A curious proposition indeed. If there are no fundamental laws of physics, then discussing the scientific method is a pointless exercise since without some underlying order to the cosmos no science can be done. Perhaps you would like to change your stance to the more moderate "We do not know how many (if any) fundamental laws of physics we know at this time." I don't have to think of you as a loon then.
Does he honsestly think that we know all of physics and that there is nothing left to learn?
I sincerely doubt Dr. Park believes that. Since I can't morph into him, I'll have to speak from my own experience as a physicist. I, like many other physicists I'm sure, have a gut feeling that this will turn out to be an erroneous measurement in the end when independent research is done to validate of the results. That being said, if said validation does occur then I will happily do an about-face and dance the proverbial dance of joy at the introduction of new and interesting physics into my worldview.
Extraordinary claims have a heavier burden of proof than do mundane claims, and they demand a higher degree of skepticism. Incidentally, Dr. Park's comment appeared to me to be directed along the lines of questioning the wisdom of funding entity, throwing good money so publicly after a project that does not appear to be good science, rather than a pronouncement on the science itself. Please understand that when this project fails (I'll play the odds here), then corporate management will be less apt to fund a legitimate R&D project, and science as a whole will suffer as a consequence.
A too-open mind is a dangerous thing--you never know what will fall in. Healthy skepticism is a much safer mindset.
One word: emacs. Bind the function keys, meta-function keys, control-function keys however you like.
But then to use emacs effectively you have the task (becoming more and more difficult nowadays) of finding a keyboard with the CAPS LOCK and CONTROL keys in the right place, i.e. CONTROL above the left shift key, CAPS LOCK below it.
Incidentally, I question strongly the wisdom in the PC arrangement with the key placement reversed. CAPS LOCK is never chorded with any other keys whereas CONTROL is always chorded; why place the chorded key in an awkward location and put the unchorded (and seldom used) key in a place where it's easy to mistype? This cannot even be useful for word-processing typists given the relative ease of mistyping the CAPS LOCK when TAB or SHIFT are intended.
It's more meaningful, therefore, to think of the number of qubits in a quantum computer as its memory size, not its addressing capacity. A 30-qubit quantum computer isn't nearly analagous to a 32-bit von Neumann machine; it's much closer to one with 4 *bytes* of RAM!
I disagree with this statement. If 30 qubits are placed into a coherent superposition, then the number of accessible states is 2^30, (about 1 billion states). This exponentiation of accessible states is part of the charm of quantum computing. To quote an article from the journal Science, A register of say 1500 qubits, if it could be placed in superposition, could access more states than there are particles in the universe. (Peter Knight, Science 2000 January 21; 287: 441-442)
I'll grant, however, that the number quoted is misleading, and hundred-qubit machines will probably be needed to factor extremely large numbers: In order to do quantum error correction (i.e., making sure the bath of states doesn't lose coherence through interactions with the outside world), then each logical qubit would need to be encoded into 5 physical qubits (minimum). In essence, if Moore's Law holds, then you will have to wait another 3 years or so for such a machine.
An advantage quantum computers have over their classical counterparts is that a qubit can be in a superposition between states |1> and |0>. In principle, each qubit can be in an infinity of possible states in between the "pure" |1> and |0> states. Like any quantum mechanical wavefunctions, these qubit states can be made to interfere with each other, which is one of the features of quantum computers that is expoited in quantum algorithms. (Interference is not strictly necessary in a quantum computer calculation, but it is a useful feature of qubits that has no classical counterpart, and it illustrates some of the power of quantum computation). By cleverly packing data into wavefunctions and then interfering wavefunctions with themselves, a quantum computer can perform calculations that would be prohibitive otherwise. As an example, by taking advantage of the quantum mechanical properties of qubits Grover's algorithm allows an unordered database to be searched in O[sqrt(N)] time rather than the O[N] time required by a classical computer.
You can read some more information about the work of the Los Alamos scientists at http://www.lanl.gov/w orldview/news/releases/archive/00-041.html. Curiously, Moore's Law seems to hold for quantum computers as well, since it was nearly 18 months since the same researchers intoduced the first 3 qubit quantum computer (using nuclear magnetic resonance and a trichloroethylene molecule). To quote the article: Of course, if Moore's Law is at work here," Laflamme added, "then we could have a 30-qubit quantum computer in less than five years." A 30 qubit machine could perform certain tasks (such as Shor's algorithm or a variant for factoring large primes) many times faster than even the most powerful present-day supercomputers.
Most likely the high speed networking hardware they refer to are not just an off-the-shelf 100baseTX switch and a bunch of ethernet cards. If I'm not mistaken, the servers themselves have very fast communications among processors inside them; the special networking software quite likely takes advantage of the network topology, and some problems (including the test problem they ran, I'd guess) may be well-suited for this type of machine. Communications latency is the bugaboo for high speed computing, so this may explain why they were apparently able to do so much with so little.
I agree with your comment that this is not necessarily entirely Boeing's fault, due to the ever-changing funding environment in which they are forced to operate, however I would counter that they are at least partially to blame for the error since the part in question was their responsibility. I'm not saying that the line workers are necessarily to blame, but the corporation, at some echelon, must accept a measure of the responsibility.
Sadly, shrinking budgets and forced reorganization seem to be endemic to large R&D projects. The vision that initiates a project seldom carries through to its fruition, and so the environment changes. In a perfect world if an organization is unable to fulfill a function within budgetary constraints, then management needs to have the spine to address their clients and inform them that they can't deliver a quality product under said constraints. This never happens in real life, unfortunately, and so to keep the cash cow alive corners get cut, errors get made, and projects which already suffer from adverse public opinion (which led to the shrinking budgets in the first place) experience failures to compound the problem. Just look at the failed Mars probes and the NIF facility at Lawrence Livermore National Labratory as cases in point.
[Note to moderators: Detritus's followup to my original post is more informative than my post, so if anyone has two points to spare, if you could moderate mine down a point and move Detritus's up a point, I'd be greatly appreciative. Thanks.]
Unless Boeing comes public with a pledge that they'll absorb the cost, the perceived effect on the taxpaying community will be that the taxpayer gets to absorb the overrun resulting from Boeing's egregious mistake, and that Boeing suffer's a mere moment's bad PR. Perhaps the rest of/. finds this to be high comedy, but I'm of the mind that the space station and the space program in general suffer enough from public image woes as it is.
I'm looking forward to the day that the public looks upon our ailing space program (and, by extension, nationally funded R&D) as something more than an enormous public works project. No amount of positive spin can undo the damage caused by a handful of silly mistakes such as this.
Wow, someone who knows what they're talking about....
I deceive people well.:)
If I understand correctly, you are describing how a "surface-to-volume" ratio goes up as the volume elements get larger, thus allowing individual processors to spend more time crunching numbers and less time waiting for boundary data. This is indeed true, and this is precisely the kind of balancing act one has to perform to compute efficiently in parallel. As you've demonstrated, the same algorithms may be more efficient on some machines than on others, but based on my (albeit limited) experience in computational physics, optimization almost always seems to boil down to how one reduces the number of messages that have to be passed in order to perform the task. This seems to be the single most important factor in the scalability of numerical calculations (how much speedup is gained by increasing the number of processors).
Disclaimer: While I have some experience in parallel computing, I am by no means an expert in this field, and I suggest you read some of the other excellent posts in this thread to hear from the real experts.
If you ever try simulating a complex physical system on a computer, you will quickly realize that without some experimental data it would be nearly impossible to reliably model the physics. In the case of nuclear weapons (among the most complicated physical systems known) computer simulation results can only be verified through testing.
To produce nuclear weapons, a state needs a program incorporating both testing and computer simulations (the U.S. approach), or else a program of testing and espionage (the U.S.S.R. approach). In either case, testing is a necessary ingredient. Moore's law pretty much assures that restricting a state's ability to perform simulations will not work, despite the U.S.'s best efforts to the contrary, so attention should probably be paid to restricting nations' abilities to test nuclear weapons and reducing their ability to acquire through espionage data which can only be obtained through testing.
Regarding ASCI, if weapons designers had complete confidence in ASCI, then the U.S. would have signed the CTBT. I think the general consensus among weapons designers is that while ASCI may replace testing in the future, it is not now at the stage of being able to replace testing, and thus the U.S. wants to keep the option of testing open in order to verify the reliability of the stockpile sometime in the future if compelling justification arises.
Open source nuclear secrets would be a Very Bad Thing [tm]. I think even Open Source zealots would admit that some information should never be released for public consumption.
I do not speak for my employer. These comments are mine alone.
Many problems do not parallelize well. For instance, to my admittedly limited knowledge no parallel version of the fast Fourier transform algorithm (which serves as the backbone of many spectral and pseudospectral codes) is known which does not require a prohibitive amount of interprocessor communications.
At the risk of being overly pedestrian, let me try tackling your differential equations question: Communications latency issues can crop up even if you have just a single equation to solve. Let's imagine, for the sake of discussion, that you wish to understand the propagation of heat on a metal plate, and you have a differential equation that describes the process. Conceptually, you might imagine solving this problem on a parallel computer by breaking the metal plate into a bunch of smaller regions, and asking each processor to compute heat flow on an individual region, as in the following, where a plate is broken into 9 regions:
OOO OOO OOO
You can see that every region borders other regions, and herein lies the difficulty: To compute how heat propagates in any one region, say the top left corner, you have to have information from each of the neighboring regions. (In the case of the top-left corner, it'd be the center-left and top-center domains). In solving differential equations, this information is called the "boundary conditions." Each time step would require sending a considerable amount of information among the processors in order to handle the boundary conditions. To use a real-world analogy, if communication latency is high, then many of the processors will end up waiting for the information they need, much like workers in a bureaucracy that has an inefficient internal mail service. In "real" supercomputers you pay big bucks for fast communications, and problems that are communications-intensive will naturally perform better on these machines than on Beowulf or Appleseed clusters. Alternatively, problems whose algorithms require few messages to be passed among processors (many Monte Carlo algorithms have this feature) may run very efficiently on a Beowulf or Appleseed cluster, where communications latency is high.
Parallel computing seems to be largely an exercise in economics. Any parallel algorithm with a nonzero number of messages to be passed will necessarily run at something less than 100% efficiency. Just how far below 100% depends on the nature of the algorithm and the machine/cluster it is running on.
Discovering 90% of what is known the universe is unconsequential if you can't feel it or touch it.
Answering a long-standing problem in astrophysics is not inconsequential in my opinion. Unlike you, many, including myself, do have a healthy interest in how the universe works, and when a problem such as this appears as if it may be answered (I'm skeptical, and I'll wait to read their paper and see some supporting evidence before I'm convinced), we understandably take interest.
An experimental/observational validation of supersymmetry would itself be a momentous achievement, even apart from the context of dark matter, as it would have a significant impact on how we understand the physical world at its most fundamental level.
I do hope it's a major break thorugh and produces something useful in every day life.
I'm of the opinion that a measure of our scientific inquiry and discovery should remain separated from the incessant "what good is it?" questions. It's bad enough that most such science can't get funded anymore, that the politics of acquiring research funds requires, typically, a "product" after a year or two [often at the expense of doing a quality job of understanding the science], that only "tried and true," conservative problems seem to get attention, and that much of the business of doing science has become entrenched in rehashing dogma and maintaining the status quo.
Sometimes human beings do things that aren't necessarily useful in everyday life, but yet enrich the human experience somehow. If you've ever posted to Slashdot, played a video game, read a novel, looked at a pretty painting, attended an opera, or cogitated your place in this universe, you might appreciate this point. It's somewhat unfortunate that science has been recognized by the layperson primarily for its commercial, and not its philosophical, value.
My apologies for the long-winded response to your post.
CMEs move at about 500 to 1000 km/sec. This gives a time of about 2 days or so between when the eruption occurs on the solar surface and when the CME plasma would arrive at earth.
Primarily ionized hydrogen, a bit of helium (a few % or so if I recall, but I haven't worked in the field for years) and also trace amounts of other species, like oxygen, iron, calcium, etc. The blob of stuff emitted is a relatively dense plasma, as compared to the solar wind plasma.
It should be noted that unless you are an astronaut en route to the moon or mars, the danger to most satellites and terrestrial systems isn't the CME, per se, but rather the geomagnetic storms that sometimes are triggered by CMEs. Generally speaking, a CME will trigger a substorm when it impinges on the earth's magnetosphere over a long period (more than a couple hours or so) with a southward pointing interplanetary magnetic field (IMF). These solar wind conditions are said to be geoeffective, and they can trigger massive changes in the earth's magnetosphere, changes which can have deleterious consequences in systems we depend on, such as power grids and communications satellites. CMEs aren't required for geoeffective conditions to be satisfied (patches of fast solar wind with the right orientation of the magnetic field can trigger substorms as well), but they are responsible for the largest geomagnetic storms.
As a matter of fact, we do have more spacecraft in place this solar maximum (SOHO, Yohkoh, ACE, and others) with which to monitor the inner heliosphere to help determine when coronal mass ejections are on their way. Unfortunately, the causes of coronal mass ejections are not well understood, and we have only a limited ability to predict when such storms will arise. Space Weather, including the prediction of geomagnetic storms [substorms] believed to result from the blob of plasma emitted in a CME striking the earth's magnetosphere with the right magnetic field alignment, has been an active area of interest in space science in the past decade.
Protecting spacecraft and terrestrial systems such as power grids requires first that we have a reliable predictive ability; the economic impact (and strategic impact, if you are in the military) of a "false positive" is high. Recent work in this field is encouraging, however, and I suggest the following site at the Naval Research Laboratory for more information: http://wwwppd.nrl.navy.mi l/whatsnew/prediction/index.html.
It is heartening to hear of the reasonableness of your schedule in the "skilled computer types" (SCT) business. Unfortunately, as a postdoctoral physicist in the USA I feel I have no such luxury. The perception among most of the physics postdocs I know is that anything short of 12 hour days (we're paid for 8) and working at least one day each weekend is slacking. Aside from our having more formal traning than the SCTs, the only significant differences I can see between our two fields, at least in this country, is the relative scarcity of career positions among physicists and the culture of "work until you drop."
In your estimation, is the level of demand for CSTs the primary factor that contributes to their empowerment in the work force in setting reasonable work hours, or is the difference cultural as well? At the risk of generalizing excessively, most Europeans I know seem to lead more balanced lives than we do here in the colonies, and I'd speculate that cultural factors may weigh more heavily than the level of demand in your particular case. I'm curious to hear what your thoughts are on this matter, and what you'd consider to be the most significant factor in your enviable freedom.
At the risk of oversimplifying the problem, the central issue with fusion has historically been how one goes about solving the "n * tau * T" problem. Namely, you have to have a high enough density "n" for a long enough time "tau" at high enough energies/temperatures "T". When this product is large enough you can get burning fusion. The challenge is that while you can indeed control plasmas well enough to get a large "tau" (e.g. stick the plasma in a magnetic bottle like a mirror device, stellerator, or tokamak), getting large enough "n T" is challenging. High densities, which give you many collisions and thus many chances at initiating fusion reactions, and high temperatures, which raise the fusion cross section considerably, lead to high plasma pressures. These plasma pressures push on the walls of the magnetic bottle and can cause the bottles to "bow out" (i.e. initiate pressure driving instabilities such as ballooning in tokamaks), which then wreck your "tau." (There are other technical issues as well, such as turbulence in the plasma which enhances cross-field transport and thus makes for leaky bottles, current-driven instabilities which can cause a catastrophic loss of confinement, etc. But I digress....) Z-pinch devices (and inertial confinement fusion schemes such as the National Ignition Facility) try to get burning fusion by just not bothering to optimize "tau," but rather just compress the plasma really hard and really fast. This cranks up "n T" to very high values and makes your "n tau T" large. It is a serious technical challenge how one might go about making these implosive events occur regularly and reproducibly enough to accommodate a fusion reactor setting.
I was an instructor for some time at UCLA. Please allow me to respond to a couple points raised here:
a) Horseapples. UCLA is a public institution in name only, really. It recieves only a minute amount of its annual revenue from the state, with the vast majority of its funding arriving from private sources, from endowments, from skimming research grants, and, yes, from royalties on their intellectual property. UCLA and Cal Berk. have both proposed removing themselves from the UC system since they would be better off financially if they were to become formally private institutions rather than just effective private institutions.
b) The professors do not own the rights to the material they generate; that's the province of the University of California. Since the University has the intention of selling this intellectual property in the future (in the form of satellite or distance classes, web-based classes, etc.), they are understandably concerned with protecting their capital. It makes good business sense, and good business at the university often translates into lower tuition costs, better facilities, and better services.
To use an analogy (though not a perfect analogy) many Slashdotters may appreciate, if I take an MCSE course and I compose a set of notes with essentially the same content, the same stucture, the same pedegogy as the course I took, do I then have the right to market my new "course" without the permission of the source I obtained the information from? If my source is a well-established and well-recognized provider of such information, do I have the right to use their name to advertise my own product without their consent?
c) It is a common practice, at least among many of the physics faculty at UCLA, to provide to the students copies of their lecture notes. Generally this is done at no charge to the student. While I personally will not contend with a student's being permitted to sit through my course, dutifully take notes of what I say, and then distribute the notes as they see fit, I do strongly question their "right" to collect the handouts I provide and sell photocopies to note-taking services. I find the practice to be particularly odious among the many professional note-taking institutions; they routinely hire students who are not even enrolled in my course to attend and grab the handouts. Then the organization sells photocopies of said handouts verbatim without giving credit to author (me in this case) or the institution that financed the compilation of the notes (The University of California). Even so, it is of questionable legality and ethics for me to allow this blatant use of taxpayers' dollars to finance commercial enterprises; my permitting it could cause the University or me to be sued, and I could lose my job.
Incidentally, I find it curious that someone with a college education believes that if someone's political alignment (...and we all know that every university professor is a card-carrying socialist--it's in the by-laws....) does not match that of the surrounding community, then they must of necessity be "out of touch." This is the kind of dogmatic ranting I'd expect from children, not allegedly educated adults. Perhaps you need new friends?
-------------------- Disclaimer: The opinions stated in this post are my own and in no way reflect or resemble (save by coincidence) those of the University of California, the Department of Energy, or the Los Alamos National Laboratory.
Perhaps you are aware that on the next census no entry for "farmer" will be listed under occupation. (Paul Harvey radio show, Oct 1 I believe). There simply aren't enough farmers left to justify the wasted ink/space on the forms, apparently.
As an aside, I'm surprised at how many on here have similar backgrounds--who grew up on a farm, chose not to farm not because of the hard lifestyle, but rather because of the hopelessness of it. (Myself included: Iowa farm, corn/soybeans/beef/cattle. My family got out of the business after the mid-80s-Reagonomics farm crisis).
I spent many summers in my youth pulling weeds out of soybean fields, and I'd like to suggest another advantage to having seeds which do not reproduce: if you rotate your crops, then you don't have to deal with seeds left over from this year's harvest being the next year's weeds.
I'd advise you to go through your house and rid yourself of your smoke alarms for safety's sake. You know those things have radioactive materials inside them. Not only that, the radioisotope in there was manufactured inside a nuclear reactor! As you are no doubt aware, nuclear reactors are very very bad, as we found out this last week. Much worse then nuclear testing, which we seem to think is nice and safe since it's done in the desert. Oh, and on your way out, please be sure you avoid breaking any "Exit" signs--you'll release more tritium into your precious Long Island environment than the incident at Brookhaven.
(In all honesty, your drinking water is probably more messed up by industrial dumping over the years than it ever was from the Brookhaven Lab's modest program).
The space station is largely irrlelvant for science as a whole; much of its capacity for making a contribution to anything beyond being a "feel-good, high-tech, public works project" has been compromised in the interest of cutting costs and getting it up into the sky.
The real embarrassment in terms of scientific spending is how NASA, the DOE, and NSF must compete with the VA, HUD, and Americorps this fiscal year in a zero-sum-game for appropriations. The latter three have highly vocal constituencies, and will almost certainly gain significant amounts of revenue which would otherwise be used for R&D spending. Given that 50 percent of the US's GDP since the end of World War II has been the result of scientific and technical innovation (70 percent over the past few years), it seems ill-advised to accept such massive cuts in scientific spending.
Federal support of R&D is now only about 45% of what it was 30 years ago. This trend is unlikely to reverse itself anytime soon without the involvement of the (largely apathetic) scientific and technical constituencies.
Perhaps we could all devote 1% of the time we spend bashing Microsoft, Apple Co., intellectual property and patent law, GNOME/KDE, RHS, the NSA, AOL, the DOJ, and Sun Co., and instead write an informed letter to a congressperson or two in support of scientific and technical spending in this country. This would do much to give the impression that some are indeed concerned about these issues.
A Wash ington Post editorial by Allan Bromley, a former president of the American Physical Society, makes a compelling case for increasing science appropriations.
(My apologies for the non-USA readers for this USA-centric post).
Statistics classes which teach Gaussian distributions are fine provided students learn what is required for the central limit theorem to apply. I think you may be mistaken regarding the "noninfinite" part. We can quantify the fluctuations from a Gaussian law in a rigorous manner in cases where a noninfinite number of variables are concerned. The real difficulty, on the other hand, is that people sometimes apply the CLT when it is not valid.
You should add "...approaches infinity, provided the fractional contribution from any one random variable to the sum uniformly converges to zero in the limit as N -> infinity." This is an important distinction. For instance, Levy distributions are a class of stable limit laws for which this is not the case--the largest variable in the sum can in fact dominate the sum. Symmetric Levy distributions may superficially resemble Gaussian laws, but with tails that decay slower (like power laws rather than exponentially fast).
This article is amusing if only because it is a nostalgic throwback to the days of P.R. and hype over "chaos theory." Call me dense, but I don't understand why something as simple as scale-invariance needs to be dressed in the extra jargon and hype. Assuming the author did not miss anything terribly fundamental, I don't see anything novel in what was reported. Perhaps someone in the know can fill me in on just how exactly this turns statistical physics on its head?
[For those who are interested, Levy distributions are treated quite adequately in Limit Distributions for Sums of Independent Random Variables (Gnedenko and Kolmogorov) (c)1954].
> We've only been able to observe the ozone layer > for about 50 years. what kind of idiot > 'scientists' jumped to the conclusion we were > causing the ozone fluctuations? It's likely been > going on for 1000s of years!
Ah, you called our bluff. You see, all of us scientists are "idiots" professionally and are not to be trusted in matters having to do with so-called scientific issues. We do however manage to play a very sophisticated political game. All alleged science research since Newton has been a big hoax, syphoning off many billions of dollars in R&D money for some nefarious purpose.
What purpose, you ask? I leave that for you to uncover. I'm sure your rapier acumen (and I must commend you on your very illuminating and scientific comparison of ACs in pickup trucks and SUVs) will undoubtedly lead you to the truth.
Just don't try to stop us--we know where you live.
Slashdot Mirror
There are no fundamental laws of physics.
A curious proposition indeed. If there are no fundamental laws of physics, then discussing the scientific method is a pointless exercise since without some underlying order to the cosmos no science can be done. Perhaps you would like to change your stance to the more moderate "We do not know how many (if any) fundamental laws of physics we know at this time." I don't have to think of you as a loon then.
Does he honsestly think that we know all of physics and that there is nothing left to learn?
I sincerely doubt Dr. Park believes that. Since I can't morph into him, I'll have to speak from my own experience as a physicist. I, like many other physicists I'm sure, have a gut feeling that this will turn out to be an erroneous measurement in the end when independent research is done to validate of the results. That being said, if said validation does occur then I will happily do an about-face and dance the proverbial dance of joy at the introduction of new and interesting physics into my worldview.
Extraordinary claims have a heavier burden of proof than do mundane claims, and they demand a higher degree of skepticism. Incidentally, Dr. Park's comment appeared to me to be directed along the lines of questioning the wisdom of funding entity, throwing good money so publicly after a project that does not appear to be good science, rather than a pronouncement on the science itself. Please understand that when this project fails (I'll play the odds here), then corporate management will be less apt to fund a legitimate R&D project, and science as a whole will suffer as a consequence.
A too-open mind is a dangerous thing--you never know what will fall in. Healthy skepticism is a much safer mindset.
One word: emacs. Bind the function keys, meta-function keys, control-function keys however you like.
But then to use emacs effectively you have the task (becoming more and more difficult nowadays) of finding a keyboard with the CAPS LOCK and CONTROL keys in the right place, i.e. CONTROL above the left shift key, CAPS LOCK below it.
Incidentally, I question strongly the wisdom in the PC arrangement with the key placement reversed. CAPS LOCK is never chorded with any other keys whereas CONTROL is always chorded; why place the chorded key in an awkward location and put the unchorded (and seldom used) key in a place where it's easy to mistype? This cannot even be useful for word-processing typists given the relative ease of mistyping the CAPS LOCK when TAB or SHIFT are intended.
It's more meaningful, therefore, to think of the number of qubits in a quantum computer as its memory size, not its addressing capacity. A 30-qubit quantum computer isn't nearly analagous to a 32-bit von Neumann machine; it's much closer to one with 4 *bytes* of RAM!
I disagree with this statement. If 30 qubits are placed into a coherent superposition, then the number of accessible states is 2^30, (about 1 billion states). This exponentiation of accessible states is part of the charm of quantum computing. To quote an article from the journal Science, A register of say 1500 qubits, if it could be placed in superposition, could access more states than there are particles in the universe. (Peter Knight, Science 2000 January 21; 287: 441-442)
I'll grant, however, that the number quoted is misleading, and hundred-qubit machines will probably be needed to factor extremely large numbers: In order to do quantum error correction (i.e., making sure the bath of states doesn't lose coherence through interactions with the outside world), then each logical qubit would need to be encoded into 5 physical qubits (minimum). In essence, if Moore's Law holds, then you will have to wait another 3 years or so for such a machine.
Haha--so much for the alleged proofread that I did of my post. As you so adroitly pointed out, it should be "factor large number."
Perhaps you can sell me a deep hole in which to stash my growing embarrassment?
An advantage quantum computers have over their classical counterparts is that a qubit can be in a superposition between states |1> and |0>. In principle, each qubit can be in an infinity of possible states in between the "pure" |1> and |0> states. Like any quantum mechanical wavefunctions, these qubit states can be made to interfere with each other, which is one of the features of quantum computers that is expoited in quantum algorithms. (Interference is not strictly necessary in a quantum computer calculation, but it is a useful feature of qubits that has no classical counterpart, and it illustrates some of the power of quantum computation). By cleverly packing data into wavefunctions and then interfering wavefunctions with themselves, a quantum computer can perform calculations that would be prohibitive otherwise. As an example, by taking advantage of the quantum mechanical properties of qubits Grover's algorithm allows an unordered database to be searched in O[sqrt(N)] time rather than the O[N] time required by a classical computer.
You can read some more information about the work of the Los Alamos scientists at http://www.lanl.gov/w orldview/news/releases/archive/00-041.html. Curiously, Moore's Law seems to hold for quantum computers as well, since it was nearly 18 months since the same researchers intoduced the first 3 qubit quantum computer (using nuclear magnetic resonance and a trichloroethylene molecule). To quote the article: Of course, if Moore's Law is at work here," Laflamme added, "then we could have a 30-qubit quantum computer in less than five years." A 30 qubit machine could perform certain tasks (such as Shor's algorithm or a variant for factoring large primes) many times faster than even the most powerful present-day supercomputers.
Most likely the high speed networking hardware they refer to are not just an off-the-shelf 100baseTX switch and a bunch of ethernet cards. If I'm not mistaken, the servers themselves have very fast communications among processors inside them; the special networking software quite likely takes advantage of the network topology, and some problems (including the test problem they ran, I'd guess) may be well-suited for this type of machine. Communications latency is the bugaboo for high speed computing, so this may explain why they were apparently able to do so much with so little.
I agree with your comment that this is not necessarily entirely Boeing's fault, due to the ever-changing funding environment in which they are forced to operate, however I would counter that they are at least partially to blame for the error since the part in question was their responsibility. I'm not saying that the line workers are necessarily to blame, but the corporation, at some echelon, must accept a measure of the responsibility.
Sadly, shrinking budgets and forced reorganization seem to be endemic to large R&D projects. The vision that initiates a project seldom carries through to its fruition, and so the environment changes. In a perfect world if an organization is unable to fulfill a function within budgetary constraints, then management needs to have the spine to address their clients and inform them that they can't deliver a quality product under said constraints. This never happens in real life, unfortunately, and so to keep the cash cow alive corners get cut, errors get made, and projects which already suffer from adverse public opinion (which led to the shrinking budgets in the first place) experience failures to compound the problem. Just look at the failed Mars probes and the NIF facility at Lawrence Livermore National Labratory as cases in point.
[Note to moderators: Detritus's followup to my original post is more informative than my post, so if anyone has two points to spare, if you could moderate mine down a point and move Detritus's up a point, I'd be greatly appreciative. Thanks.]
Unless Boeing comes public with a pledge that they'll absorb the cost, the perceived effect on the taxpaying community will be that the taxpayer gets to absorb the overrun resulting from Boeing's egregious mistake, and that Boeing suffer's a mere moment's bad PR. Perhaps the rest of /. finds this to be high comedy, but I'm of the mind that the space station and the space program in general suffer enough from public image woes as it is.
I'm looking forward to the day that the public looks upon our ailing space program (and, by extension, nationally funded R&D) as something more than an enormous public works project. No amount of positive spin can undo the damage caused by a handful of silly mistakes such as this.
Wow, someone who knows what they're talking about....
:)
I deceive people well.
If I understand correctly, you are describing how a "surface-to-volume" ratio goes up as the volume elements get larger, thus allowing individual processors to spend more time crunching numbers and less time waiting for boundary data. This is indeed true, and this is precisely the kind of balancing act one has to perform to compute efficiently in parallel. As you've demonstrated, the same algorithms may be more efficient on some machines than on others, but based on my (albeit limited) experience in computational physics, optimization almost always seems to boil down to how one reduces the number of messages that have to be passed in order to perform the task. This seems to be the single most important factor in the scalability of numerical calculations (how much speedup is gained by increasing the number of processors).
Disclaimer: While I have some experience in parallel computing, I am by no means an expert in this field, and I suggest you read some of the other excellent posts in this thread to hear from the real experts.
If you ever try simulating a complex physical system on a computer, you will quickly realize that without some experimental data it would be nearly impossible to reliably model the physics. In the case of nuclear weapons (among the most complicated physical systems known) computer simulation results can only be verified through testing.
To produce nuclear weapons, a state needs a program incorporating both testing and computer simulations (the U.S. approach), or else a program of testing and espionage (the U.S.S.R. approach). In either case, testing is a necessary ingredient. Moore's law pretty much assures that restricting a state's ability to perform simulations will not work, despite the U.S.'s best efforts to the contrary, so attention should probably be paid to restricting nations' abilities to test nuclear weapons and reducing their ability to acquire through espionage data which can only be obtained through testing.
Regarding ASCI, if weapons designers had complete confidence in ASCI, then the U.S. would have signed the CTBT. I think the general consensus among weapons designers is that while ASCI may replace testing in the future, it is not now at the stage of being able to replace testing, and thus the U.S. wants to keep the option of testing open in order to verify the reliability of the stockpile sometime in the future if compelling justification arises.
Open source nuclear secrets would be a Very Bad Thing [tm]. I think even Open Source zealots would admit that some information should never be released for public consumption.
I do not speak for my employer. These comments are mine alone.
Many problems do not parallelize well. For instance, to my admittedly limited knowledge no parallel version of the fast Fourier transform algorithm (which serves as the backbone of many spectral and pseudospectral codes) is known which does not require a prohibitive amount of interprocessor communications.
At the risk of being overly pedestrian, let me try tackling your differential equations question: Communications latency issues can crop up even if you have just a single equation to solve. Let's imagine, for the sake of discussion, that you wish to understand the propagation of heat on a metal plate, and you have a differential equation that describes the process. Conceptually, you might imagine solving this problem on a parallel computer by breaking the metal plate into a bunch of smaller regions, and asking each processor to compute heat flow on an individual region, as in the following, where a plate is broken into 9 regions:
OOO
OOO
OOO
You can see that every region borders other regions, and herein lies the difficulty: To compute how heat propagates in any one region, say the top left corner, you have to have information from each of the neighboring regions. (In the case of the top-left corner, it'd be the center-left and top-center domains). In solving differential equations, this information is called the "boundary conditions." Each time step would require sending a considerable amount of information among the processors in order to handle the boundary conditions. To use a real-world analogy, if communication latency is high, then many of the processors will end up waiting for the information they need, much like workers in a bureaucracy that has an inefficient internal mail service. In "real" supercomputers you pay big bucks for fast communications, and problems that are communications-intensive will naturally perform better on these machines than on Beowulf or Appleseed clusters. Alternatively, problems whose algorithms require few messages to be passed among processors (many Monte Carlo algorithms have this feature) may run very efficiently on a Beowulf or Appleseed cluster, where communications latency is high.
Parallel computing seems to be largely an exercise in economics. Any parallel algorithm with a nonzero number of messages to be passed will necessarily run at something less than 100% efficiency. Just how far below 100% depends on the nature of the algorithm and the machine/cluster it is running on.
Discovering 90% of what is known the universe is unconsequential if you can't feel it or touch it.
Answering a long-standing problem in astrophysics is not inconsequential in my opinion. Unlike you, many, including myself, do have a healthy interest in how the universe works, and when a problem such as this appears as if it may be answered (I'm skeptical, and I'll wait to read their paper and see some supporting evidence before I'm convinced), we understandably take interest.
An experimental/observational validation of supersymmetry would itself be a momentous achievement, even apart from the context of dark matter, as it would have a significant impact on how we understand the physical world at its most fundamental level.
I do hope it's a major break thorugh and produces something useful in every day life.
I'm of the opinion that a measure of our scientific inquiry and discovery should remain separated from the incessant "what good is it?" questions. It's bad enough that most such science can't get funded anymore, that the politics of acquiring research funds requires, typically, a "product" after a year or two [often at the expense of doing a quality job of understanding the science], that only "tried and true," conservative problems seem to get attention, and that much of the business of doing science has become entrenched in rehashing dogma and maintaining the status quo.
Sometimes human beings do things that aren't necessarily useful in everyday life, but yet enrich the human experience somehow. If you've ever posted to Slashdot, played a video game, read a novel, looked at a pretty painting, attended an opera, or cogitated your place in this universe, you might appreciate this point. It's somewhat unfortunate that science has been recognized by the layperson primarily for its commercial, and not its philosophical, value.
My apologies for the long-winded response to your post.
CMEs move at about 500 to 1000 km/sec. This gives a time of about 2 days or so between when the eruption occurs on the solar surface and when the CME plasma would arrive at earth.
Primarily ionized hydrogen, a bit of helium (a few % or so if I recall, but I haven't worked in the field for years) and also trace amounts of other species, like oxygen, iron, calcium, etc. The blob of stuff emitted is a relatively dense plasma, as compared to the solar wind plasma.
It should be noted that unless you are an astronaut en route to the moon or mars, the danger to most satellites and terrestrial systems isn't the CME, per se, but rather the geomagnetic storms that sometimes are triggered by CMEs. Generally speaking, a CME will trigger a substorm when it impinges on the earth's magnetosphere over a long period (more than a couple hours or so) with a southward pointing interplanetary magnetic field (IMF). These solar wind conditions are said to be geoeffective, and they can trigger massive changes in the earth's magnetosphere, changes which can have deleterious consequences in systems we depend on, such as power grids and communications satellites. CMEs aren't required for geoeffective conditions to be satisfied (patches of fast solar wind with the right orientation of the magnetic field can trigger substorms as well), but they are responsible for the largest geomagnetic storms.
As a matter of fact, we do have more spacecraft in place this solar maximum (SOHO, Yohkoh, ACE, and others) with which to monitor the inner heliosphere to help determine when coronal mass ejections are on their way. Unfortunately, the causes of coronal mass ejections are not well understood, and we have only a limited ability to predict when such storms will arise. Space Weather, including the prediction of geomagnetic storms [substorms] believed to result from the blob of plasma emitted in a CME striking the earth's magnetosphere with the right magnetic field alignment, has been an active area of interest in space science in the past decade.
Protecting spacecraft and terrestrial systems such as power grids requires first that we have a reliable predictive ability; the economic impact (and strategic impact, if you are in the military) of a "false positive" is high. Recent work in this field is encouraging, however, and I suggest the following site at the Naval Research Laboratory for more information: http://wwwppd.nrl.navy.mi l/whatsnew/prediction/index.html.
It is heartening to hear of the reasonableness of your schedule in the "skilled computer types" (SCT) business. Unfortunately, as a postdoctoral physicist in the USA I feel I have no such luxury. The perception among most of the physics postdocs I know is that anything short of 12 hour days (we're paid for 8) and working at least one day each weekend is slacking. Aside from our having more formal traning than the SCTs, the only significant differences I can see between our two fields, at least in this country, is the relative scarcity of career positions among physicists and the culture of "work until you drop."
In your estimation, is the level of demand for CSTs the primary factor that contributes to their empowerment in the work force in setting reasonable work hours, or is the difference cultural as well? At the risk of generalizing excessively, most Europeans I know seem to lead more balanced lives than we do here in the colonies, and I'd speculate that cultural factors may weigh more heavily than the level of demand in your particular case. I'm curious to hear what your thoughts are on this matter, and what you'd consider to be the most significant factor in your enviable freedom.
At the risk of oversimplifying the problem, the central issue with fusion has historically been how one goes about solving the "n * tau * T" problem. Namely, you have to have a high enough density "n" for a long enough time "tau" at high enough energies/temperatures "T". When this product is large enough you can get burning fusion. The challenge is that while you can indeed control plasmas well enough to get a large "tau" (e.g. stick the plasma in a magnetic bottle like a mirror device, stellerator, or tokamak), getting large enough "n T" is challenging. High densities, which give you many collisions and thus many chances at initiating fusion reactions, and high temperatures, which raise the fusion cross section considerably, lead to high plasma pressures. These plasma pressures push on the walls of the magnetic bottle and can cause the bottles to "bow out" (i.e. initiate pressure driving instabilities such as ballooning in tokamaks), which then wreck your "tau." (There are other technical issues as well, such as turbulence in the plasma which enhances cross-field transport and thus makes for leaky bottles, current-driven instabilities which can cause a catastrophic loss of confinement, etc. But I digress....) Z-pinch devices (and inertial confinement fusion schemes such as the National Ignition Facility) try to get burning fusion by just not bothering to optimize "tau," but rather just compress the plasma really hard and really fast. This cranks up "n T" to very high values and makes your "n tau T" large. It is a serious technical challenge how one might go about making these implosive events occur regularly and reproducibly enough to accommodate a fusion reactor setting.
I was an instructor for some time at UCLA. Please allow me to respond to a couple points raised here:
a) Horseapples. UCLA is a public institution in name only, really. It recieves only a minute amount of its annual revenue from the state, with the vast majority of its funding arriving from private sources, from endowments, from skimming research grants, and, yes, from royalties on their intellectual property. UCLA and Cal Berk. have both proposed removing themselves from the UC system since they would be better off financially if they were to become formally private institutions rather than just effective private institutions.
b) The professors do not own the rights to the material they generate; that's the province of the University of California. Since the University has the intention of selling this intellectual property in the future (in the form of satellite or distance classes, web-based classes, etc.), they are understandably concerned with protecting their capital. It makes good business sense, and good business at the university often translates into lower tuition costs, better facilities, and better services.
To use an analogy (though not a perfect analogy) many Slashdotters may appreciate, if I take an MCSE course and I compose a set of notes with essentially the same content, the same stucture, the same pedegogy as the course I took, do I then have the right to market my new "course" without the permission of the source I obtained the information from? If my source is a well-established and well-recognized provider of such information, do I have the right to use their name to advertise my own product without their consent?
c) It is a common practice, at least among many of the physics faculty at UCLA, to provide to the students copies of their lecture notes. Generally this is done at no charge to the student. While I personally will not contend with a student's being permitted to sit through my course, dutifully take notes of what I say, and then distribute the notes as they see fit, I do strongly question their "right" to collect the handouts I provide and sell photocopies to note-taking services. I find the practice to be particularly odious among the many professional note-taking institutions; they routinely hire students who are not even enrolled in my course to attend and grab the handouts. Then the organization sells photocopies of said handouts verbatim without giving credit to author (me in this case) or the institution that financed the compilation of the notes (The University of California). Even so, it is of questionable legality and ethics for me to allow this blatant use of taxpayers' dollars to finance commercial enterprises; my permitting it could cause the University or me to be sued, and I could lose my job.
Incidentally, I find it curious that someone with a college education believes that if someone's political alignment (...and we all know that every university professor is a card-carrying socialist--it's in the by-laws....) does not match that of the surrounding community, then they must of necessity be "out of touch." This is the kind of dogmatic ranting I'd expect from children, not allegedly educated adults. Perhaps you need new friends?
--------------------
Disclaimer: The opinions stated in this post are my own and in no way reflect or resemble (save by coincidence) those of the University of California, the Department of Energy, or the Los Alamos National Laboratory.
Perhaps you are aware that on the next census no entry for "farmer" will be listed under occupation. (Paul Harvey radio show, Oct 1 I believe). There simply aren't enough farmers left to justify the wasted ink/space on the forms, apparently.
As an aside, I'm surprised at how many on here have similar backgrounds--who grew up on a farm, chose not to farm not because of the hard lifestyle, but rather because of the hopelessness of it. (Myself included: Iowa farm, corn/soybeans/beef/cattle. My family got out of the business after the mid-80s-Reagonomics farm crisis).
I spent many summers in my youth pulling weeds out of soybean fields, and I'd like to suggest another advantage to having seeds which do not reproduce: if you rotate your crops, then you don't have to deal with seeds left over from this year's harvest being the next year's weeds.
Just a thought.
I'd advise you to go through your house and rid yourself of your smoke alarms for safety's sake. You know those things have radioactive materials inside them. Not only that, the radioisotope in there was manufactured inside a nuclear reactor! As you are no doubt aware, nuclear reactors are very very bad, as we found out this last week. Much worse then nuclear testing, which we seem to think is nice and safe since it's done in the desert. Oh, and on your way out, please be sure you avoid breaking any "Exit" signs--you'll release more tritium into your precious Long Island environment than the incident at Brookhaven.
(In all honesty, your drinking water is probably more messed up by industrial dumping over the years than it ever was from the Brookhaven Lab's modest program).
The space station is largely irrlelvant for science as a whole; much of its capacity for making a contribution to anything beyond being a "feel-good, high-tech, public works project" has been compromised in the interest of cutting costs and getting it up into the sky.
The real embarrassment in terms of scientific spending is how NASA, the DOE, and NSF must compete with the VA, HUD, and Americorps this fiscal year in a zero-sum-game for appropriations. The latter three have highly vocal constituencies, and will almost certainly gain significant amounts of revenue which would otherwise be used for R&D spending. Given that 50 percent of the US's GDP since the end of World War II has been the result of scientific and technical innovation (70 percent over the past few years), it seems ill-advised to accept such massive cuts in scientific spending.
Federal support of R&D is now only about 45% of what it was 30 years ago. This trend is unlikely to reverse itself anytime soon without the involvement of the (largely apathetic) scientific and technical constituencies.
Perhaps we could all devote 1% of the time we spend bashing Microsoft, Apple Co., intellectual property and patent law, GNOME/KDE, RHS, the NSA, AOL, the DOJ, and Sun Co., and instead write an informed letter to a congressperson or two in support of scientific and technical spending in this country. This would do much to give the impression that some are indeed concerned about these issues.
A Wash ington Post editorial by Allan Bromley, a former president of the American Physical Society, makes a compelling case for increasing science appropriations.
(My apologies for the non-USA readers for this USA-centric post).
Statistics classes which teach Gaussian distributions are fine provided students learn what is required for the central limit theorem to apply. I think you may be mistaken regarding the "noninfinite" part. We can quantify the fluctuations from a Gaussian law in a rigorous manner in cases where a noninfinite number of variables are concerned. The real difficulty, on the other hand, is that people sometimes apply the CLT when it is not valid.
You should add "...approaches infinity, provided the fractional contribution from any one random variable to the sum uniformly converges to zero in the limit as N -> infinity." This is an important distinction. For instance, Levy distributions are a class of stable limit laws for which this is not the case--the largest variable in the sum can in fact dominate the sum. Symmetric Levy distributions may superficially resemble Gaussian laws, but with tails that decay slower (like power laws rather than exponentially fast).
This article is amusing if only because it is a nostalgic throwback to the days of P.R. and hype over "chaos theory." Call me dense, but I don't understand why something as simple as scale-invariance needs to be dressed in the extra jargon and hype. Assuming the author did not miss anything terribly fundamental, I don't see anything novel in what was reported. Perhaps someone in the know can fill me in on just how exactly this turns statistical physics on its head?
[For those who are interested, Levy distributions are treated quite adequately in Limit Distributions for Sums of Independent Random Variables (Gnedenko and Kolmogorov) (c)1954].
Did the pickup come with a gun rack?
> We've only been able to observe the ozone layer
> for about 50 years. what kind of idiot
> 'scientists' jumped to the conclusion we were
> causing the ozone fluctuations? It's likely been
> going on for 1000s of years!
Ah, you called our bluff. You see, all of us scientists are "idiots" professionally and are not to be trusted in matters having to do with so-called scientific issues. We do however manage to play a very sophisticated political game. All alleged science research since Newton has been a big hoax, syphoning off many billions of dollars in R&D money for some nefarious purpose.
What purpose, you ask? I leave that for you to uncover. I'm sure your rapier acumen (and I must commend you on your very illuminating and scientific comparison of ACs in pickup trucks and SUVs) will undoubtedly lead you to the truth.
Just don't try to stop us--we know where you live.