Two things strike me about this: (1) As several people have already pointed out, Helium fusion is, to put it politely, "technically challenging." While both major tracks of fusion research (inertial and magnetic confinement) have made substantial progress in the past decade, they're nowhere near being able to build production hardware of any sort.
But second, and maybe a bit more relevant - if we're already setting up that sort of infrastructure on the moon, why the hell not just use solar energy? The main limiting factor in solar energy today is atmospheric damping - solar energy is more efficient on Mars than it is on Earth (despite its getting only a quarter as much sunlight) just because it has less atmosphere. If you're going to be on the moon, or for that matter in any vaguely stable fairly high orbit, it seems much more straightforward to simply set up large photovoltaic arrays.
(Of course, this brings up the problem of how one transports this power back down to the ground - not hard if the "ground" in question is a space station or a lunar facility, but somewhat trickier if the ground is on Earth. But I suspect that it wouldn't be much harder than setting up a permanent gas-mining facility on the moon - something which would be complicated significantly by the fact that, even if the moon has a lot of 3He, it's not particularly _dense_ there)
I found a fairly straightforward solution to this problem. I wrote a small wrapper around a known-good md5 function, compiled it and placed it in a nonstandard location. (Thus it doesn't have a widely recognizeable filesize or md5 to be detected and stomped) Then I wrote a simple shell script which checksums various critical files on a regular basis and tests the MD5 values against a record it keeps, again in a private location. Whenver a change happens, it sets off alarm bells all over the place, both in syslogs and on the console.
On top of this I stuck in one small bit of shell script that allowed me to modify a file myself without setting off alarms - it simply recalculated the md5 value and updated the record files.
I suppose this is theoretically vulnerable to an attacker reading through/etc/crontab, then checking each local shell script for a sensor and carefully overwriting my own nonstandard code - but if any attacker has that much free time on his hands, there's a limit to how much of a sensor I can implement.
The nice thing about this code is that it also implicitly tests for corruption of critical files after fsck-triggering events like kernel panics or total power failures. (That's actually what prompted its initial writing) And it's remarkably trivial to implement, even more so if one simply copies an off-the-shelf md5 binary rather than compiling one's own wrapper.
The original post didn't make it clear which category was asked about: Completely non-technical books, introductory technical books, or books for people with a technical background (engineers or math-oriented geeks of various sorts) who want an introduction to the field?
Since the first two have already been covered pretty thoroughly, here are a few suggestions for the third, based on classes I've taught (undergrad physics) in the past few years:
For generic physics, there are standard freshman physics books like Halliday & Resnick or Serway, but these are good mostly if you want to learn basic mechanics of balls rolling down inclined planes and so on. If you're not interested in carefully learning the mathematics and techniques but would rather go for the ideas, a slightly harder book is a gem, Feynman's "Lecture notes in physics." Three volumes, each worth their weight in gold. That man can explain things.
For quantum mechanics, the best intro book for someone with a technical background is probably D. J. Griffith's "Introduction to Quantum Mechanics." Very friendly writing style and overall good, although the way he covers a few topics still makes me wince a bit. The background needed for this is calculus up through basic differential equations, and linear algebra.
For general relativity, AFAIK there aren't any books which are both good and "exciting." The two most common books to start with are either Schutz's "A first course in general relativity" or Hawking & Ellis' "The large-scale structure of space-time." Both are a bit on the dry side, but can quickly get you up to spin. I'd recommend the latter more for someone who just wants to survey the field, since although it's harder to learn the technical basics from it, it gets on more quickly to exciting topics such as black holes and cosmology.
For things like field theory and string theory, unfortunately your choice is either Brian Greene's book and similar nontechnical works, or jumping straight into hardcore texts like Green, Schwarz & Witten or Polchinski. Not for the faint of heart, but quite nifty. If anyone knows any mid-range-technical books on this, I'd be glad to hear about it...
And finally, an ObPlug for an upcoming book: In the not-too-far future there should be a book by Lindesay and Susskind on the quantum theory of black holes which promises to be very neat, though definitely on the technical side. (Conflict of interest notice: Susskind is my thesis advisor. But he's remarkably good at teaching this)
I wonder if the author has considered that the primary applications of this work are probably not in influencing file-sharing networks so much as in politics. The P2P network that first comes to mind is ordinary web access within China. This is a situation where the government has an active interest in preventing any politically sensitive information from being propagated within the country, and so the ideas of this paper are directly applicable.
I'll leave the relevant ethical issues as a matter of discussion -- but I would suggest that this is a far more serious reason to be concerned about corporate research into network interruption.
Apart from the issue of whether any legislation whatsoever would be reasonable, I would be strongly opposed to the suggested law for a more practical reason:
There is no way to search for copyrighted material without examining the content of all traffic going across networks. Such a law would require the police to continually monitor large amounts of all network conversation, with active attention to their content, in direct contravention of all established law and precedent regarding wiretaps and other surveillance techniques. It essentially amounts to a call for blanket surveillance of the population in order to protect the copyrights of certain business interests.
Given that I have opposed such expansions of federal authority in the past, when the excuse was the (arguably much better one) of capturing terrorists, I cannot imagine countenancing such a law when the only incentive is to help some groups make a buck.
Hmm... so after reading this story, I'm not certain whether this is meant to be a user-installed software package, a trojan, or a remote exploit of a vulnerability in IE and Netscape.
If the former, what benefit does it claim to give the user in exchange for the obvious annoyance?
If the second, how much damage will it do to the system in the process of installation in order to make it difficult to remove, and will this damage be actionable? (I'm mentally comparing it to the story about the Celine Dion CD above...)
If the latter, how complex a firewall filter will it take to splatter this? (Since it goes along the HTML channel obviously this is much more sophisticated than packet filtering...)
If you are going the homebrew route, one case that may be worth checking out is the case from an old compact SPARC. They had very small, very dense cases which are just great for luggable applications, and quite robust. You can probably pick one up very cheap as scrap.
As a minimal solution, it isn't too hard to actually build a box to this sort of spec. For a case, start with a toolbox and hollow it out; then strap in a power supply, a small motherboard, and all the goodies. A bit of cutting work should let the ports and so on come out.
This is different from trying to build a portable or luggable since it doesn't need its own power source -- if you're doing music, you probably have access to 120VAC somewhere. So a traditional power supply can work.
A setup like this could easily come down to the $1000 price range, and open you to putting more money into a really good sound card...
The thing we would get if someone were to find a polynomial-time algorithm for any NP-complete problem is an immediate, poly-time algorithm for every NP-complete problem. This is because the definition of NP-complete is that there is a (known) poly-time algorithm to turn any one NP-complete algorithm into any other, so just by composing these two you get them all. (I'll attach a glossary at the bottom -- most people on this list probably aren't mathematicians:)
But OK, what does this mean realistically? The good news is that there are several very useful NP-complete problems; probably the best known (as someone has already mentioned) is the travelling salesman, and being able to do fast TS problems could mean incredible reductions in cost for shipping of goods and things like that. All sorts of problems in computer network architecture are also NP-complete; think about trying to design an internet which is both fault-tolerant and maximizing bandwidth.
The bad news: There are two things. First of all: This does not mean encryption of any sort is broken! The heart of public-key crypto is that factorization takes exponential time (or more specifically, the discrete logarithm, which is at the heart of fast factorization, takes exp time) and so if you could do poly-time factorization, you could break various algorithms like RSA. But factorization is only conjectured to be NP-complete; there is no proof, and in particular the explicit algorithm which would be needed to use a poly-time algorithm for some other NP-complete problem for factorization isn't known. This doesn't mean it can't be done; it just means that there's one other significant step between finding such an algorithm and breaking crypto.
The second problem is that even a poly-time algorithm isn't necessarily useful if the coefficients are large. What poly-time really means is that, in the limit of very large inputs, computation time doesn't go completely out of control; the fact that (to the best of current knowledge) factorization isn't poly is what makes adding one digit to key size enough to increase the difficulty of decryption by a factor of two. (i.e., the work increases as an exponential of the input) So this is important when you're trying to create "sufficiently large" inputs to jam up an algorithm. But for real-world problems that people are trying to solve apart from crypto, an O(N^1000) algorithm might technically be "better" than an O(e^N) algorithm but practically still be way out of reach.
In fact, most of the interesting NP-complete problems such as travelling salesman are routinely worked on by methods which give approximate answers in fairly short time; this turns out to be more than good enough for a remarkable range of uses, which means that the advance of getting poly time wouldn't be as earth-shaking for most real-world applications.
Hey, cool! (I'm a CU alum as well...) Some ideas for your project:
* Robotics is an obvious one -- the term itself was coined by Asimov, and the term "robot" by Capek. Asimov's collected works are definitely the most influential on this subject, to the extent that his "laws of robotics" are in fact quoted at the opening of one of the standard reference works on the subject.
* For computers in the most modern era, Gibson's _Neuromancer_ had a surprisingly deep impact; a large fraction of modern terminology (even the general usage of the word "Net") stems from there, as do many of the ways in which programmers visualized what they were trying to achieve.
* One thing that might not be so obvious is nuclear everything. There was actually a series of short stories published by various authors in the early 1940's (pre-Hiroshima!) which were remarkably technically accurate and which were being read at places like Los Alamos. Some stories on this thread:
"Nerves" by Lester del Rey -- this appeared orignally as a short story and was then expanded into a novella. The former is better written and more historically significant; you can find it in "My Favorite Science Fiction Story," edited by Greenberg.
"Solution Unsatisfactory" by Robert Heinlein, in his anthology "Expanded Universe." "Blowups Happen," by the same author and in "The Past Through Tomorrow," is about nuclear power.
OK, really this is such a broad topic that it's impossible to generate even a basic list; science fiction was so influential because the scientists read it, as children and as adults, and their notions of what sort of projects ought to be attacked were deeply shaped by this. If you go from a historical perspective, things to go for might be
* Hugo Gernsback was the first of the "great editors;" he edited some of the chief science fiction pubs back in the 20's and 30's, and was largely responsible for a vision of a technological utopia. He wrote some books of his own as well; this was all tied in heavily with things ranging from the Art Deco movement in architecture to the technological movement in Fascism. His books can be good refs. "The Jetsons" is a direct descendant of this line of writing, to give you an idea...
* There's a huge amount of "golden age" (40's-50's) SF which really shaped ideas about space travel, robotics, and nuclear energy. For this it might be best to go to anthologies from the period; the one edited by Greenberg mentioned above is good, as are any edited by John W. Campbell. (The second of the "great editors")
* The modern discussion of computers really started around the 60's, but I don't know this era as well. But from the perspective of a computer programmer, (and thus the receiving end of this cultural influx:) I can't think of any writers that made a huge impact in this regard between Asimov and the other golden age writers and Gibson in the mid-80's. And at that point it gets rather hard to tell what had historical significance, if only because it's so recent.
Hope some of this helps, and good luck on your course!
Well, I'm seeing a completely different issue here, beyond other people being able to craft virii exploiting the same holes that this Magic Lantern does. (Although I'm assuming that as security holes get patched, Magic Lantern will ultimately refer to a family of virii rather than any single virus; it's going to make McAfee's job of trying to explicitly exclude it from virus searches all the more ridiculous)
The thing that occurs to me is that, back when I was an easily amused kid I used to capture computer viruses, dissect them and study them. If Magic Lantern is genuinely going to be an effective way to retreive data -- and if it's a virus designed by a team of top-level professionals, which it is likely to be, then it should be so -- then how long a matter of time is it going to be before everyone and his mad bastard cousin starts to make copies of this virus and mutate it for their own ends? This seems like it would quickly become a valuable corporate espionage tool, and then a personal espionage tool, and then just a total disaster area.
The problem with this is, if they design a powerful cracking tool which by its nature must be primarily built out of code resident on the target's machine, it's only a brief matter of time before such software and any upgrades thereof enter the mainstream of black-hat equipment.
Frankly, I'm not looking forward to script kiddies with tools like this...
HDCP uses a linear system for generating the shared secret.
From a part-time mathematician's perspective (ok, actually a physicist) this was the line that just made my jaw drop. What were they thinking?! If this text is correct, this algorithm may as well have been designed by a high-school student.
As several people have pointed out already, this is really one of the big threats of the DMCA -- that companies will go around using incredibly poor standards like this, and be immune to any pressure to improve their quality because their customers are legally forbidden to ask what they are receiving. It says a great deal about the present legal climate that anyone could get away with a mess like this cryptosystem in a commercial product.
All the talk I've heard about diamond IC's is for very specialized applications like radiation-hardened chips (EMP's from a tactical nuke won't fry the diamonds themselves, though the rest of the circuit may be in more trouble) or very high-temp applications where for some reason you can't put the logic somewhere else.
But I thought the major problem with diamond chips wasn't fault lines so much as the fact that you can't p-type dope diamonds; for reasons still unknown they simply boot out any such dopants. Which would make it kinda tricky to do anything useful as far as diodes or transistors. Are these people claiming any sort of innovation in that direction? Or is this result more useful for using diamond as a structural material?
Guys, this needs to go out in book form, not as an e-book or anything less. It needs to go out as a book because this is something that needs to be used in classes in high schools, and to get that done, it has to be out under a publisher's banner in dead-tree format.
Will it be hard to get this into schools? Yes, maybe. But school boards and schools have been much better in the past few years about getting this sort of thing into classes, especially English classes, under the rubric of "diversity." And in the aftermath of Columbine this book would fill a deep and urgent need that I believe even the people in decision-making positions will understand.
Will it be hard to get into print? Somewhat. But this is something we could realistically do as a community effort. There are only a few things we need. First, to get permissions from all posters and clear up the legal issues. This means we need everyone who contributed to come forth and give an OK if they want their stuff used. Second, the thing needs to be pulled together and properly formatted; John Katz's excellent introduction, plus possibly more pieces similar to that interspersed with the community contributions. Then the thing needs to be formatted nicely (not difficult; we have plenty of page-layout geeks here. And this part is important; anything well-formatted looks profound and sells much better)
Finally, and most importantly, a publisher has to be found. But the more discussion we manage to generate about the Hellmouth between now and when the formatting is done, the more likely we are to be able to convince a major house to have an interest in this book.
I would like to make a RFD: would people be willing to join in organizing this project and bringing it to completion? We can get this moving quite quickly if we all work together.
This question honestly seems a bit strange to me. The major reason that we haven't moved to something else is that there is no reasonable alternative yet out there.
There are two kinds of problem, that of choosing what "new engine" to use and the costs of converting. Since someone else has already spoken very well about the latter problem I'll just say something about the former.
What's needed to power cars is a fuel that is highly portable, (e.g. no solar panels or wind turbines) capable of producing not only large amounts of power but high impulses, (i.e. no burning pure ethanol) is reasonably safe, (no nuclear-powered rocket engines) and reasonably inexpensive. Unfortunately, that rules out virtually all of the proposals on the market.
One of the things that has killed most of these (such as electric cars) is that they fail on one of the first two criteria. Insufficient power means that the car won't zoom fast enough, and yes, that is an issue; nobody is going to buy a car that tops out at 60mph. It's simply not a practical vehicle in a society built around major roads and highways. (Maybe in Europe or Japan one has a better chance) Insufficient portability means that there's not enough range, which requires any number of Rube Goldberg schemes to work around but always comes out to meaning that the car is good for commuting between a few nearby points with refueling docks already at them. This is why you only see electric vehicles driving around on large campuses or whatever.
Fuel cells suffer from an even more serious problems: They really don't exist yet. Cells capable of powering a car, satisfying anything even remotely like the above requirements, are still a few years away at least. Fuel containment, ease of recharging, inhibition of flammability, recoverability of fuel, and high-drain performance are all still significant issues. Once those are fixed, we get to talk about conversion costs.
The one thing that is making progress is hybrid systems, which use gasoline for the highest-drain parts of driving (acceleration) and switch to electric or other motors for low-drain parts. (highways, etc.) The technology isn't simple but it's advancing rapidly, and several such vehicles are already on the road. More are almost certain to come in the near future.
Of course, software configuration means that it would require all-new drivers to work under any other OS. But provided it doesn't look too much like a food product, it may turn out to be a somewhat useful gizmo. Think of it as a mouse with multiple input senses; like having meta-keys for mouse input.
(Maybe this will give some incentive for my hands to ever leave the keyboard? Nah...)
Krasnikov's Subway is an old idea; it was written as a response to Alcubierre's warp drive article, which I think we talked about here a while ago. It is perfectly consistent as a solution to classical general relativity, but the requirement for this is an enormous (about 10^80 times the mass of the universe) amount of negative mass. There are various quantum theorems that tell us that QM prevents anything more than infinitesimal amounts of negative mass from forming, so I wouldn't bother planning any Alpha Centauri commutes aboard this subway. (BTW, the Alcubierre warp drive has very similar problems. Both of these came out in the early '90s.)
A poor man once went to his wealthier neighbor and told him, "Sir, my son is soon to be married, and his bride's parents will be coming to visit. But my house is bare and I am ashamed that they will see it like this. Could I borrow from you your silver spoons?"
The rich man considered it - he was not one to lend out his things lightly - but ultimately agreed. The next day the poor man returned with the spoons, and another small silver spoon as well.
"What is this?" the rich man asked.
"During the night one of your spoons gave birth. Since it is your spoon, the child should be yours as well."
The rich man thought this was ludicrous, but he was not one to pass up good fortune, so smiling he accepted the small spoon.
Several weeks passed, and again the poor man came to ask to borrow the spoons; remembering his previous good fortune, the rich man assented, and once again found himself presented with another spoon.
A few weeks after that, the poor man came to the rich man again, explaining that the wedding itself was coming up, and asking to borrow his silver candlesticks. At this the man hesitated; spoons, yes, but those candlesticks were pure silver! But mindful of his previous luck, and with visions of money dancing in his head, he agreed to let the poor man borrow them.
The next day the poor man came back, empty-handed. "Where are my candlesticks?" the rich man asked.
"It is horrible! Last night they were beautiful - but this morning, I came and they were both dead!"
At this ridiculous line the rich man flew into a rage. He accused the poor man of being a thief, and soon the two came up before the town rabbi, who heard their case.
The rabbi considered it carefully, and said: "If a spoon can be born, why can a candlestick not die? If you chose to accept nonsense when it was profitable to you, you can accept the same nonsense when it brings you loss."
So it is with Amazon; as they accepted the idea of one-click patents to protect their money, they can accept the idea of sampling audio as well. Perhaps once this is finished neither side will be so foolish again - and the patent office will not encourage them.
Remember to physically unplug your computers from the network, since we are doing line maintenance that could cause your computer to explode. In fact, it's best if you wrap your computer in blankets so that if the worst should happen you won't be injured.
Well... just remember that all high-energy papers are preprints before they're published. 99% of the stuff on xxx really does get published and is meaningful science.
But as you point out, it isn't peer-reviewed, and anything that shows up there needs to be taken with a serious grain of salt until it's checked.
(1) Very possibly. But the theory itself is still cracked.
(2) Nonetheless, de Aquino claims that the null mass of the photon (the vanishing of its inertial rest mass) implies that it is outside the effects of gravity. I argue that the way de Aquino referred to the mass of the photon as having to do with gravity is simply wrong.
(3) This is a good question and one that bugged me for a long time. The real answer is that the graviton picture - the approximation that the gravitational field can be decomposed into small excitations which behave more or less like free particles, with some additional interactions - is only valid in the limit of weak gravitational fields. For black holes and similar strong fields, the self-interactions of the gravitons are so strong that they can no longer even approximately be thought of as particles.
This actually is fairly closely associated with why quantum theories of gravity are hard. Basically one can show that a quantum field theory of spin-2 particles is inconsistent. It reduces to something consistent in the low-energy approximation, where gravity can be thought of as the exchange of individual gravitons. But as soon as the self-interactions of gravitons becomes significant, this picture of gravitons as particles stops producing meaningful predictions.
And this is exactly the point where we have to turn to our one well-defined and reasonably well-understood theory of quantum gravity: superstring theory. While this theory is not experimentally verified, there is substantial reason to believe it will be. (Which I won't go into now:) Most importantly, string theory can correctly predict aspects of the low-energy behavior of black holes which show up when classical GR starts to fail. (Hawking's "information paradox" is such an issue)
So it's possible to calculate the gravitational properties of black holes using string theory and then back away and say: If I'm far away from the black hole, it just looks like an ordinary gravitating object, gravity is weak, and the graviton approximation should apply. Indeed it does; it turns out that, from far away, the black hole looks like a uniform sphere of radius equal to its event horizon, and as far as a distant observer is concerned, it does indeed emit gravitons from this distance. But all information about the internal structure of the black hole is kept within its event horizon; we can't see inside the horizon using these gravitons.
One-line summary: These guys are better at graphics than they are at physics. No, it doesn't work.
Since the discussion below is a bit more technical than I usually send to Slashdot, here's a brief summary:
* No, it doesn't work. There are in fact some very fundamental reasons why it can't be made to work, either.
* It can be shown (through some fairly messy math) that it is not possible to cancel gravity by introducing other particles, unless you somehow had negative-mass particles to strap to your spaceship.
* Even if you could do all of this, how would you hold a sphere of photons in place?!
The rest of this is a bit heavy, so you may just want to skim. If you want to get a fairly easy-to-manage and good intro to this topic, you may want to check out Rindler's _Introduction to Relativity_.
Okay, on with the show. There are three basic problems with this suggestion.
Problem 1: "Photons have a null gravitational mass." It is true that the rest mass of a photon is zero. (The rest mass of a particle is the mass of that particle as measured by an observer at rest with respect to it. Special relativity tells us that objects with nonzero rest mass must travel slower than light, and objects with zero rest mass travel exactly at the speed of light.) However, the "effective mass" of an object for gravity purposes is <i>not</i> its rest mass. The quantity which generates gravitational fields is a more complicated quantity called the stress tensor, which is nonzero any time there is energy or momentum in a system. Photons definitely have a nonzero stress tensor and as a result do produce gravity.
In fact, because of the way the stress tensor is defined, it is impossible to entirely cancel it with <i>any</i> configuration of masses, unless you somehow had something with a negative mass. If you find something like this please let the rest of us know; there are several well-known, mathematically valid, ways to design superluminal drives based on them. However there are also several results that quantum mechanics prevents the formation of more than infinitesimal quantities of such matter.
So that's problem one; photons don't have "null mass" (whatever that means) and so don't automatically cancel gravity. On to
Problem 2: If we had a shell of photons around our ship, they could exactly cancel the gravitons. Not quite. First of all, I should say that gravitons are the smallest quanta of gravity and are in many ways analogous to photons. (In the language of general relativity, where gravity is a stretching of spacetime, a graviton represents a localized deformation of spacetime. Similarly, a photon represents a local deformation of the electromagnetic field, which is the particle physics way of stating the fact that light is an electromagnetic wave.)
Now, the way gravity works (in graviton language) is that gravitons can be emitted or absorbed by anything with a nonvanishing stress tensor. (The bigger the stress tensor, the more easily this happens) So say two massive objects are moving along; Alice (having a nonzero stress) emits a graviton, and recoils by conservation of momentum. Bob (also having a nonzero stress) absorbs this graviton, and also recoils. So effectively both Alice and Bob have changed their momentum, i.e. exerted a force on one another.
Now, say I wanted to cancel this out with some shell of Mystery Particles which would give the exactly opposite forces. These particles would have to do a couple of things:
* First of all, they had better travel at the speed of light so that they arrive at the same time as the gravitons in order to cancel them out. So these had better be massless. (See note above)
* Second, they should be able to interact with any particle that has mass, and they should interact equally strongly with any two particles of the same mass. (e.g., if our mystery particle hit electrons twice as hard per unit mass as they did protons, they wouldn't just cancel the gravity, they would also knock about everybody's particles quite a bit.)
Now it turns out (and this would require some lengthy tangents to explain in detail) that these two conditions are enough to <i>uniquely</i> specify the properties of this particle! In particular, the particle must have zero mass, and (for those of you with some physics background) have spin 2 and be symmetric tensor fields. These are exactly the same as the properties of the graviton, which isn't surprising since these things have to cancel them. But...
* It's fairly straightforward to show (using a bit of field theory) that forces mediated by spin-2 symmetric particles are always attractive. (Like gravity, and as opposed to electromagnetism. EM is mediated by photons, which have spin 1 and can be either attractive [opposite charges] or repulsive. [like charges]) So unfortunately, no Mystery Particles can cancel out gravity.
OK, and just in case this isn't enough, one other point:
Problem 3: Even if you had some magic way of getting around this, e.g. introducing multiple species of graviton and giving them some extra interactions which let them exactly cancel gravity while somehow not affecting the ordinary operation of gravity in <i>any way</i>, (Which, incidentally, doesn't work; if these particles are just like gravitons, they should be produced in nature just like gravitons and so would be everywhere) How the hell are you going to keep a bunch of massless particles sitting in a perfect shell? They're <i>massless</i>. They tend to fly away.
(OK, and now a more technical note for those of you who bothered to read this far: There is actually one other possible way to satisfy the constraints for a mystery particle, allowing it to couple to matter identically to gravitons without also being a graviton. Namely, the particles can be part of a multiplet of particles which are related to gravitons by some continuous symmetry [discrete symmetries would just give multiple graviton species] and so all matter would have to couple universally to them as well. This could be done even for particles that did not have spin 2.
But some math prevents this from working. If we want our mystery particle not to have spin 2, the symmetry must relate particles of different spin. By a key result known as the Coleman-Mandula Theorem, the only way to achieve this is with a type of symmetry known as supersymmetry. If you do this, you find that it is possible to construct a complete suite of particles called a "gravity multiplet," consisting of the graviton and several particles with spins ranging from zero to 3/2, all of which couple universally to matter and do everything we wanted.
So does this solve it? Well, not quite. The condition for these supermultiplets to cancel gravity is very well-known; it's called the BPS condition, and it says that the objects exerting the forces on one another have to obey very specific relationships between their charges (electric and other) and their masses, as well as some requirements about their relative shapes and orientations. When you work it all out, there are a few very specific configurations where this happens, but (1) It doesn't happen for arbitrarily shaped objects like spacecraft, and (2) All of this would require that supersymmetry be present and unbroken in nature. This would have some <i>really</i> visible experimental consequences, e.g. atoms looking pretty much nothing like they actually do. Such a symmetry is quite definitely ruled out.
So I'm afraid that there's nothing of this sort that can cancel out gravity. Sorry...
Call me a heretic of the Open Source movement, but:
I use Mathematica regularly. Its syntax is arcane only to the extent that it is itself a programming language with a complex instruction set; and the source is closed. But it has two features which I believe counter this. First, there are simply no programs of comparable power for complex symbolic manipulation; and yes, I am familiar with the open source packages. But algorithms for solving symbolic differential equations and large integrals are simply too much for small groups of people to do; their design requires substantial teams of very skilled people. And while the open source community has mustered many (most?) of the best programmers in the world, the skills of applied mathematicians simply aren't as prevalent in this world.
And second, Wolfram Research (the company which makes Mathematica) has systematically made itself as open as possible; they routinely solicit user suggestions and input, and sometimes incorporate user-submitted packages and code into their own releases. While the core code itself is compiled, a large fraction of the program comes in the form of modular packages which come in the form of Mathematica source code.
In short, I'll say that Not All Closed Source is Bad. The modularity of Mathematica, the publication of the API's and the source to all of the interpreter-level packages, and the responsiveness of the company to its users have given it most of the same advantages that true Open Source posesses.
(All of this applies as well to Maple; that system is oriented more towards large data set manipulation rather than pure symbolics, however, so the situation is slightly - but not very - different.)
So call me a heretic; but I believe that, when the cost of a large number of specialists needed to develop a package is high, the creation of a closed-source, sold-for-money package is reasonable so long as the company does not behave in a manner detrimental to its users. Therefore I would suggest that the continued use and active support of systems such as Mathematica and Maple is beneficial to the community as a whole and should be continued, even in the presence of open-source alternatives.
Slashdot Mirror
Two things strike me about this: (1) As several people have already pointed out, Helium fusion is, to put it politely, "technically challenging." While both major tracks of fusion research (inertial and magnetic confinement) have made substantial progress in the past decade, they're nowhere near being able to build production hardware of any sort.
But second, and maybe a bit more relevant - if we're already setting up that sort of infrastructure on the moon, why the hell not just use solar energy? The main limiting factor in solar energy today is atmospheric damping - solar energy is more efficient on Mars than it is on Earth (despite its getting only a quarter as much sunlight) just because it has less atmosphere. If you're going to be on the moon, or for that matter in any vaguely stable fairly high orbit, it seems much more straightforward to simply set up large photovoltaic arrays.
(Of course, this brings up the problem of how one transports this power back down to the ground - not hard if the "ground" in question is a space station or a lunar facility, but somewhat trickier if the ground is on Earth. But I suspect that it wouldn't be much harder than setting up a permanent gas-mining facility on the moon - something which would be complicated significantly by the fact that, even if the moon has a lot of 3He, it's not particularly _dense_ there)
I remember, many years ago, discussing with people how one day all of our ordinary home appliances would be computerized.
Then four or five years ago, two things happened: I moved into an apartment with inadequate heating and insulation, and I bought a P2-266.
And now, my space heater runs UNIX! I just put xflame on, and it's an instant fireplace...
I found a fairly straightforward solution to this problem. I wrote a small wrapper around a known-good md5 function, compiled it and placed it in a nonstandard location. (Thus it doesn't have a widely recognizeable filesize or md5 to be detected and stomped) Then I wrote a simple shell script which checksums various critical files on a regular basis and tests the MD5 values against a record it keeps, again in a private location. Whenver a change happens, it sets off alarm bells all over the place, both in syslogs and on the console.
/etc/crontab, then checking each local shell script for a sensor and carefully overwriting my own nonstandard code - but if any attacker has that much free time on his hands, there's a limit to how much of a sensor I can implement.
On top of this I stuck in one small bit of shell script that allowed me to modify a file myself without setting off alarms - it simply recalculated the md5 value and updated the record files.
I suppose this is theoretically vulnerable to an attacker reading through
The nice thing about this code is that it also implicitly tests for corruption of critical files after fsck-triggering events like kernel panics or total power failures. (That's actually what prompted its initial writing) And it's remarkably trivial to implement, even more so if one simply copies an off-the-shelf md5 binary rather than compiling one's own wrapper.
The original post didn't make it clear which category was asked about: Completely non-technical books, introductory technical books, or books for people with a technical background (engineers or math-oriented geeks of various sorts) who want an introduction to the field?
Since the first two have already been covered pretty thoroughly, here are a few suggestions for the third, based on classes I've taught (undergrad physics) in the past few years:
For generic physics, there are standard freshman physics books like Halliday & Resnick or Serway, but these are good mostly if you want to learn basic mechanics of balls rolling down inclined planes and so on. If you're not interested in carefully learning the mathematics and techniques but would rather go for the ideas, a slightly harder book is a gem, Feynman's "Lecture notes in physics." Three volumes, each worth their weight in gold. That man can explain things.
For quantum mechanics, the best intro book for someone with a technical background is probably D. J. Griffith's "Introduction to Quantum Mechanics." Very friendly writing style and overall good, although the way he covers a few topics still makes me wince a bit. The background needed for this is calculus up through basic differential equations, and linear algebra.
For general relativity, AFAIK there aren't any books which are both good and "exciting." The two most common books to start with are either Schutz's "A first course in general relativity" or Hawking & Ellis' "The large-scale structure of space-time." Both are a bit on the dry side, but can quickly get you up to spin. I'd recommend the latter more for someone who just wants to survey the field, since although it's harder to learn the technical basics from it, it gets on more quickly to exciting topics such as black holes and cosmology.
For things like field theory and string theory, unfortunately your choice is either Brian Greene's book and similar nontechnical works, or jumping straight into hardcore texts like Green, Schwarz & Witten or Polchinski. Not for the faint of heart, but quite nifty. If anyone knows any mid-range-technical books on this, I'd be glad to hear about it...
And finally, an ObPlug for an upcoming book: In the not-too-far future there should be a book by Lindesay and Susskind on the quantum theory of black holes which promises to be very neat, though definitely on the technical side. (Conflict of interest notice: Susskind is my thesis advisor. But he's remarkably good at teaching this)
I wonder if the author has considered that the primary applications of this work are probably not in influencing file-sharing networks so much as in politics. The P2P network that first comes to mind is ordinary web access within China. This is a situation where the government has an active interest in preventing any politically sensitive information from being propagated within the country, and so the ideas of this paper are directly applicable.
I'll leave the relevant ethical issues as a matter of discussion -- but I would suggest that this is a far more serious reason to be concerned about corporate research into network interruption.
Apart from the issue of whether any legislation whatsoever would be reasonable, I would be strongly opposed to the suggested law for a more practical reason:
There is no way to search for copyrighted material without examining the content of all traffic going across networks. Such a law would require the police to continually monitor large amounts of all network conversation, with active attention to their content, in direct contravention of all established law and precedent regarding wiretaps and other surveillance techniques. It essentially amounts to a call for blanket surveillance of the population in order to protect the copyrights of certain business interests.
Given that I have opposed such expansions of federal authority in the past, when the excuse was the (arguably much better one) of capturing terrorists, I cannot imagine countenancing such a law when the only incentive is to help some groups make a buck.
Yonatan
Hmm... so after reading this story, I'm not certain whether this is meant to be a user-installed software package, a trojan, or a remote exploit of a vulnerability in IE and Netscape.
If the former, what benefit does it claim to give the user in exchange for the obvious annoyance?
If the second, how much damage will it do to the system in the process of installation in order to make it difficult to remove, and will this damage be actionable? (I'm mentally comparing it to the story about the Celine Dion CD above...)
If the latter, how complex a firewall filter will it take to splatter this? (Since it goes along the HTML channel obviously this is much more sophisticated than packet filtering...)
If you are going the homebrew route, one case that may be worth checking out is the case from an old compact SPARC. They had very small, very dense cases which are just great for luggable applications, and quite robust. You can probably pick one up very cheap as scrap.
As a minimal solution, it isn't too hard to actually build a box to this sort of spec. For a case, start with a toolbox and hollow it out; then strap in a power supply, a small motherboard, and all the goodies. A bit of cutting work should let the ports and so on come out.
This is different from trying to build a portable or luggable since it doesn't need its own power source -- if you're doing music, you probably have access to 120VAC somewhere. So a traditional power supply can work.
A setup like this could easily come down to the $1000 price range, and open you to putting more money into a really good sound card...
The thing we would get if someone were to find a polynomial-time algorithm for any NP-complete problem is an immediate, poly-time algorithm for every NP-complete problem. This is because the definition of NP-complete is that there is a (known) poly-time algorithm to turn any one NP-complete algorithm into any other, so just by composing these two you get them all. (I'll attach a glossary at the bottom -- most people on this list probably aren't mathematicians :)
But OK, what does this mean realistically? The good news is that there are several very useful NP-complete problems; probably the best known (as someone has already mentioned) is the travelling salesman, and being able to do fast TS problems could mean incredible reductions in cost for shipping of goods and things like that. All sorts of problems in computer network architecture are also NP-complete; think about trying to design an internet which is both fault-tolerant and maximizing bandwidth.
The bad news: There are two things. First of all: This does not mean encryption of any sort is broken! The heart of public-key crypto is that factorization takes exponential time (or more specifically, the discrete logarithm, which is at the heart of fast factorization, takes exp time) and so if you could do poly-time factorization, you could break various algorithms like RSA. But factorization is only conjectured to be NP-complete; there is no proof, and in particular the explicit algorithm which would be needed to use a poly-time algorithm for some other NP-complete problem for factorization isn't known. This doesn't mean it can't be done; it just means that there's one other significant step between finding such an algorithm and breaking crypto.
The second problem is that even a poly-time algorithm isn't necessarily useful if the coefficients are large. What poly-time really means is that, in the limit of very large inputs, computation time doesn't go completely out of control; the fact that (to the best of current knowledge) factorization isn't poly is what makes adding one digit to key size enough to increase the difficulty of decryption by a factor of two. (i.e., the work increases as an exponential of the input) So this is important when you're trying to create "sufficiently large" inputs to jam up an algorithm. But for real-world problems that people are trying to solve apart from crypto, an O(N^1000) algorithm might technically be "better" than an O(e^N) algorithm but practically still be way out of reach.
In fact, most of the interesting NP-complete problems such as travelling salesman are routinely worked on by methods which give approximate answers in fairly short time; this turns out to be more than good enough for a remarkable range of uses, which means that the advance of getting poly time wouldn't be as earth-shaking for most real-world applications.
Hey, cool! (I'm a CU alum as well...) Some ideas for your project:
:) I can't think of any writers that made a huge impact in this regard between Asimov and the other golden age writers and Gibson in the mid-80's. And at that point it gets rather hard to tell what had historical significance, if only because it's so recent.
* Robotics is an obvious one -- the term itself was coined by Asimov, and the term "robot" by Capek. Asimov's collected works are definitely the most influential on this subject, to the extent that his "laws of robotics" are in fact quoted at the opening of one of the standard reference works on the subject.
* For computers in the most modern era, Gibson's _Neuromancer_ had a surprisingly deep impact; a large fraction of modern terminology (even the general usage of the word "Net") stems from there, as do many of the ways in which programmers visualized what they were trying to achieve.
* One thing that might not be so obvious is nuclear everything. There was actually a series of short stories published by various authors in the early 1940's (pre-Hiroshima!) which were remarkably technically accurate and which were being read at places like Los Alamos. Some stories on this thread:
"Nerves" by Lester del Rey -- this appeared orignally as a short story and was then expanded into a novella. The former is better written and more historically significant; you can find it in "My Favorite Science Fiction Story," edited by Greenberg.
"Solution Unsatisfactory" by Robert Heinlein, in his anthology "Expanded Universe." "Blowups Happen," by the same author and in "The Past Through Tomorrow," is about nuclear power.
OK, really this is such a broad topic that it's impossible to generate even a basic list; science fiction was so influential because the scientists read it, as children and as adults, and their notions of what sort of projects ought to be attacked were deeply shaped by this. If you go from a historical perspective, things to go for might be
* Hugo Gernsback was the first of the "great editors;" he edited some of the chief science fiction pubs back in the 20's and 30's, and was largely responsible for a vision of a technological utopia. He wrote some books of his own as well; this was all tied in heavily with things ranging from the Art Deco movement in architecture to the technological movement in Fascism. His books can be good refs. "The Jetsons" is a direct descendant of this line of writing, to give you an idea...
* There's a huge amount of "golden age" (40's-50's) SF which really shaped ideas about space travel, robotics, and nuclear energy. For this it might be best to go to anthologies from the period; the one edited by Greenberg mentioned above is good, as are any edited by John W. Campbell. (The second of the "great editors")
* The modern discussion of computers really started around the 60's, but I don't know this era as well. But from the perspective of a computer programmer, (and thus the receiving end of this cultural influx
Hope some of this helps, and good luck on your course!
Well, I'm seeing a completely different issue here, beyond other people being able to craft virii exploiting the same holes that this Magic Lantern does. (Although I'm assuming that as security holes get patched, Magic Lantern will ultimately refer to a family of virii rather than any single virus; it's going to make McAfee's job of trying to explicitly exclude it from virus searches all the more ridiculous)
The thing that occurs to me is that, back when I was an easily amused kid I used to capture computer viruses, dissect them and study them. If Magic Lantern is genuinely going to be an effective way to retreive data -- and if it's a virus designed by a team of top-level professionals, which it is likely to be, then it should be so -- then how long a matter of time is it going to be before everyone and his mad bastard cousin starts to make copies of this virus and mutate it for their own ends? This seems like it would quickly become a valuable corporate espionage tool, and then a personal espionage tool, and then just a total disaster area.
The problem with this is, if they design a powerful cracking tool which by its nature must be primarily built out of code resident on the target's machine, it's only a brief matter of time before such software and any upgrades thereof enter the mainstream of black-hat equipment.
Frankly, I'm not looking forward to script kiddies with tools like this...
From a part-time mathematician's perspective (ok, actually a physicist) this was the line that just made my jaw drop. What were they thinking?! If this text is correct, this algorithm may as well have been designed by a high-school student.
As several people have pointed out already, this is really one of the big threats of the DMCA -- that companies will go around using incredibly poor standards like this, and be immune to any pressure to improve their quality because their customers are legally forbidden to ask what they are receiving. It says a great deal about the present legal climate that anyone could get away with a mess like this cryptosystem in a commercial product.
*sigh*
All the talk I've heard about diamond IC's is for very specialized applications like radiation-hardened chips (EMP's from a tactical nuke won't fry the diamonds themselves, though the rest of the circuit may be in more trouble) or very high-temp applications where for some reason you can't put the logic somewhere else.
But I thought the major problem with diamond chips wasn't fault lines so much as the fact that you can't p-type dope diamonds; for reasons still unknown they simply boot out any such dopants. Which would make it kinda tricky to do anything useful as far as diodes or transistors. Are these people claiming any sort of innovation in that direction? Or is this result more useful for using diamond as a structural material?
Guys, this needs to go out in book form, not as an e-book or anything less. It needs to go out as a book because this is something that needs to be used in classes in high schools, and to get that done, it has to be out under a publisher's banner in dead-tree format.
Will it be hard to get this into schools? Yes, maybe. But school boards and schools have been much better in the past few years about getting this sort of thing into classes, especially English classes, under the rubric of "diversity." And in the aftermath of Columbine this book would fill a deep and urgent need that I believe even the people in decision-making positions will understand.
Will it be hard to get into print? Somewhat. But this is something we could realistically do as a community effort. There are only a few things we need. First, to get permissions from all posters and clear up the legal issues. This means we need everyone who contributed to come forth and give an OK if they want their stuff used. Second, the thing needs to be pulled together and properly formatted; John Katz's excellent introduction, plus possibly more pieces similar to that interspersed with the community contributions. Then the thing needs to be formatted nicely (not difficult; we have plenty of page-layout geeks here. And this part is important; anything well-formatted looks profound and sells much better)
Finally, and most importantly, a publisher has to be found. But the more discussion we manage to generate about the Hellmouth between now and when the formatting is done, the more likely we are to be able to convince a major house to have an interest in this book.
I would like to make a RFD: would people be willing to join in organizing this project and bringing it to completion? We can get this moving quite quickly if we all work together.
This question honestly seems a bit strange to me. The major reason that we haven't moved to something else is that there is no reasonable alternative yet out there.
There are two kinds of problem, that of choosing what "new engine" to use and the costs of converting. Since someone else has already spoken very well about the latter problem I'll just say something about the former.
What's needed to power cars is a fuel that is highly portable, (e.g. no solar panels or wind turbines) capable of producing not only large amounts of power but high impulses, (i.e. no burning pure ethanol) is reasonably safe, (no nuclear-powered rocket engines) and reasonably inexpensive. Unfortunately, that rules out virtually all of the proposals on the market.
One of the things that has killed most of these (such as electric cars) is that they fail on one of the first two criteria. Insufficient power means that the car won't zoom fast enough, and yes, that is an issue; nobody is going to buy a car that tops out at 60mph. It's simply not a practical vehicle in a society built around major roads and highways. (Maybe in Europe or Japan one has a better chance) Insufficient portability means that there's not enough range, which requires any number of Rube Goldberg schemes to work around but always comes out to meaning that the car is good for commuting between a few nearby points with refueling docks already at them. This is why you only see electric vehicles driving around on large campuses or whatever.
Fuel cells suffer from an even more serious problems: They really don't exist yet. Cells capable of powering a car, satisfying anything even remotely like the above requirements, are still a few years away at least. Fuel containment, ease of recharging, inhibition of flammability, recoverability of fuel, and high-drain performance are all still significant issues. Once those are fixed, we get to talk about conversion costs.
The one thing that is making progress is hybrid systems, which use gasoline for the highest-drain parts of driving (acceleration) and switch to electric or other motors for low-drain parts. (highways, etc.) The technology isn't simple but it's advancing rapidly, and several such vehicles are already on the road. More are almost certain to come in the near future.
Of course, software configuration means that it would require all-new drivers to work under any other OS. But provided it doesn't look too much like a food product, it may turn out to be a somewhat useful gizmo. Think of it as a mouse with multiple input senses; like having meta-keys for mouse input.
(Maybe this will give some incentive for my hands to ever leave the keyboard? Nah...)
Krasnikov's Subway is an old idea; it was written as a response to Alcubierre's warp drive article, which I think we talked about here a while ago. It is perfectly consistent as a solution to classical general relativity, but the requirement for this is an enormous (about 10^80 times the mass of the universe) amount of negative mass. There are various quantum theorems that tell us that QM prevents anything more than infinitesimal amounts of negative mass from forming, so I wouldn't bother planning any Alpha Centauri commutes aboard this subway. (BTW, the Alcubierre warp drive has very similar problems. Both of these came out in the early '90s.)
A poor man once went to his wealthier neighbor and told him, "Sir, my son is soon to be married, and his bride's parents will be coming to visit. But my house is bare and I am ashamed that they will see it like this. Could I borrow from you your silver spoons?"
The rich man considered it - he was not one to lend out his things lightly - but ultimately agreed. The next day the poor man returned with the spoons, and another small silver spoon as well.
"What is this?" the rich man asked.
"During the night one of your spoons gave birth. Since it is your spoon, the child should be yours as well."
The rich man thought this was ludicrous, but he was not one to pass up good fortune, so smiling he accepted the small spoon.
Several weeks passed, and again the poor man came to ask to borrow the spoons; remembering his previous good fortune, the rich man assented, and once again found himself presented with another spoon.
A few weeks after that, the poor man came to the rich man again, explaining that the wedding itself was coming up, and asking to borrow his silver candlesticks. At this the man hesitated; spoons, yes, but those candlesticks were pure silver! But mindful of his previous luck, and with visions of money dancing in his head, he agreed to let the poor man borrow them.
The next day the poor man came back, empty-handed.
"Where are my candlesticks?" the rich man asked.
"It is horrible! Last night they were beautiful - but this morning, I came and they were both dead!"
At this ridiculous line the rich man flew into a rage. He accused the poor man of being a thief, and soon the two came up before the town rabbi, who heard their case.
The rabbi considered it carefully, and said: "If a spoon can be born, why can a candlestick not die? If you chose to accept nonsense when it was profitable to you, you can accept the same nonsense when it brings you loss."
So it is with Amazon; as they accepted the idea of one-click patents to protect their money, they can accept the idea of sampling audio as well. Perhaps once this is finished neither side will be so foolish again - and the patent office will not encourage them.
Remember to physically unplug your computers from the network, since we are doing line maintenance that could cause your computer to explode. In fact, it's best if you wrap your computer in blankets so that if the worst should happen you won't be injured.
Well... just remember that all high-energy papers are preprints before they're published. 99% of the stuff on xxx really does get published and is meaningful science.
But as you point out, it isn't peer-reviewed, and anything that shows up there needs to be taken with a serious grain of salt until it's checked.
(1) Very possibly. But the theory itself is still cracked.
:) Most importantly, string theory can correctly predict aspects of the low-energy behavior of black holes which show up when classical GR starts to fail. (Hawking's "information paradox" is such an issue)
(2) Nonetheless, de Aquino claims that the null mass of the photon (the vanishing of its inertial rest mass) implies that it is outside the effects of gravity. I argue that the way de Aquino referred to the mass of the photon as having to do with gravity is simply wrong.
(3) This is a good question and one that bugged me for a long time. The real answer is that the graviton picture - the approximation that the gravitational field can be decomposed into small excitations which behave more or less like free particles, with some additional interactions - is only valid in the limit of weak gravitational fields. For black holes and similar strong fields, the self-interactions of the gravitons are so strong that they can no longer even approximately be thought of as particles.
This actually is fairly closely associated with why quantum theories of gravity are hard. Basically one can show that a quantum field theory of spin-2 particles is inconsistent. It reduces to something consistent in the low-energy approximation, where gravity can be thought of as the exchange of individual gravitons. But as soon as the self-interactions of gravitons becomes significant, this picture of gravitons as particles stops producing meaningful predictions.
And this is exactly the point where we have to turn to our one well-defined and reasonably well-understood theory of quantum gravity: superstring theory. While this theory is not experimentally verified, there is substantial reason to believe it will be. (Which I won't go into now
So it's possible to calculate the gravitational properties of black holes using string theory and then back away and say: If I'm far away from the black hole, it just looks like an ordinary gravitating object, gravity is weak, and the graviton approximation should apply. Indeed it does; it turns out that, from far away, the black hole looks like a uniform sphere of radius equal to its event horizon, and as far as a distant observer is concerned, it does indeed emit gravitons from this distance. But all information about the internal structure of the black hole is kept within its event horizon; we can't see inside the horizon using these gravitons.
Yonatan
OK, some physics perspective on this...
One-line summary: These guys are better at graphics than they are at physics. No, it doesn't work.
Since the discussion below is a bit more technical than I usually send to Slashdot, here's a brief summary:
* No, it doesn't work. There are in fact some very fundamental reasons why it can't be made to work, either.
* It can be shown (through some fairly messy math) that it is not possible to cancel gravity by introducing other particles, unless you somehow had negative-mass particles to strap to your spaceship.
* Even if you could do all of this, how would you hold a sphere of photons in place?!
The rest of this is a bit heavy, so you may just want to skim. If you want to get a fairly easy-to-manage and good intro to this topic, you may want to check out Rindler's _Introduction to Relativity_.
Okay, on with the show. There are three basic problems with this suggestion.
Problem 1: "Photons have a null gravitational mass." It is true that the rest mass of a photon is zero. (The rest mass of a particle is the mass of that particle as measured by an observer at rest with respect to it. Special relativity tells us that objects with nonzero rest mass must travel slower than light, and objects with zero rest mass travel exactly at the speed of light.) However, the "effective mass" of an object for gravity purposes is <i>not</i> its rest mass. The quantity which generates gravitational fields is a more complicated quantity called the stress tensor, which is nonzero any time there is energy or momentum in a system. Photons definitely have a nonzero stress tensor and as a result do produce gravity.
In fact, because of the way the stress tensor is defined, it is impossible to entirely cancel it with <i>any</i> configuration of masses, unless you somehow had something with a negative mass. If you find something like this please let the rest of us know; there are several well-known, mathematically valid, ways to design superluminal drives based on them. However there are also several results that quantum mechanics prevents the formation of more than infinitesimal quantities of such matter.
So that's problem one; photons don't have "null mass" (whatever that means) and so don't automatically cancel gravity. On to
Problem 2: If we had a shell of photons around our ship, they could exactly cancel the gravitons. Not quite. First of all, I should say that gravitons are the smallest quanta of gravity and are in many ways analogous to photons. (In the language of general relativity, where gravity is a stretching of spacetime, a graviton represents a localized deformation of spacetime. Similarly, a photon represents a local deformation of the electromagnetic field, which is the particle physics way of stating the fact that light is an electromagnetic wave.)
Now, the way gravity works (in graviton language) is that gravitons can be emitted or absorbed by anything with a nonvanishing stress tensor. (The bigger the stress tensor, the more easily this happens) So say two massive objects are moving along; Alice (having a nonzero stress) emits a graviton, and recoils by conservation of momentum. Bob (also having a nonzero stress) absorbs this graviton, and also recoils. So effectively both Alice and Bob have changed their momentum, i.e. exerted a force on one another.
Now, say I wanted to cancel this out with some shell of Mystery Particles which would give the exactly opposite forces. These particles would have to do a couple of things:
* First of all, they had better travel at the speed of light so that they arrive at the same time as the gravitons in order to cancel them out. So these had better be massless. (See note above)
* Second, they should be able to interact with any particle that has mass, and they should interact equally strongly with any two particles of the same mass. (e.g., if our mystery particle hit electrons twice as hard per unit mass as they did protons, they wouldn't just cancel the gravity, they would also knock about everybody's particles quite a bit.)
Now it turns out (and this would require some lengthy tangents to explain in detail) that these two conditions are enough to <i>uniquely</i> specify the properties of this particle! In particular, the particle must have zero mass, and (for those of you with some physics background) have spin 2 and be symmetric tensor fields. These are exactly the same as the properties of the graviton, which isn't surprising since these things have to cancel them. But...
* It's fairly straightforward to show (using a bit of field theory) that forces mediated by spin-2 symmetric particles are always attractive. (Like gravity, and as opposed to electromagnetism. EM is mediated by photons, which have spin 1 and can be either attractive [opposite charges] or repulsive. [like charges]) So unfortunately, no Mystery Particles can cancel out gravity.
OK, and just in case this isn't enough, one other point:
Problem 3: Even if you had some magic way of getting around this, e.g. introducing multiple species of graviton and giving them some extra interactions which let them exactly cancel gravity while somehow not affecting the ordinary operation of gravity in <i>any way</i>, (Which, incidentally, doesn't work; if these particles are just like gravitons, they should be produced in nature just like gravitons and so would be everywhere) How the hell are you going to keep a bunch of massless particles sitting in a perfect shell? They're <i>massless</i>. They tend to fly away.
(OK, and now a more technical note for those of you who bothered to read this far: There is actually one other possible way to satisfy the constraints for a mystery particle, allowing it to couple to matter identically to gravitons without also being a graviton. Namely, the particles can be part of a multiplet of particles which are related to gravitons by some continuous symmetry [discrete symmetries would just give multiple graviton species] and so all matter would have to couple universally to them as well. This could be done even for particles that did not have spin 2.
But some math prevents this from working. If we want our mystery particle not to have spin 2, the symmetry must relate particles of different spin. By a key result known as the Coleman-Mandula Theorem, the only way to achieve this is with a type of symmetry known as supersymmetry. If you do this, you find that it is possible to construct a complete suite of particles called a "gravity multiplet," consisting of the graviton and several particles with spins ranging from zero to 3/2, all of which couple universally to matter and do everything we wanted.
So does this solve it? Well, not quite. The condition for these supermultiplets to cancel gravity is very well-known; it's called the BPS condition, and it says that the objects exerting the forces on one another have to obey very specific relationships between their charges (electric and other) and their masses, as well as some requirements about their relative shapes and orientations. When you work it all out, there are a few very specific configurations where this happens, but (1) It doesn't happen for arbitrarily shaped objects like spacecraft, and (2) All of this would require that supersymmetry be present and unbroken in nature. This would have some <i>really</i> visible experimental consequences, e.g. atoms looking pretty much nothing like they actually do. Such a symmetry is quite definitely ruled out.
So I'm afraid that there's nothing of this sort that can cancel out gravity. Sorry...
Yonatan
Whoops - I didn't mean to Anonymous Coward that. I'll stand by my statements publicly. Yonatan Zunger
Call me a heretic of the Open Source movement, but:
I use Mathematica regularly. Its syntax is arcane only to the extent that it is itself a programming language with a complex instruction set; and the source is closed. But it has two features which I believe counter this. First, there are simply no programs of comparable power for complex symbolic manipulation; and yes, I am familiar with the open source packages. But algorithms for solving symbolic differential equations and large integrals are simply too much for small groups of people to do; their design requires substantial teams of very skilled people. And while the open source community has mustered many (most?) of the best programmers in the world, the skills of applied mathematicians simply aren't as prevalent in this world.
And second, Wolfram Research (the company which makes Mathematica) has systematically made itself as open as possible; they routinely solicit user suggestions and input, and sometimes incorporate user-submitted packages and code into their own releases. While the core code itself is compiled, a large fraction of the program comes in the form of modular packages which come in the form of Mathematica source code.
In short, I'll say that Not All Closed Source is Bad. The modularity of Mathematica, the publication of the API's and the source to all of the interpreter-level packages, and the responsiveness of the company to its users have given it most of the same advantages that true Open Source posesses.
(All of this applies as well to Maple; that system is oriented more towards large data set manipulation rather than pure symbolics, however, so the situation is slightly - but not very - different.)
So call me a heretic; but I believe that, when the cost of a large number of specialists needed to develop a package is high, the creation of a closed-source, sold-for-money package is reasonable so long as the company does not behave in a manner detrimental to its users. Therefore I would suggest that the continued use and active support of systems such as Mathematica and Maple is beneficial to the community as a whole and should be continued, even in the presence of open-source alternatives.