be. It has nothing to do with the camera and everything to do with the eyes far behind it.
You're right. It should be gosub for recursion. My bad.
Depends on the level you're talking about and the interface used. I'd consider many MMORGs, MUDs and "interactive fiction" games to make it extremely hard to differentiate between the character and the player. I'd consider games like Elite, Virus 2000 and TORCS to likewise blur the boundaries substantially. If you add in the current work on CAVE systems, neural interfaces and other such gizmos, you can definitely see the possibilities of a total collapse between virtual environments and reality. Mind you, I'm not so sure I'd be opposed to that. You'll notice that most (if not all) of the types of game, or examples of games, I gave have a very dedicated core "cult" following bordering on addiction (although apparently not everyone agrees, according to other stories on the front page). The "addictive" element, in my opinion, is exactly what is being talked about - the total collapse of the fourth wall.
Now, on the flip-side, if you look at games that are perhaps "popular" for a while but aren't really considered that addictive and are soon forgotten when something else comes along, you will see that they rigidly differentiate between player and character. It is my belief that it is this element that makes games exciting for people, that if you actually try to keep players at a distance from their characters, you will destroy the magic of the game entirely.
Is this just true of computer games? I'll argue no. RPGs that keep the player at a distance die off. Those that immerse the player survive and thrive. Is it only true of games? I'll say no to that as well. Books, movies and songs that draw the audience into the "sub-creation" (as Tolkien called it) are longer-lasting that those that don't.
Indeed, Tolkien's call for MORE sub-creation and GREATER destruction of the "fourth wall" is IMHO a better guide for how popular culture should develop. We should not fear our own creations, only our fear of them.
1. Recursive logic works because of (goto 1)
If myspace or facebook started asking for hair clippings from subscribers, the DNA database would choke.
Anyone treated in hospital is likely to fill out a form asking if their DNA can be used for medical research. Anyone who has used a genealogical DNA service (and there are many) or one of the genetic disease detection services (there are rather fewer of those, deCODEme being one of the better-known) has their DNA on file on the service's database. Now, whilst there is some protection from unscrupulous departments abusing these kinds of socially-provided services, "law enforcement" gets a LOT of leeway when it comes to bending/breaking the rules. Also, bear in mind that the FBI has been caught using anti-terror legislation to cart off all of the computers in libraries and other communal services. All you need is one "potential suspect" on one of these massive databases and suddenly the powers-that-be have authorization to confiscate/duplicate the entire system, protections be damned. This is not the fault of such services, and they can have all the good intentions they like when it comes to privacy protection, but none of your records have been safe for many many years now. If there is a "fault" in this, it is the fault of those Americans who encouraged (through votes, money or even simply not telling their senators and representatives they were being stupid - inaction is just as bad as incorrect action) Congress to pass such paranoid legislation.
The British have been just as foolish, trusting (despite all evidence to the contrary) that it was safe to provide Government with such sweeping powers. Iceland has no doubt passed on it's thanks. I'm not a proponent of small Government by any means. Size doesn't matter, it's what you do with it that counts. And what's been done is the encouraging of the paranoid delusions of a few, to the detriment of many.
...Dexter got hauled off by Major Glory to see his lawyers for IP infringement.
Ok, that does it. Gotten is legit American English? I'm going to patent the English language and sue Websters.
But... but... I want things to just work when I tweak things! And I want the latest of everything! And I want all the tools that I've bookmarked on Freshmeat that the distros don't carry! And... and... and I want a pony!
People are exposed to 2.5 tesla magnetic fields in conventional MRIs, and there are MRIs that go up to 9.3 tesla (even though that's a bit beyond what the FDA approves). There are scanners for animals that go up to 12 tesla. Even the least of these is well beyond anything the Earth generates. Clearly, however, magnetic fields are not as dangerous as has been assumed. At least, within reasonable limits.
This is a localized magnetic field, however. If you were to set up a Faraday cage, so that you were enclosed in a uniform magnetic field, you shouldn't ever detect it. A totally uniform field has zero effect on anything inside that region of uniformity. This would offer some interesting possibilities, as it would be a great deal cheaper to rig up a Faraday cage of some given internal volume than a totally pressurized dome of equal internal volume. The rest of the air would be breathable, and short exposure to the localized fields outside should be no more hazardous than short exposure to the magnetic fields in an MRI, and simpler forms of life (plants, for example) should be safe enough, allowing you to place the terraforming mechanism outside the inhabitable zone.
There is another option, however. It is radioactive material that keeps the Earth's core as hot as it is. We have plenty of such material we desperately want to get rid of. Provided you could find a method of getting the nuclear waste into the Martial core, there should be more than enough by now to strengthen and stabilize the natural magnetic field to usable levels for longer than we'd ever care about. Getting it there would be a problem and a half - mechanical drilling would be impossible and relying on maintaining a sub-critical mass to achieve the "China Syndrome" long enough to nuclear drill through would at best risk landing you with a supervolcano in your lap. However, I'm going to call that an "implementation detail" and not worry about it.
It would be more correct to say we lack evidence for viable alternatives, assuming the current models used, for which we now lack evidence unless evidence has been lacking on the existence of dark matter. Which may be great for grant checks, but it's lousy science.
Technically speaking, there is no such thing as a "rest mass". In fact, at the most fundamental level (that of energy that contains information), there is no such thing as mass at all. Ignoring that, for a moment, there can be no invariants except from the perspective of an observer who is outside of space/time. Trying to Newtonianize space/time may simplify things, but I'm not convinced it's terribly useful. It seems like it would be as effective as using Plato's model of the solar system, adjusted for Kepler's orbits and relativistic motion, by adding more wheels. If you add enough, it will work. Well, up until the point where you need to split the atom to get the wheels to fit.
There is a value in moving out of primitive Cartesian space/time, as the equations for relativistic properties are horrible in that notation, but this doesn't remove the need for relativistic properties. It simply allows you to codify them in simpler ways. But it's no wonder that the top minds have gone off their rockers, round the bend, and completely up the spout, on mad quests for undetectable, invisible matter, gravity waves that you could no more sense than the ether even if they did exist, and other insanities that contradict the very science from which their existence is supposedly derived, if they are trying to cram 20th/21st century physics into 17th century concepts. What are they planning on doing next week - reinventing the LHC using flint implements?
The problem with scientists trying to "make things easy" is that science is inherently hard. sin(x) != x, except where x=0. The fundamental rule, attributed to a great many scientists - Einstein included, is that you should indeed make things as simple as possible, BUT NO SIMPLER. Those last three words are absolutely critical. If a rest mass has no meaning in the system you are using, then substituting the relativistic mass for the rest mass will break the system. Do Not Want. If that means they need to define macros on their wordprocessors to spell everything out, then by all means do so. If that means we have to live with 12 dimensional silly string, I am far more willing to tolerate that than trying to divide by zero.