I respectfully disagree with the notion that dynamically generated sound is impossible to do with today's hardware. Yes, perfectly raytracing sound would be computationally infeasible, but guess what...... that's exactly what the grandparent was asking for.
Please re-read the grandparent post. He called for simulation from scratch, not prepared sounds mixed together.
I'll actually raise you one on the rest of your (misaimed) message... we should already have had dynamic sounds for about five years at least. At the very least, they could fucking shift the pitch simply by modifying the playback speed a touch every once in a while, so that every single scream from the dying demon in Doom doesn't sound exactly the same every single time. If I had a job in the industry writing sound engines that's one feature I'd sneak in even if it weren't called for.
physical modeling is just beginning to come of its own
Grandparent was not talking about "physical modelling". "Physical modelling" isn't. "Physical modelling" is when you create a filter that represents a part of an instrument, and string them together. While it has tolerably good results, and it quite fun for musicians to play with, it has no ability to simulate arbitrary objects doing arbitrary things; the process of creating the model is long and tedious, and what you can do with it is pretty sharply constrained.
Please re-read the grandparent of this post more carefully; what he's asking for bears the same resemblence to "physical modelling" that a water rocket packed in with your box of Trix bears to the Space Shuttle.
We already have Toy Story-level graphics and we've had them for years... Sony promised us Toy Story-level graphics in real time years ago, and by golly, they released the PlayStation 2 and behold, we did!
Which is to say... been here, heard this. Don't believe it till you see it, and are playing it. And a gentle reminder to the Sony fanboys to take the claims Sony makes about the PS/3 with a grain of salt; they played you for chumps last time, with the worst visual quality of the current generation. (There are many times where even the DreamCast beats the PS/2, because of Sony's poor decisions w.r.t. anti-aliasing.) Don't let them do it again.
Re:Where are the "Sound Acceleration" cards?
on
Cinematic Game Graphics
·
· Score: 5, Interesting
But, where are the cards that can generate the sound of one arbitrary object hitting another? I don't just mean positional sound of pre-recorded samples, but really create the sounds from scratch (or an "audio-enabled model").
Not to put too fine a point on it, but... we don't know how to do that.
Even what you may have heard of is wild, wild cheating compared to what you describe.
I've given some (intelligent, educated) thought to this issue, and here's the problem. Light is adequately simulated with a line in the macroscopic world. It is, technically, a wave, but it is so high frequency that we can ignore it unless we're trying to simulate at quantum scales. This keeps the complexity down to merely polynomial, and some smart people have figured out how to shrink that polynomial down to a surprising degree.
Sound is not so friendly. Imagine some reasonably complex object, like, oh, a chair. We'll even cheat and make it all out of one material and go ahead and make it as geometrically simple as you like, as long as it's a chair, which is to say, at least three legs and a sitting surface, preferably with some sort of back support. Now, mentally give the chair a tap.
Now, odds are you're not too practiced at this sort of mental visualization, but here's what happens. For the sake of argument, let's tap it on the direct middle top of the sitting surface. Let's cheat some more and assume that one impact is what makes the sound, rather then an oscillation at the tap point. (See how much we're cheating, and we'll still end up with an uncomputable scenario.) So the tap radiates outward from there and starts wiggling the legs. The sound partially bounces off the legs, and goes back, and some of the sound wiggles into the legs. The sound bounces all around in the leg, and every time it bounces off of the edge, it loses some of the sound to the atmosphere and some of it bounces back. By the time it hits the bottom, it's bounced several times. (See, sound can't turn except at one of those boundaries, where it is essentially absorbed and re-emitted.)
Meanwhile, the same thing is happening in the back of the chair. Plus, on the sitting surface, we have reflections of reflections of reflections to deal with, and thanks to the wonders of resonance, we absolutely have to track each and every one of them until they hit a really low level. In a fraction of a second, we have hundreds upon hundreds of seperate waves to track, and they aren't even rays, they are "wavefronts".... imagine a wavefront hitting the edge of the chair at an angle, like an ocean wave bouncing off the beach. It doesn't bounce like a particle, it is two entirely new waves, the one that reflects and the one that continues on.
Basically, we can't even simulate this horrid simplification, the real world is even worse. Sound is highly, highly parallel. Ultimately, sound simulation is firmly exponential and the constants are very, very high. Maybe if those magical quantum computers come online we'll get this, but we'll quite possibly never get it with conventional technology; we're always going to have to cheat.
(One can try to imagine a transformation of the chair where the sound travels in a straight line in some space, but I'll be damned if I know what that actually looks like in real code, nor am I sure that it would be any easier to compute then a straight-forward simulation anyhow. Bright ideas in this regard should probably not be posted on Slashdot and saved for your PhD thesis in Mathematics/Physics/Computer Science; they'll all be waiting with bated breath.)
I usually just deal with it by dedicating an entire 36 gig disk to swap in servers and use a much smaller swap partition on workstations.
You need to ask yourself, is there anything that server does that I'm willing to wait for thirty-six gigabytes of disk space to fill up, rather then just have it throw an "out of memory error" immediately?
I don't care how fast your drive is, 36GB takes a long time to fill up.
Lately, with all the RAM we have, I've been seriously toying with a "no-swap" policy. I'm not quite ready to take the plunge at 512MB of ram (dip into swap about once a month on average; gentoo emerge + mozilla + Open Office + a few other things and you do get there eventually), but once I have a gig I think I'm going to stop swapping.
Why? Because I can not think of a circumstance where I'd rather see the machine bog down for potentially hours, trying to use even a gig of swap, rather then just immediately throw an "out-of-memory" error. Nowadays, if the computer is reaching for that much swap it's most likely an errant task in an endless alloc loop anyhow!
Now, there are exceptions, but I think they are even rarer then most people think. For instance, if you're running a database server, you might need the swap... but again, are there any circumstances where you absolutely need that query to run... but don't mind waiting literally millions of times longer for it to run, probably causing a cascade failure anyhow (as lack of memory impacts more and more queries until the system can't catch up)?
It's not a popular opinion yet, but I think the era of "swap" is coming to an end, obsoleted by hard drives that are, relatively speaking, thousands of times slower then they used to be (relative to the processor times), and memories thousands of times bigger. The set of situations swap makes sense in is shrinking rapidly, and will soon be gone.
Eh? With modern computers, I can attain a frag rate of millions per second with Conway's Life running at full tilt. You must play a mean game of Quake.
Enterprise users [are lazy people with no time to do anything with their computers]
What crack are you smoking? It's the enterprise users who have stuff that's so mission critical they buy Windows source code from Microsoft and do stuff with it. You know, because tons of money is at stake.
It's the poor schmoes with three computers and a network hub that just want to plug things in and make it work, because one person-week wasted is a significant percentage of the company's time.
You obvious have no clue. If the enterprise users can score any kind of 10% improvement enterprise-wide with merely a few thousand man-hours invested, that's a good deal.
Any time you argue against the feasibility of nano-factories, you need to make sure your arguments do not equally argue against the possibility of life in our universe.
I don't think your post meets that bar. We know self-replicating factories exist. We are one. The question is, what improvements can we make and how well can we control and miniaturize them, not whether they are possible.
Immune systems and viruses have co-evolved. If a nano-bot develops that is made of a metal skin, that pretty much bypasses anything the immune system can throw at it, even if it's almost biological inside the metal skin, since most (all?) of the immune system keys off of proteins on the surface of cells.
Immune systems work because viruses have an evolutionary barrier to get to anywhere the immune systems won't work (i.e., a "half metal" virus can't mutate into being; such a thing may be possible but the gulf to get there is too wide; evolution is powerful but kooky and definately not omnipotent, it does have limits and in many ways, people overestimate as much as they underestimate). Nanotech will have no such restrictions. A self-replicating plague of some kind would still be limited by what kind of elements we have in our bodies, but there's enough iron and a few other metals to make enough nano-bots to kill us... and the nanobots have all day, metaphorically speaking, because the immune system will never even see them, let alone attack them, so they can kill cells at their leisure.
Not to mention the biological judo a deliberately designed killer could apply, recruiting the body's own immune system to help.
Cute, but I think your program may have a race condition in the ID assignment; I've tried several times now and I think I'm getting other people's information.
I think you're doing a "SELECT MAX(id) FROM database" when writing the link out onto the screen but you can't do that; other records are being inserted before the link is written. But that is just a guess.
I was able to hack the URL to find the one I want to send to a friend, though, so thanks from me.
What are you talking about, not everyone is supposed to get 100%. Everyone is supposed to TRY and get 100%, but the test is made hard enougth that the mean is a B or a C. That is pretty standard for all education levels. There are always a few teachers who make their tests real easy, but generally it isn't a big problem
No, that's the theory, a theory which is now being paid less and less lip service, let alone actually being implemented. Even so, if the system doesn't have built-in retries with the expectation that every student does better then last time (meaning there is a wide range of questions), it's still expecting 100% performance first try and penalizing anything else. (And re-learning the same thing year after year isn't the same; the retries need to be in a fast feedback loop, and you can't get much slower then "yearly".) And the system is largely "Learn A, test A, ignore results of test", "Learn B, test B, ignore results of test", with a final "average" at the end of the year.
One could hardly imagine a more learning-hostile environment. For all the efforts educators claim to put into developmental psychology, and studying how people learn, they seem remarkably impervious to the results of such things. They seem incapable of learning themselves.
"Global Warming" is a debate, albeit a lop-sided one. The causes and ratios of climate change is a fierce debate. (Have humans had an effect? Oh, sure. But are they anywhere near 100% responsible? Now thats a much more inflammatory question. Personally, while I know humans aren't 0% responsible, people trying to put the number in, oh, say, the high 90%'s or even 100% I find much less compelling then those with lower ratios.)
The inevitability of some change is not a subjuct of debate... except among some environmentalists who seem to think the status quo is the only good, and the only reason it's not being maintained is human action, and that if we only did the right things, somehow magically stasis would occur.
I've never understood the obsession with "everybody must understand everything". By the time you're dumbing down the content for the lowest common denominator, you've got nearly nothing of substance beyond:
The magnets attract each other because of magic. We label this magic "magnetism". It's really complicated, but there are wizards who understand it. The magic is said to involve "poles", like the "north pole" and the "south pole", which is related to the north and south pole of the Earth in ways you can't understand. You now know nothing really about magnetism, but you can now sling around the labels "north magnetic pole" and "south magnetic pole" and sound like you understand something, just like the engineers on Star Trek! Speaking of which, here's a few pictures from Star Trek.
Now, I understand and totally agree that people can't jump from ignorance to Maxwell's equations, nor should they have to. And there's good reason to believe that Maxwell's equations are totally beyond most children (see developmental psychology; the cognitive skills necessary to understand calculus typically do not develop until the kid hits double-digits in the age).
On the other hand, why must the whole exhibit be geared at the introductory level? A museum is a big place. Surely at least a little bit of room could be spared for some more sophisticated information in parallel with the simplified stuff? 10-year-old and Dad ought to be able to learn something.
(I have a similar criticism of the educational system. Why should we expect every child to 100% master the same math? Instead, set a baseline, and include varying levels of math in the same lessons. Especially as you get into Algebra and beyond, it's increasingly easy to challenge your students while making sure everyone understands the baseline, even in the exact same classroom. The myth that every student should perform 100% on every assignment is one of the worst blocks to educational reform today. We should expect children to get things wrong... because next time they try, they'll do better, and next time, they'll do better, and next time, they'll do better, etc.... and those children end up way ahead of the ones confined to just what they can do ~100% the first time... and as we've seen, 100% perfection has a habit of receding over time, instead of advancing as we need.
It's all the same fallacy, playing out over and over again, museums, schools, college, television shows, everywhere.)
Well, linear time == linear time. But I know what you mean.
The problem is that "Quantum Computing" is, in most people's minds, where Nuclear Power was fifty years ago. The fact is that if you throw enough bits at even conventional encryption, as I understand it, you still can't build even a theoretical QC that can crack the encryption, because I can always push the real solution below the noise limit.
QC isn't magical. Assuming it works at all, which I think there is stronger reason to doubt then believe*, it still has limits. QC works by running the answers all at once and then filtering the resulting wave for the answer. ("Filter" here has technical meaning.) If you run too large of a problem, you can't "amplify" the correct answer with an acceptably high probability, you're just as likely to get a wrong one. We may need all 20480 bits, or 204,800 bits, but eventually you're going to need a machine that is too large to actually be built to crack it. (Because of the way they works, qubit-based machines have rather firm upper limits to the complexity of a problem any given machine can compute, so even if a theoretical machine could crack our best encryption, it will probably still be the case that no real machine can do it.)
*: Large qubit machines, that is. I am quite aware small ones have been built and there has been progress in the field but I am of the opinion that we have already hit the point of diminishing returns. A qubit-machine requires too much isolation from the real world to ever be practical in this Universe.
Now, if you're like a PhD student or something in quantum computing, I welcome correction. But if you're just waving qubits around like a magic wand, please put the wand down and stop worrying that encryption is going to stop working. Besides, qubit-based machines only get us into exponential land, and there are an infinite number of super-exponential functions (using computer science lingo freely). I would be absolutely stunned if quantum computing didn't have a corresponding encryption algorithm it can feasibly implement that quantum computing must then brute force in its own way. That's just the way the complexity domains work; an infinite number of them, and my intution strongly suspects that the encryptors will always have a simpler problem then the decryptors.
The Discovery channel did one of its credulous stories on The Ark, and, well, it was better then the family reunion I was attending and I wasn't in a position to change the channel, so...
The CIA has bona fide pictures of... steel yourself... something. It looks like it might be a boat. (Personally, I think it looks too much like a boat; if you go looking for a thousand-year-old, unmaintained building and snap a picture from orbit of something that looks like a building, you've eliminated that site, because it should look like rubble, not a building.) A CIA photography analyst took bona-fide interest in it, but I'd lay money that since they are geeks, they take interest in a lot of things. (It's kind of their job, and part of that job is to shrink the class of "things we can't identify".)
The CIA legitimately classified it, and while the Discovery channel made hay out of that, my guess is that given the time frame, the image was classified, along with all other satellite images, on the (correct) grounds that even if you give someone a harmless image, the image itself will tell a lot about the capabilities and resolution of the satellite. Thus, the classification itself isn't really a mystery, only the people who think it's a mystery are a mystery.
I'm a theology-reading, bona-fide, for-real Christian... but I still classify the CIA evidence as another "Cydonia"-type picture, since I still apply the skeptical eye to things like this. (I respect the image analysts of the CIA a lot, it is a highly skilled job, but when faced with a random photo of a mountainside, the fact that they work for the CIA doesn't mean anything except to conspiracy theorists; the CIA, in this situation, is most likely just the group that happened to be holding the camera.)
The parent is modded as funny but it's true. As long as the errors are evenly distributed (i.e., a fundamental effect, vs. a huge-ass scratch across the platter which isn't evenly distributed at all), you can throw enough error correction at the medium to make the unreliability go away.
This isn't theoretical at all; CDs are routinely "destroyed", but they have a lot of error correction built into them so you don't even notice.
Computing the exact probabilities left as an exercise to the reader... but given any level of reliability there is some error correction scheme that can bring it up to any other given level of reliability (short of perfect, of course). Of course you can construct pathological cases that need as many bits as you like, the equations work that way too.
We do, say, 2048-bit encryption (asymmetric), because it would be "too slow" to do 20480-bit encryption. "Too slow" here is a fuzzy term, but generally speaking, if you're sending an encrypted email you don't want to hit "send" and have it delayed for three weeks while it gets encrypted. There's no real reason we couldn't do it today.
As computers speed up, both encryption and decryption get faster. However, while adding another 128 bits to 128-bit symmetric cipher may be "free" with newer computers (and eventually will be), the 2^128 multiplicitave increase to the space the decrypters have to search is not free. To increase encryption power, the encryptors merely double their work. (To an approximation; I don't think the work load is strictly linear but it's a lot closer to that then exponential, and that's all that matters.) Meanwhile, for that relatively modest investment in encryption power, the decrypter's jobs got 340,282,366,920,938,463,463,374,607,431,768,211,45 6 times harder.
This is why, in the relatively near future, we'll all have encryption that is effectively "unbreakable", because no conceivable decrypter could be built that could do the calculations to crack the encryptions, even with the raw materials of the entire Universe.
Practically speaking, most of us already have damn-near unbreakable encryption today; if you're connecting to a computer with SSH, SSH is most likely the strongest link in your security chain by far; the weak links are the computers on each end of the link, the humans on each end of the link, and possibly the facilities the computers are in. Nobody is going to tap your ssh stream and get any value from the massive decryption effort that would be necessary unless you're trading secrets worth billions.
Specialized hardware can only gain you a linear speed up, at best, and those calculations for "minimal computer" to crack a given encryption key are not extrapolated from modern computers, they are extrapolated from the maximum computation possible to do, given a finite energy supply. (QM-based computation advocates may wait until they have a large-scale (multi-thousand-qubit) machine to jump in here.)
Along with jsac's comment (more processor power exponentially benefits encryptors, only linearly benefits crackers, on the whole more power means a win for encryptors), I'd like point out this is only a set-back for encyption in-as-much as encryptors claim that their encryption will keep your data safe for all time. Which is to say, at least for the reputable encryptors, this isn't a set back at all.
If you insist on putting words in their mouth, then yeah, you might consider it a set back. But that's your misunderstanding, not theirs. All reputable encryptors have accounted for Moore's Law in their cost/benefits tradeoffs. Since it doesn't take much encryption power before it requires computers larger then the Universe to crack it via brute force (and since "cracks" on good encryption are really typically just ways of collapsing the search space, not procedures that give immediate answers, often adding more bits will require Universe sized machines, too), this isn't that big a deal for encryption. Push your key size up and be done with it. Even conventional machines can handle that today, it just takes longer.
every University and College *I know of* has an incredibly grueling music theory degree, and after taking a simple piano appreciation class, this CS student knows better than to take any more music courses regarding song analysis!
Hmmm. Speaking as someone who got a Masters in Comp. Sci., I found music analysis to be almost trivial, certainly I found it much easier then my fellow musicians. In particular, I was very easily able to straddle the line between "the rules" and "the feel".
(For those who have never done it, music analysis is interesting and useful for composers and players, but there is a strong element of "post hoc" analysis to it; analysis is really more interested in exploring the effect music has and sometimes a given theory will say X is happening when a quick and critical listening will say Y is happening. In the complementary direction, you'll see musicians use things like double-flats because even though E-double-flat is "literally" D (tonal pedants need not apply), in the theoretical context it makes more sense as an E, doubly-flatted. This is almost isomorphic to the relationship between software engineering theory and software engineering, complete with the "theory uber alles!" contingent and the "who the hell needs theory?!?" contingent. I'm one of the few people in the early theory classes who correctly used a double-flat on a test.)
(Then again, to be fair, I'm one of the rare comp. sci. types able to navigate theory and practice easily, so I'm probably an odd bird anyhow.)
What would be interesting would be to qualify various music genres along various lines and see which qualifications match their observations. Mozart was a genius and his music is like no other, but I can't think of any single attribute he holds a monopoly on (though the combination is unique, IMHO). A lot of "concept albums" share some of the off-beat regularity and interest in themes, for instance, though they obviously differ from Mozart in more ways then they are in common.
I'm looking for something that would cover such topics as basic desktop do's and don'ts, like 'do choose a non-dictionary password' and 'don't blindly drop to root and install an unverified/unauthenticated RPM that you receive via email,' etc. Anyone seen a guide like this?
Why?
Do you expect anyone to actually read this document?
Oh, I wish I were being sarcastic.
Either enforce things (your password policy), or wait for people to have trouble so you know what to document (every installation is unique, and you're wasting time trying to predict how your users will react when you could just wait and see).
The only purpose of such a document, in the end, is CYA anyhow. And again, I wish I were being sarcastic. If you can't enforce it, people are going to do it.
The only possible exception is if this is a technical group of users who will be daily and strongly held accountable for violations. Basically, the only group of people who meet these two criteria are Computer Science (or related disciplines) students.
In order for the TiVo to know what is on, it needs to dial-up and download the lineup. TiVo has streamlined some stuff by broadcasting some "infomercials", but you still have to have that dial-up to get the lineup.
Of course, maintaining the system, keeping it up to date, lining up the contracts for the modem banks, etc. isn't free, and it is also TiVo's primary money source.
One could make a case that a pre-existing Internet connection could be used, but that's too complicated for most people.
You can buy a TiVo and not use their provided service, but the majority of the good stuff, which involves understanding the TV lineup, won't work.
I would love to see a movie devoted to the great star trek time warp.
You mean, other then Star Trek 4, Star Trek: Generations, and Star Trek: First Contact? And pretty much the entire run of Voyager?
"Time Warping" is the worst problem with Star Trek. You can't build drama because nothing is ever at stake that won't be wiped away by next week's time travel episode or movie. That problem alone is enough to sink the series; the entire concept has devolved into some of the worst "collaborative writing" I've ever seen. (Ever done a "collaborative novel" online? The resulting continuity trainwreck bears an uncomfortable resemblence to Star Trek now...)
No, his point is that one of the favored FTL communication methods of the "I have a middle-school education in Quantum Mechanics" contingent doesn't work, ye olde "light pulse through a material speeding up the speed of light". The light travels through ever so slightly faster then the speed of light, but he still couldn't push information through any faster.
You missed the point, you're talking about quantum cryptography, which is quite a different thing.
No, you missed the point. Quantum cryptography takes advantage of the same principles, and you'll note that nobody is using it to communicate faster then light either.
You can not force the polarization of light. There is no way whatsoever to do it. Without that, your scheme doesn't work.
Look, if it were that easy, we'd have done it already. It's not like we don't have the equipment or the ability. All it takes is a bit of money to set your system up. All you psuedo-physicists who think that FTL communication is so simple, go set this system up and make a few billion selling it. Or maybe, just maybe, it's not as easy as your oh-so-thin "understanding" of quantum physics makes it out to be?
Put up or shut up. It's not like we discussing something that would be several trillion dollars just to build a prototype, it's down in the merely several thousand.
Slashdot Mirror
I respectfully disagree with the notion that dynamically generated sound is impossible to do with today's hardware. Yes, perfectly raytracing sound would be computationally infeasible, but guess what... ... that's exactly what the grandparent was asking for.
Please re-read the grandparent post. He called for simulation from scratch, not prepared sounds mixed together.
I'll actually raise you one on the rest of your (misaimed) message... we should already have had dynamic sounds for about five years at least. At the very least, they could fucking shift the pitch simply by modifying the playback speed a touch every once in a while, so that every single scream from the dying demon in Doom doesn't sound exactly the same every single time. If I had a job in the industry writing sound engines that's one feature I'd sneak in even if it weren't called for.
physical modeling is just beginning to come of its own
Grandparent was not talking about "physical modelling". "Physical modelling" isn't. "Physical modelling" is when you create a filter that represents a part of an instrument, and string them together. While it has tolerably good results, and it quite fun for musicians to play with, it has no ability to simulate arbitrary objects doing arbitrary things; the process of creating the model is long and tedious, and what you can do with it is pretty sharply constrained.
Please re-read the grandparent of this post more carefully; what he's asking for bears the same resemblence to "physical modelling" that a water rocket packed in with your box of Trix bears to the Space Shuttle.
We already have Toy Story-level graphics and we've had them for years... Sony promised us Toy Story-level graphics in real time years ago, and by golly, they released the PlayStation 2 and behold, we did!
Which is to say... been here, heard this. Don't believe it till you see it, and are playing it. And a gentle reminder to the Sony fanboys to take the claims Sony makes about the PS/3 with a grain of salt; they played you for chumps last time, with the worst visual quality of the current generation. (There are many times where even the DreamCast beats the PS/2, because of Sony's poor decisions w.r.t. anti-aliasing.) Don't let them do it again.
But, where are the cards that can generate the sound of one arbitrary object hitting another? I don't just mean positional sound of pre-recorded samples, but really create the sounds from scratch (or an "audio-enabled model").
.
Not to put too fine a point on it, but... we don't know how to do that
Even what you may have heard of is wild, wild cheating compared to what you describe.
I've given some (intelligent, educated) thought to this issue, and here's the problem. Light is adequately simulated with a line in the macroscopic world. It is, technically, a wave, but it is so high frequency that we can ignore it unless we're trying to simulate at quantum scales. This keeps the complexity down to merely polynomial, and some smart people have figured out how to shrink that polynomial down to a surprising degree.
Sound is not so friendly. Imagine some reasonably complex object, like, oh, a chair. We'll even cheat and make it all out of one material and go ahead and make it as geometrically simple as you like, as long as it's a chair, which is to say, at least three legs and a sitting surface, preferably with some sort of back support. Now, mentally give the chair a tap.
Now, odds are you're not too practiced at this sort of mental visualization, but here's what happens. For the sake of argument, let's tap it on the direct middle top of the sitting surface. Let's cheat some more and assume that one impact is what makes the sound, rather then an oscillation at the tap point. (See how much we're cheating, and we'll still end up with an uncomputable scenario.) So the tap radiates outward from there and starts wiggling the legs. The sound partially bounces off the legs, and goes back, and some of the sound wiggles into the legs. The sound bounces all around in the leg, and every time it bounces off of the edge, it loses some of the sound to the atmosphere and some of it bounces back. By the time it hits the bottom, it's bounced several times. (See, sound can't turn except at one of those boundaries, where it is essentially absorbed and re-emitted.)
Meanwhile, the same thing is happening in the back of the chair. Plus, on the sitting surface, we have reflections of reflections of reflections to deal with, and thanks to the wonders of resonance, we absolutely have to track each and every one of them until they hit a really low level. In a fraction of a second, we have hundreds upon hundreds of seperate waves to track, and they aren't even rays, they are "wavefronts".... imagine a wavefront hitting the edge of the chair at an angle, like an ocean wave bouncing off the beach. It doesn't bounce like a particle, it is two entirely new waves, the one that reflects and the one that continues on.
Basically, we can't even simulate this horrid simplification, the real world is even worse. Sound is highly, highly parallel. Ultimately, sound simulation is firmly exponential and the constants are very, very high. Maybe if those magical quantum computers come online we'll get this, but we'll quite possibly never get it with conventional technology; we're always going to have to cheat.
(One can try to imagine a transformation of the chair where the sound travels in a straight line in some space, but I'll be damned if I know what that actually looks like in real code, nor am I sure that it would be any easier to compute then a straight-forward simulation anyhow. Bright ideas in this regard should probably not be posted on Slashdot and saved for your PhD thesis in Mathematics/Physics/Computer Science; they'll all be waiting with bated breath.)
I usually just deal with it by dedicating an entire 36 gig disk to swap in servers and use a much smaller swap partition on workstations.
You need to ask yourself, is there anything that server does that I'm willing to wait for thirty-six gigabytes of disk space to fill up, rather then just have it throw an "out of memory error" immediately?
I don't care how fast your drive is, 36GB takes a long time to fill up.
Lately, with all the RAM we have, I've been seriously toying with a "no-swap" policy. I'm not quite ready to take the plunge at 512MB of ram (dip into swap about once a month on average; gentoo emerge + mozilla + Open Office + a few other things and you do get there eventually), but once I have a gig I think I'm going to stop swapping.
Why? Because I can not think of a circumstance where I'd rather see the machine bog down for potentially hours, trying to use even a gig of swap, rather then just immediately throw an "out-of-memory" error. Nowadays, if the computer is reaching for that much swap it's most likely an errant task in an endless alloc loop anyhow!
Now, there are exceptions, but I think they are even rarer then most people think. For instance, if you're running a database server, you might need the swap... but again, are there any circumstances where you absolutely need that query to run... but don't mind waiting literally millions of times longer for it to run, probably causing a cascade failure anyhow (as lack of memory impacts more and more queries until the system can't catch up)?
It's not a popular opinion yet, but I think the era of "swap" is coming to an end, obsoleted by hard drives that are, relatively speaking, thousands of times slower then they used to be (relative to the processor times), and memories thousands of times bigger. The set of situations swap makes sense in is shrinking rapidly, and will soon be gone.
Eh? With modern computers, I can attain a frag rate of millions per second with Conway's Life running at full tilt. You must play a mean game of Quake.
Enterprise users [are lazy people with no time to do anything with their computers]
What crack are you smoking? It's the enterprise users who have stuff that's so mission critical they buy Windows source code from Microsoft and do stuff with it. You know, because tons of money is at stake.
It's the poor schmoes with three computers and a network hub that just want to plug things in and make it work, because one person-week wasted is a significant percentage of the company's time.
You obvious have no clue. If the enterprise users can score any kind of 10% improvement enterprise-wide with merely a few thousand man-hours invested, that's a good deal.
Any time you argue against the feasibility of nano-factories, you need to make sure your arguments do not equally argue against the possibility of life in our universe.
I don't think your post meets that bar. We know self-replicating factories exist. We are one. The question is, what improvements can we make and how well can we control and miniaturize them, not whether they are possible.
Immune systems and viruses have co-evolved. If a nano-bot develops that is made of a metal skin, that pretty much bypasses anything the immune system can throw at it, even if it's almost biological inside the metal skin, since most (all?) of the immune system keys off of proteins on the surface of cells.
Immune systems work because viruses have an evolutionary barrier to get to anywhere the immune systems won't work (i.e., a "half metal" virus can't mutate into being; such a thing may be possible but the gulf to get there is too wide; evolution is powerful but kooky and definately not omnipotent, it does have limits and in many ways, people overestimate as much as they underestimate). Nanotech will have no such restrictions. A self-replicating plague of some kind would still be limited by what kind of elements we have in our bodies, but there's enough iron and a few other metals to make enough nano-bots to kill us... and the nanobots have all day, metaphorically speaking, because the immune system will never even see them, let alone attack them, so they can kill cells at their leisure.
Not to mention the biological judo a deliberately designed killer could apply, recruiting the body's own immune system to help.
Cute, but I think your program may have a race condition in the ID assignment; I've tried several times now and I think I'm getting other people's information.
I think you're doing a "SELECT MAX(id) FROM database" when writing the link out onto the screen but you can't do that; other records are being inserted before the link is written. But that is just a guess.
I was able to hack the URL to find the one I want to send to a friend, though, so thanks from me.
What are you talking about, not everyone is supposed to get 100%. Everyone is supposed to TRY and get 100%, but the test is made hard enougth that the mean is a B or a C. That is pretty standard for all education levels. There are always a few teachers who make their tests real easy, but generally it isn't a big problem
No, that's the theory, a theory which is now being paid less and less lip service, let alone actually being implemented. Even so, if the system doesn't have built-in retries with the expectation that every student does better then last time (meaning there is a wide range of questions), it's still expecting 100% performance first try and penalizing anything else. (And re-learning the same thing year after year isn't the same; the retries need to be in a fast feedback loop, and you can't get much slower then "yearly".) And the system is largely "Learn A, test A, ignore results of test", "Learn B, test B, ignore results of test", with a final "average" at the end of the year.
One could hardly imagine a more learning-hostile environment. For all the efforts educators claim to put into developmental psychology, and studying how people learn, they seem remarkably impervious to the results of such things. They seem incapable of learning themselves.
"Global Warming" is a debate, albeit a lop-sided one. The causes and ratios of climate change is a fierce debate. (Have humans had an effect? Oh, sure. But are they anywhere near 100% responsible? Now thats a much more inflammatory question. Personally, while I know humans aren't 0% responsible, people trying to put the number in, oh, say, the high 90%'s or even 100% I find much less compelling then those with lower ratios.)
The inevitability of some change is not a subjuct of debate... except among some environmentalists who seem to think the status quo is the only good, and the only reason it's not being maintained is human action, and that if we only did the right things, somehow magically stasis would occur.
On the other hand, why must the whole exhibit be geared at the introductory level? A museum is a big place. Surely at least a little bit of room could be spared for some more sophisticated information in parallel with the simplified stuff? 10-year-old and Dad ought to be able to learn something.
(I have a similar criticism of the educational system. Why should we expect every child to 100% master the same math? Instead, set a baseline, and include varying levels of math in the same lessons. Especially as you get into Algebra and beyond, it's increasingly easy to challenge your students while making sure everyone understands the baseline, even in the exact same classroom. The myth that every student should perform 100% on every assignment is one of the worst blocks to educational reform today. We should expect children to get things wrong... because next time they try, they'll do better, and next time, they'll do better, and next time, they'll do better, etc.... and those children end up way ahead of the ones confined to just what they can do ~100% the first time... and as we've seen, 100% perfection has a habit of receding over time, instead of advancing as we need.
It's all the same fallacy, playing out over and over again, museums, schools, college, television shows, everywhere.)
Linear time == instantly.
Well, linear time == linear time. But I know what you mean.
The problem is that "Quantum Computing" is, in most people's minds, where Nuclear Power was fifty years ago. The fact is that if you throw enough bits at even conventional encryption, as I understand it, you still can't build even a theoretical QC that can crack the encryption, because I can always push the real solution below the noise limit.
QC isn't magical. Assuming it works at all, which I think there is stronger reason to doubt then believe*, it still has limits. QC works by running the answers all at once and then filtering the resulting wave for the answer. ("Filter" here has technical meaning.) If you run too large of a problem, you can't "amplify" the correct answer with an acceptably high probability, you're just as likely to get a wrong one. We may need all 20480 bits, or 204,800 bits, but eventually you're going to need a machine that is too large to actually be built to crack it. (Because of the way they works, qubit-based machines have rather firm upper limits to the complexity of a problem any given machine can compute, so even if a theoretical machine could crack our best encryption, it will probably still be the case that no real machine can do it.)
*: Large qubit machines, that is. I am quite aware small ones have been built and there has been progress in the field but I am of the opinion that we have already hit the point of diminishing returns. A qubit-machine requires too much isolation from the real world to ever be practical in this Universe.
Now, if you're like a PhD student or something in quantum computing, I welcome correction. But if you're just waving qubits around like a magic wand, please put the wand down and stop worrying that encryption is going to stop working. Besides, qubit-based machines only get us into exponential land, and there are an infinite number of super-exponential functions (using computer science lingo freely). I would be absolutely stunned if quantum computing didn't have a corresponding encryption algorithm it can feasibly implement that quantum computing must then brute force in its own way. That's just the way the complexity domains work; an infinite number of them, and my intution strongly suspects that the encryptors will always have a simpler problem then the decryptors.
Just provide the documents the CIA has.
The Discovery channel did one of its credulous stories on The Ark, and, well, it was better then the family reunion I was attending and I wasn't in a position to change the channel, so...
The CIA has bona fide pictures of... steel yourself... something. It looks like it might be a boat. (Personally, I think it looks too much like a boat; if you go looking for a thousand-year-old, unmaintained building and snap a picture from orbit of something that looks like a building, you've eliminated that site, because it should look like rubble, not a building.) A CIA photography analyst took bona-fide interest in it, but I'd lay money that since they are geeks, they take interest in a lot of things. (It's kind of their job, and part of that job is to shrink the class of "things we can't identify".)
The CIA legitimately classified it, and while the Discovery channel made hay out of that, my guess is that given the time frame, the image was classified, along with all other satellite images, on the (correct) grounds that even if you give someone a harmless image, the image itself will tell a lot about the capabilities and resolution of the satellite. Thus, the classification itself isn't really a mystery, only the people who think it's a mystery are a mystery.
I'm a theology-reading, bona-fide, for-real Christian... but I still classify the CIA evidence as another "Cydonia"-type picture, since I still apply the skeptical eye to things like this. (I respect the image analysts of the CIA a lot, it is a highly skilled job, but when faced with a random photo of a mountainside, the fact that they work for the CIA doesn't mean anything except to conspiracy theorists; the CIA, in this situation, is most likely just the group that happened to be holding the camera.)
The parent is modded as funny but it's true. As long as the errors are evenly distributed (i.e., a fundamental effect, vs. a huge-ass scratch across the platter which isn't evenly distributed at all), you can throw enough error correction at the medium to make the unreliability go away.
This isn't theoretical at all; CDs are routinely "destroyed", but they have a lot of error correction built into them so you don't even notice.
Computing the exact probabilities left as an exercise to the reader... but given any level of reliability there is some error correction scheme that can bring it up to any other given level of reliability (short of perfect, of course). Of course you can construct pathological cases that need as many bits as you like, the equations work that way too.
We do, say, 2048-bit encryption (asymmetric), because it would be "too slow" to do 20480-bit encryption. "Too slow" here is a fuzzy term, but generally speaking, if you're sending an encrypted email you don't want to hit "send" and have it delayed for three weeks while it gets encrypted. There's no real reason we couldn't do it today.
5 6 times harder.
As computers speed up, both encryption and decryption get faster. However, while adding another 128 bits to 128-bit symmetric cipher may be "free" with newer computers (and eventually will be), the 2^128 multiplicitave increase to the space the decrypters have to search is not free. To increase encryption power, the encryptors merely double their work. (To an approximation; I don't think the work load is strictly linear but it's a lot closer to that then exponential, and that's all that matters.) Meanwhile, for that relatively modest investment in encryption power, the decrypter's jobs got 340,282,366,920,938,463,463,374,607,431,768,211,4
This is why, in the relatively near future, we'll all have encryption that is effectively "unbreakable", because no conceivable decrypter could be built that could do the calculations to crack the encryptions, even with the raw materials of the entire Universe.
Practically speaking, most of us already have damn-near unbreakable encryption today; if you're connecting to a computer with SSH, SSH is most likely the strongest link in your security chain by far; the weak links are the computers on each end of the link, the humans on each end of the link, and possibly the facilities the computers are in. Nobody is going to tap your ssh stream and get any value from the massive decryption effort that would be necessary unless you're trading secrets worth billions.
Specialized hardware can only gain you a linear speed up, at best, and those calculations for "minimal computer" to crack a given encryption key are not extrapolated from modern computers, they are extrapolated from the maximum computation possible to do, given a finite energy supply. (QM-based computation advocates may wait until they have a large-scale (multi-thousand-qubit) machine to jump in here.)
Along with jsac's comment (more processor power exponentially benefits encryptors, only linearly benefits crackers, on the whole more power means a win for encryptors), I'd like point out this is only a set-back for encyption in-as-much as encryptors claim that their encryption will keep your data safe for all time. Which is to say, at least for the reputable encryptors, this isn't a set back at all.
If you insist on putting words in their mouth, then yeah, you might consider it a set back. But that's your misunderstanding, not theirs. All reputable encryptors have accounted for Moore's Law in their cost/benefits tradeoffs. Since it doesn't take much encryption power before it requires computers larger then the Universe to crack it via brute force (and since "cracks" on good encryption are really typically just ways of collapsing the search space, not procedures that give immediate answers, often adding more bits will require Universe sized machines, too), this isn't that big a deal for encryption. Push your key size up and be done with it. Even conventional machines can handle that today, it just takes longer.
every University and College *I know of* has an incredibly grueling music theory degree, and after taking a simple piano appreciation class, this CS student knows better than to take any more music courses regarding song analysis!
Hmmm. Speaking as someone who got a Masters in Comp. Sci., I found music analysis to be almost trivial, certainly I found it much easier then my fellow musicians. In particular, I was very easily able to straddle the line between "the rules" and "the feel".
(For those who have never done it, music analysis is interesting and useful for composers and players, but there is a strong element of "post hoc" analysis to it; analysis is really more interested in exploring the effect music has and sometimes a given theory will say X is happening when a quick and critical listening will say Y is happening. In the complementary direction, you'll see musicians use things like double-flats because even though E-double-flat is "literally" D (tonal pedants need not apply), in the theoretical context it makes more sense as an E, doubly-flatted. This is almost isomorphic to the relationship between software engineering theory and software engineering, complete with the "theory uber alles!" contingent and the "who the hell needs theory?!?" contingent. I'm one of the few people in the early theory classes who correctly used a double-flat on a test.)
(Then again, to be fair, I'm one of the rare comp. sci. types able to navigate theory and practice easily, so I'm probably an odd bird anyhow.)
What would be interesting would be to qualify various music genres along various lines and see which qualifications match their observations. Mozart was a genius and his music is like no other, but I can't think of any single attribute he holds a monopoly on (though the combination is unique, IMHO). A lot of "concept albums" share some of the off-beat regularity and interest in themes, for instance, though they obviously differ from Mozart in more ways then they are in common.
Hey, how about a link?
I'm looking for something that would cover such topics as basic desktop do's and don'ts, like 'do choose a non-dictionary password' and 'don't blindly drop to root and install an unverified/unauthenticated RPM that you receive via email,' etc. Anyone seen a guide like this?
Why?
Do you expect anyone to actually read this document?
Oh, I wish I were being sarcastic.
Either enforce things (your password policy), or wait for people to have trouble so you know what to document (every installation is unique, and you're wasting time trying to predict how your users will react when you could just wait and see).
The only purpose of such a document, in the end, is CYA anyhow. And again, I wish I were being sarcastic. If you can't enforce it, people are going to do it.
The only possible exception is if this is a technical group of users who will be daily and strongly held accountable for violations. Basically, the only group of people who meet these two criteria are Computer Science (or related disciplines) students.
Otherwise, don't bother. Not sarcasm.
In order for the TiVo to know what is on, it needs to dial-up and download the lineup. TiVo has streamlined some stuff by broadcasting some "infomercials", but you still have to have that dial-up to get the lineup.
Of course, maintaining the system, keeping it up to date, lining up the contracts for the modem banks, etc. isn't free, and it is also TiVo's primary money source.
One could make a case that a pre-existing Internet connection could be used, but that's too complicated for most people.
You can buy a TiVo and not use their provided service, but the majority of the good stuff, which involves understanding the TV lineup, won't work.
I would love to see a movie devoted to the great star trek time warp.
You mean, other then Star Trek 4, Star Trek: Generations, and Star Trek: First Contact? And pretty much the entire run of Voyager?
"Time Warping" is the worst problem with Star Trek. You can't build drama because nothing is ever at stake that won't be wiped away by next week's time travel episode or movie. That problem alone is enough to sink the series; the entire concept has devolved into some of the worst "collaborative writing" I've ever seen. (Ever done a "collaborative novel" online? The resulting continuity trainwreck bears an uncomfortable resemblence to Star Trek now...)
No, his point is that one of the favored FTL communication methods of the "I have a middle-school education in Quantum Mechanics" contingent doesn't work, ye olde "light pulse through a material speeding up the speed of light". The light travels through ever so slightly faster then the speed of light, but he still couldn't push information through any faster.
You missed the point, you're talking about quantum cryptography, which is quite a different thing.
No, you missed the point. Quantum cryptography takes advantage of the same principles, and you'll note that nobody is using it to communicate faster then light either.
You can not force the polarization of light. There is no way whatsoever to do it. Without that, your scheme doesn't work.
Look, if it were that easy, we'd have done it already. It's not like we don't have the equipment or the ability. All it takes is a bit of money to set your system up. All you psuedo-physicists who think that FTL communication is so simple, go set this system up and make a few billion selling it. Or maybe, just maybe, it's not as easy as your oh-so-thin "understanding" of quantum physics makes it out to be?
Put up or shut up. It's not like we discussing something that would be several trillion dollars just to build a prototype, it's down in the merely several thousand.