Not really, but your logic makes about as much sense as the conspiracy theorists. Just because one idiot who works for the government screwed up, doesn't imply anything about other people, and other agencies? Why would it?
Although I agree the logic of the argument doesn't work, the poster's point was more subtle. Yes, everyone agrees when circumstances are right the "government" would rather pull a fast one on us instead of look bad. However, the sheer scale and intricacy of many proposed conspiracy theories require entire civilian agencies to work like so many well-oiled, evil organizations from a James Bond movie rather than the behemoths of bureaucratic inefficiency that they actually are. Not to mention, *every* single employee of said organizations involved in such conspiracies (which by sheer scale alone require tens of thousands of highly technically proficient personnel to pull off) would have to be prevented from intentionally or inadvertently leaking any the juicy details for the last 40 years. I mean, a military of paramilitary organization might have a chance at keeping a big secret, but we're talking about agencies like NASA here.
If you want to have a good laugh, screen the movie "capricorn one" with a bunch of NASA employees (it involves a vast NASA conspiracy complete with armed thugs and a secret sound-state in the desert, starring O.J. Simpson as an astronaut!!!) Amusingly, the latter is slightly more believable than the former if you've ever dealt with the NASA bureaucracy.
So not only would doctors have to get light inside the brain, they'd first have to genetically engineer the neurons to include and express the halorhodopsin gene. The right neurons: the ones that will later have Parkinson's Disease or whatever is being treated.
Right on about the bottleneck being the genetic engineering bit. But the whole advantage of the light-triggered shutdown is precisely that you could introduce it into an entire swath of neurons, and then only shut down the misfiring neurons (and you can reactivate them if you make a mistake by taking the light away). Otherwise, with a working gene delivery system, you could in principle just try to fix the faulty genes themselves - while bearing the risk of messing up already healthy cells.
Existing treatments for Epilepsy are very invasive, involving open brain surgery, recording an actual seizure, and then burning out (irreversibly!) the neurons that start the seizure. If you burn the wrong place, or too large a spot, you could permanently alter brain functionality. Advanced Parkingsons treatment also involves implanting electrodes deep into the brain , except they use a small tunable electrode grid so that they can optimize/refine the area it is stimulating without repeated invasive surgeries.
Bingo, that's exactly right. While I do agree that this is an extremely clever approach, it has the same Achilles heel as any other genetic treatment, i.e. the lack of any viable, working delivery system. Putting in new genetic material typically involves some sort of stem-cell graft for regenerative tissues (i.e. bone marrow) or some sort of modified virus delivery system. Since neurons don't regenerate, that leaves option 2 (directly injecting a virus into the brain) which is, well, a pretty bad idea right now considering the disasters that have occured in most gene therapy trials to date.
The bright side is, if a safe and effective delivery system were invented, you wouldn't need to worry as much about delivery to non-target tissues (i.e. the neurons that are working perfectly) since a doctor could reversibly choose and optimize which specific areas to shut down, they way they do now with impanted electrodes for Parkingson's patients.
No, wbren is right. The actual performance numbers for the current F@H ps3 build should be similar to late model PPC mac since the Pande group
has simply recompiled the gromacs core to run on the Cell PPE (that is the general purpose core of CELL). NONE of the 8 vector processors are being utilized, which are the source for the speculative performance ratios in TFA. When there is sufficient coding expertise in Cell to optimize for all the processors, then it might be significantly faster than a PC, but not before. . . I'm sure the Pande group has this in the works but it won't be ready for the PS3 launch. I think the upstream gromacs developers are working on Cell optimizations but they've mentioned its very, very difficult compared to programming for any other architecture they have come across (these are people who wrote ASSEMBLY inner loops for every major workstation architecture).
see
http://www.mail-archive.com/gmx-users@gromacs.org/ msg03335.html
robotkid
But, what about hypothesizing certain criteria for the state of a folded protein? Such as... minimization of free energies, auto/cross correlation functions (pair distribution functions) going to zero (likely shell values), ergodicity in ensemble of states (time average = ensemble average). These are all indications of a system gone to equilibrium. Now, that's not necessarily the folded state..but it is a stable state of the protein (of which the folded state might be the dominant one).
All good ideas in principle but very difficult in practice. Directly calculating free energies for something as large as a protein is probably even harder than folding a protein because now you have to have good sampling of all possible paths to fold, not just one. You also need to know what the unfolded ensemble (i.e. starting states) are and that's an open question right now.
Ergodicity is also one of the hardest things to quantify - and with Pande's approach he's already banked on ergodicity allowing him to stitch loads of tiny simulations into one long composite simulation so one can't double dip and test for ergodicity after having assumed ergodicity.
The Pande group shoot 100,000s of trajectories through space state space to calculate those rates. Most of those trajectories do not go in the right direction, but some do. It's certainly possible to start testing some of the criteria listed above and see which ones tell you which directions are the "right ones" (and then see if those criteria are universal, of course).
I assure you the Pande group is mining their data as much as is feasible. Unfortunately most of the assumptions needed to make these computations distributable also make the data only useful for the exact quantity they decide to interrogate, it's just the nature of the beast.
I don't see anything in the article saying that this is the completed product. In fact, the second paragraph states
"The simulation pushes today's computing power to the limit. But it is only a first step. In future researchers hope that bigger, longer simulations will reveal details about how viruses invade cells and cause disease."
This sounds like it is meant more as a proof of concept, not a complete simulation.
Noone is accusing them of lying, it's just that this is a field where there are more "proof of concept" simulations than actual, scientifically insightful simulations. Between the "look we're pushing the envelope" papers (which get into sexy publications) and "look we did a simulation without crashing" papers (which this would be, BTW, if the system was 10x smaller) there's a little too much "hoping that bigger and longer simulations will lead to insight".
The Pande group does great work. . but to be precise they can calculate
the folding rate of small, two-state folders. They still need to know what
the folded structure is before they even start so that they know when something is heading in the right direction. This is different then actually predicting the folded structure.
An analogous statement would have been made about the manpower needed to sequence a genome back in the 70's. With current models, yes it might take that much sand. But that just means our models are broken, and an infinite amount of computing power will still not make up for lack of new scientific insight on the problem.
So there's a tree called the Copaifera that grows in brazil that, when tapped, produces 40+ liters of biodiesel sap a year with minimal processing. Of course, growing our own rainforest is probably not a very efficient or practical way to solve our energy needs.
If, however, the pathway of genes responsible for creating the sap can be isolated and cloned into a plant more suitable for crop farming, that would be mind-bogglingly cool. So in that sense some serious biotech tinkering is certainly in order.
But in terms of reverse engineering how a plant actually does photosynthesis, plant physiologists and biophysicists have been doing this for years and basically it appears to be so complex and highly tuned it's unlikely we can outdesign billions of years of selection for efficiency.
That said it's so tempting to think about optimizing it (at least for me). The enzyme that actually does the carbon fixation is called RuBisCo, arguably the most important enzyme on the entire planet, and its a really, really lousy enzyme. Its a protein of molecular weight ~5,000,000 which can process a lousy 3 CO2 molecules per second. The only reason plants can grow at all is by brute force, something like 60% of the dry weight of a plant is rubisco. If we could make it just a bit more efficient . . .
It is interesting to note the biochemistry of carbon fixation was discovered by the nobel laureate Melvin Calvin (which is why it's called the calvin cycle). He was also one of the pioneer researchers of Copaifera trees as a source of biodiesel, and dreamed of splicing the genes into weeds to solve our dependence on foreign oil. Sadly I've seen very little work published about this since his death in 1997.
Unfortunately, the missing context is that the top 50 billboard songs are chosen along very similar algorithms, so it's hardly surprising (talking about tracking audience response, not the sound itself). A modern clearchannel radio station doesnt' even use a cd-player. All the songs they need for an entire month fit on a small harddrive, and what songs get played and how often is automated in response to instant surveys and "focus groups" (the same type of info this program tracks). They say a radio station owner can't even go to the loo without checking if that's supported by the most recent ratings.
Alan
Sidestepping the threadmongous ID vs evolution debate geomon has kicked up, I actually beg to differ on his interpretation of why science is declining. Biology is as healthy as ever in the states, with the NIH extramural grant system supporting the lions share of the nations academic biomedical research. It's the basic sciences that are hurting most, but not because of the religious right (unless you're talking about stem cells). It's the fact that if you can't get on the NIH gravy train, ALL of the basic sciences combined then have to duke it out for NSF funds, funds which are constantly being cut and only add up to a few percent of the NIH budget. The pickings are really slim for my friends in geology or physics all the way to the aforementioned fields of evolutionary biology.
In physics the loss of the superconducting supercollider and the successful bid for next-generation accelerators and fusion reactors to be in Europe have pretty much dashed the hopes of an entire generation of american physicists. In chemistry, most schools are struggling to recruit ANY american students alt all to fill their graduate rosters because demand is so low. A hostile climate for foreign graduate students (visas tend to expire now if you sneeze on them) coupled with the fact that most foreign nationals view their dream job as going home to their mother countries (with newly expanding science budgets) should make it obvious the days of the US as being the undisputed scientific leader are numbered if not gone already.
Not that a little competition isn't good for us, certainly the pace of progress in the developing world is a good thing. But combine the fact that we're scaring off foreign talent from staying, and that our native talent is dwindling fast (or choosing to go into IT or IS where they can actually have a decent chance at getting a job) this is a recipe for disaster. It will only take a generation or two before the dropoff in basic science really starts to hurt our applied sciences, and at that point the science infrastructure will be beyond rescuscitation.
A final problem is that the collapse of the soviet union displaced a tremendous amount of talent trained in the hard sciences. The research associate/visiting professor market is completely flooded with tens of thousands of PhD's from eastern europe who are trying to make ends meet. Inevitably, hiring decisions have to be made between young, unproven U.S. scientists and a senior, experienced expatriate scientists who need to feed their children. All from the same piece of ever-shrinking pie. Everyone looses.
Well, while Mac is technically correct on all accounts, couching things in terms of "does or does not" violate laws of physics is tricky at best. Just because there are suggestive explanations that exist in the land of emiment theorists doesn't change the fact that it's all incredibly, incredibly improbable and there's a disconnect between sci-fi's claim "this will be as easy as buying milk someday" and today's reality which estimates these things to be more probable than hell freezing over but slightly less probable than pigs flying.
I'm not saying I demand my sci-fi to be nonfiction, mind you, but once you loose the impact of the "what if" causing a flash of thoughtfulness to the audience you have to resort to things like changing nose appendages or poorly scripted love interests to keep them watching.
I'm not a soothsayer and can't say for certain that we won't have things like this someday, but I guarantee if I told any of my physics professors I was going to do a thesis investigating the plausibility of warp drives I would find myself in the loony bin thaaaaat fast..
I certainly agree everyone shouldn't be taught as if they were physicists. And engineers do care about completely different things than physicists, granted.
But they way it is now, they don't even speak each other's language.
I guess my problem is that physicists, especially in academia, are encouraged to pursue things with no obvious practical use. Excessive attention to practicality is viewed as pandering for funding. Engineers, on the other hand, are taught the complete opposite. ..practicality is king.
Some of the worst examples of this come from NASA. Engineers trying to design space probes to specs written by scientists who have no idea how to write specs. . the engineers do their job and optimize something that completely erases the scientific value of the data. Of course, add 3 layers of management separating every scientist from an engineer and you've got absolute bedlam. And people wonder why they go over budget.
As for "teaching only as much as a major needs", that's the exact problem right there. How does a junior physics professor who studies general relativity have any idea at all what type of physics a neurobiologist is going to find useful? He doesn't, so they learn how balls bounce instead of how an MRI works. That's a big disconnect. I also never understood why future doctors need to know organic synthesis. But that's what they get taught. And it perpetuates itself, the pigeonholing of academic interests because then you can have a whole field "all to yourself" so you can get tenure. Great.
I once heard Richard Ernst (nobel laureate) speak about how it is a crime when academia sees their only real educational duties as the nurturing of future academics to someday replace them. I couldn't agree more.
Have you watched it? It's classic Kurosawa, BTW, with Toshiro Mifune as the lead.
There's the defeated clan whose only hope lies in smuggling their princess to friendly territory. She is guarded by an old general dressed incognito (help me obi wan!). The comic relief is slapstick from a duo of clueless peasants. . who agree to help smuggle the strangers for money...one tall and skinny, the other short and fat . ..always bickering.
At one point, they get captured but the enemy leader recognizes the general and says he would have liked to face him on the battlefield. So instead he calls off his troops and they have a one on one spearfight. (the fight choreography is strikingly similar to a light-saber fight. FYI many of the actors were trained by real samurai, the very last of their kind).
Darth vader's helmet is very similar to a samurai helmet, especially when they wear the facemask and slatted armor.
I'm not saying Lucas didn't add alot of his own. But the influence is striking. Not quite as blatant as "magnificent seven" vs "the seven samurai" but still there.
I've digressed really far. I was trying to assert that given Lucas' fondness for Kurosawa, any perceived "cowboy movie" influence is more likely to be samurai movie influence. That's all.
As for what is called "science" in science fiction, that completely depends on your goal (star wars is practically fantasy but we still like it fine). A. Clarke always maintained that suffiently advanced technology is indistinguishable from magic. You could take the smartest neanderthal ever and probably never be able to convince it that a VCR and TV are anything other than magic. So while there is value in making the science believable for stories set in the near-future, it's pretty ridiculous to even pretend we can imagine what the distant future will be like. So the choice is to choose "hard" science fiction so that we can consider society today and how slight changes might affect it, or we can erase all the rules and just make it plain old fiction and try to stand on the quality of the writing without worrying about "scientific" merit (Dune qualifies as a good example here - the author was genuinely trying to do more than just entertain the reader but without "hard" explanations).
A friend of mine (whom I'm pretty sure only had theoretical knowledge of women) often liked to quip (in a yoda voice)
"social life. . heh. ..women...heh. ..a physicist craves not these things..."
I think he went on to say econ majors were the dark side of the force but he was usually drowned out by hysteric laughter at this point.
I have to tell you, though, it doesn't help that most of the women in physics I knew were completely asexual. Except for the one diamond in the rough that was an avid video gamer, RPG freak, complete sci-fi dork, and I have no doubt she making some lucky geek very happy right now:-)
Centrifugal "force" and the Coriolis "force" are not real forces because in both cases you would see it's just inertia and nothing more if you were observing from the proper inertial frame . It just looks like an extra force when we make the assumption that our rotating Earth is an inertial frame or that the merry-go-round we are on is an inertial frame (both cases are rotating which means they are being accellerated and are therefore improper frames).
So the short answer is, we understand those "forces" as much as we understand inertia. . the "pseudo" part is just an artifact of its a real pain in the ass to have to deal with the earth's rotation in every little computation we do - so like the lazy bums we are, we just ignore it if we can get away with it.
I wholeheartedly agree. I think in gradeschool the emphasis is always on learning facts instead of concepts especially where science is concerned. . . I'm certain we'd have alot more interest in science if we could convey just how much we don't know instead of giving the impression that everything's been worked out and it's all unimportant minutiae you could always look up in the encyclopedia if you were really desperate. . .
I was floored on my first day of intro physics when the professor told us it's just an assumption that intertial mass was the same as gravitational mass. It's been shown to 10 decimal places or so but not formally proven (I think it stems from the same inability to describe gravity that the original post was about).
My college profs were always very good about pointing out what we do and don't know. my only beef there is that only the physics majors get the good profs who tell us this stuff. . someone who was taking physics for engineers or physics for biologists probably ended up doing alot of arithmetic about bouncing balls without ever being told the parts that are actually interesting.
I do realize, however, that's is much harder to teach a subject to an auditorium full of people taking the subject for a requirement than a small group of nerds who already have proven commitment given that of their own free will they signed up for masochistic science course 666. I wish schools would try and find professors who are good at teaching non-majors instead of relegating that duty to whichever disgruntled professor is on the department's naughty list.
There actually was a chemistry 666 in my school, it was advanced synthetic organic chemistry I think:-)
The reason why this is interesting is not because
"oh look, gravity still works". Most physicists have no doubt that whenever you test it, gravity will still work. In that sense, it is kindof what another poster said, "everything's normal, big deal!"
But what's eating all the theorists is that they have absolutely no idea why. The venerable laws of gravitation are empirical, in the sense that noone knows where it comes from other than the fact that it is associated with mass. All the other forces of nature have a quantum explanation, and have a particle that transmits them (most notably electromagnetism and photons). Noone has been able to satisfactorily reconcile gravity with any fundamental (quantum mechanical) nature of a particle.
It's almost scary that we know more about what binds subatomic particles together than what keeps the moon orbiting the earth. It's also ironic that most people's only introduction to physics is newtonian physics which is presented in textbooks as complete and understood. It's true we have the math to predict the effect of gravity to arbitrary precision, but I'm sure engineers can back me up that just because something has a robust empirical law doesn't mean anyone really understands how it works.
I've been following this debate, alot of high-profile journals are full of opinion articles on this. But here's some context for all those too lazy to RTFA:
ACS publishes "SCIFINDER" and "CHEMICAL ABSTRACTS" which is NOT a database of journal papers. It is a database of chemical structures and properties which is invaluable to any research chemist because a) most information on exotic compounds are not published in journals and b) even if they were you'd have a very hard time searching journals for occurences of chemical compounds.
(FYI most of this info was probably NOT gained through tax-funded research, it's mostly industrial)
Just about every chemistry lab in the nation has to pay a subscription for this service, but it obviously requires many, many curators to keep up to date because of the crazy amount of info out there. ACS is a nonprofit organization and it uses the proceeds to fund things like scientific meetings and putting every journal article from the last 100 years online (they are way ahead of most journals that only have 5 or 10 years online).
That being said it is strange that they are so vehemently against an NIH database which is primary geared towards biological compounds (i.e. proteins and nucleic acids and derivatives) which is pretty orthogonal to most of the chemical research world. But it would be a gross oversimplification to paint the ACS as an evil money grubbing organization.
Besides, chemists are rarely evil. Science fiction proves it's always an overweight doctor come-geneticist played by marlon brando that's evil.
There is plenty of written science-fiction with as much or as little science in it as you want - I think your generalization about modern sci-fi not having science only applies to TV series and big-budget movies.
There is always something wrong with being arrogant, agreed, but there's nothing wrong with liking science fiction better than any other genre.
But I do think you are missing out on a crucial observation: real scientists love science fiction. Especially physicists, we crave that stuff, the good , the bad, and sometimes the ugly (I've also met more than a few chemists and engineers who feel the same way). Why? Because if we didn't believe that ass-boring and ass-hard equations could somehow have a big effect in the long run then we are just a bunch of masochistic idealists or incredibly underpaid technicians:-)
I think the focus on the "hardness" of sci-fi is completely misplaced. Granted, there is a divide between science fantasy (say, star wars) and there's science fiction (say 2001). But ultimately, other than being fun escapism (which I wholeheartedly endorse BTW) the function of science fiction has always been to explore the societal impacts of technology. It's not good science fiction just because the author is a PhD and spends an entire chapter lecturing me on a spaceship propulsion system that adds nothing to the plot (although that is ever so slightly better than an author who in real life couldn't build a model rocket from a kit who cribs something he read in scientific american). It's good science fiction if it makes you think hard about something you never thought about before. A good science fiction author immerses you in a world that you, by the end of the book, could actually believe exisited in some future history book. Technology is not a plot gimmick or a crutch but something that brings a new aspect of humanity under the magnifying glass, and none of us would be even reading this news site if we weren't fascinated by how society is shaped by technology, right?
My 2 cents
(and don't get me wrong, I do like hard-science fiction if its done well, but being a graduate student in science, I have to read enough hokey articles obfuscated by irrelevant technical details in modern journals as it is!)
That was one of the original motivations for the ISS.
(to server as a refueling station/staging point). In fact, it's pretty much the only reason we signed up to build it. Buti t's quickly become a moot point because the ISS has been scaled back so many times due to budget cuts that the 2 crewmen there have to spend their entire efforts focused on not dying. I'm afraid the ISS is all show and no substance, they had to cut out any potential for innovative science when they were cutting it's budget. And yet it continues to drain valuable $$ that NASA could really use right now.
Ehh, NASA's been shaking themselves up ever since the end of the cold war and it hasn't done much except change the logos and buzzwords. All their cold war talent is either retired, dead, or went to industry. Hiring new talent, well, they contract out everything. I'm not sure the concept of "lowest bidder" is even sound when there are only two mega-aerospace contractors left, and whoever wins the contract subcontracts out to the other guy.
I worked at a NASA center briefly as a computer technician. The high level bureaucrats had plush carpeting, receptionists, potted palms trees, and $20,000 sparc workstations to check their email with. While the engineers in the basement actually designing a *&(* satellites would come and asked me what type of computer I could acquire for them with their discretionary fund of 50$. I scrounged around and they were the new, proud owners of a 386sx with a yellow screen (monochrome!) that noone else wanted (I'm dating myself here).
There was plenty of lower middle management that had engineering backgrounds, which you'd think would be a good thing (i.e. being able to communicate with subordinates and translate to superiors, etc). Thing was, ones I met were ex-engineers who were lucky enough to be civil servants. If their project failed miserably, but not so miserably to cause an inquiry, they got reassigned to lower management. Civil servants have tenure and are hard to fire, so obviously contractors are preferred. Which means all your talent leaves the moment a project ends and then the next project doing the exact same thing has to re-learn everything the last guys figured out.
It also doesn't help that between the space station and the space shuttle (neither of which can be lowered any more due to politics), and salaries for 10 levels of bureaucracy, thats already 90% of the entire budget. AFAIK the only thing keeping any science alive is the Jet Propulsion Lab. Caltech is still attracting young talent away from industry instead of vice versa.
Well, there's a big difference between "useful for limited sets of problems" and useful to biology in general. I'll point out you don't even NEED molecular dynamics if all you wanted was a static h-bonding pattern, steric clashes, and distance contraints. Rasmol and a structure file will do fine. The force fields are getting better but with the current machinery in place, we can never even hope in our wildest dreams that MD is going to tell us the mechanistic details of catalysis or protein-protein recognition if you just load a structure and hit "go". We still have to do a boatload of biochemistry, alot of mutational analysis (with hypothesis as deep as "this part looks important so lets kill it"). The "promise" of accurate simulations on the same timescale as the actual biochemical and (gasp) biological events the macromolecules perform is still a pipe dream.
Slashdot Mirror
Although I agree the logic of the argument doesn't work, the poster's point was more subtle. Yes, everyone agrees when circumstances are right the "government" would rather pull a fast one on us instead of look bad. However, the sheer scale and intricacy of many proposed conspiracy theories require entire civilian agencies to work like so many well-oiled, evil organizations from a James Bond movie rather than the behemoths of bureaucratic inefficiency that they actually are. Not to mention, *every* single employee of said organizations involved in such conspiracies (which by sheer scale alone require tens of thousands of highly technically proficient personnel to pull off) would have to be prevented from intentionally or inadvertently leaking any the juicy details for the last 40 years. I mean, a military of paramilitary organization might have a chance at keeping a big secret, but we're talking about agencies like NASA here.
If you want to have a good laugh, screen the movie "capricorn one" with a bunch of NASA employees (it involves a vast NASA conspiracy complete with armed thugs and a secret sound-state in the desert, starring O.J. Simpson as an astronaut!!!) Amusingly, the latter is slightly more believable than the former if you've ever dealt with the NASA bureaucracy.
Right on about the bottleneck being the genetic engineering bit. But the whole advantage of the light-triggered shutdown is precisely that you could introduce it into an entire swath of neurons, and then only shut down the misfiring neurons (and you can reactivate them if you make a mistake by taking the light away). Otherwise, with a working gene delivery system, you could in principle just try to fix the faulty genes themselves - while bearing the risk of messing up already healthy cells.
Existing treatments for Epilepsy are very invasive, involving open brain surgery, recording an actual seizure, and then burning out (irreversibly!) the neurons that start the seizure. If you burn the wrong place, or too large a spot, you could permanently alter brain functionality. Advanced Parkingsons treatment also involves implanting electrodes deep into the brain , except they use a small tunable electrode grid so that they can optimize/refine the area it is stimulating without repeated invasive surgeries.Bingo, that's exactly right. While I do agree that this is an extremely clever approach, it has the same Achilles heel as any other genetic treatment, i.e. the lack of any viable, working delivery system. Putting in new genetic material typically involves some sort of stem-cell graft for regenerative tissues (i.e. bone marrow) or some sort of modified virus delivery system. Since neurons don't regenerate, that leaves option 2 (directly injecting a virus into the brain) which is, well, a pretty bad idea right now considering the disasters that have occured in most gene therapy trials to date.
The bright side is, if a safe and effective delivery system were invented, you wouldn't need to worry as much about delivery to non-target tissues (i.e. the neurons that are working perfectly) since a doctor could reversibly choose and optimize which specific areas to shut down, they way they do now with impanted electrodes for Parkingson's patients.
Well, maybe he was really uploading a windows install disk to the aliens . . .
No, wbren is right. The actual performance numbers for the current F@H ps3 build should be similar to late model PPC mac since the Pande group has simply recompiled the gromacs core to run on the Cell PPE (that is the general purpose core of CELL). NONE of the 8 vector processors are being utilized, which are the source for the speculative performance ratios in TFA. When there is sufficient coding expertise in Cell to optimize for all the processors, then it might be significantly faster than a PC, but not before. . . I'm sure the Pande group has this in the works but it won't be ready for the PS3 launch. I think the upstream gromacs developers are working on Cell optimizations but they've mentioned its very, very difficult compared to programming for any other architecture they have come across (these are people who wrote ASSEMBLY inner loops for every major workstation architecture). see http://www.mail-archive.com/gmx-users@gromacs.org/ msg03335.html
robotkid
Ergodicity is also one of the hardest things to quantify - and with Pande's approach he's already banked on ergodicity allowing him to stitch loads of tiny simulations into one long composite simulation so one can't double dip and test for ergodicity after having assumed ergodicity.
I assure you the Pande group is mining their data as much as is feasible. Unfortunately most of the assumptions needed to make these computations distributable also make the data only useful for the exact quantity they decide to interrogate, it's just the nature of the beast.To paraphrase southpark:
step 1) Run longer, bigger simulation.
step 2) . . . .
step 3) Insight!!
The Pande group does great work. . but to be precise they can calculate the folding rate of small, two-state folders. They still need to know what the folded structure is before they even start so that they know when something is heading in the right direction. This is different then actually predicting the folded structure.
An analogous statement would have been made about the manpower needed to sequence a genome back in the 70's. With current models, yes it might take that much sand. But that just means our models are broken, and an infinite amount of computing power will still not make up for lack of new scientific insight on the problem.
If, however, the pathway of genes responsible for creating the sap can be isolated and cloned into a plant more suitable for crop farming, that would be mind-bogglingly cool. So in that sense some serious biotech tinkering is certainly in order.
But in terms of reverse engineering how a plant actually does photosynthesis, plant physiologists and biophysicists have been doing this for years and basically it appears to be so complex and highly tuned it's unlikely we can outdesign billions of years of selection for efficiency.
That said it's so tempting to think about optimizing it (at least for me). The enzyme that actually does the carbon fixation is called RuBisCo, arguably the most important enzyme on the entire planet, and its a really, really lousy enzyme. Its a protein of molecular weight ~5,000,000 which can process a lousy 3 CO2 molecules per second. The only reason plants can grow at all is by brute force, something like 60% of the dry weight of a plant is rubisco. If we could make it just a bit more efficient . . .
It is interesting to note the biochemistry of carbon fixation was discovered by the nobel laureate Melvin Calvin (which is why it's called the calvin cycle). He was also one of the pioneer researchers of Copaifera trees as a source of biodiesel, and dreamed of splicing the genes into weeds to solve our dependence on foreign oil. Sadly I've seen very little work published about this since his death in 1997.
Unfortunately, the missing context is that the top 50 billboard songs are chosen along very similar algorithms, so it's hardly surprising (talking about tracking audience response, not the sound itself). A modern clearchannel radio station doesnt' even use a cd-player. All the songs they need for an entire month fit on a small harddrive, and what songs get played and how often is automated in response to instant surveys and "focus groups" (the same type of info this program tracks). They say a radio station owner can't even go to the loo without checking if that's supported by the most recent ratings. Alan
In physics the loss of the superconducting supercollider and the successful bid for next-generation accelerators and fusion reactors to be in Europe have pretty much dashed the hopes of an entire generation of american physicists. In chemistry, most schools are struggling to recruit ANY american students alt all to fill their graduate rosters because demand is so low. A hostile climate for foreign graduate students (visas tend to expire now if you sneeze on them) coupled with the fact that most foreign nationals view their dream job as going home to their mother countries (with newly expanding science budgets) should make it obvious the days of the US as being the undisputed scientific leader are numbered if not gone already.
Not that a little competition isn't good for us, certainly the pace of progress in the developing world is a good thing. But combine the fact that we're scaring off foreign talent from staying, and that our native talent is dwindling fast (or choosing to go into IT or IS where they can actually have a decent chance at getting a job) this is a recipe for disaster. It will only take a generation or two before the dropoff in basic science really starts to hurt our applied sciences, and at that point the science infrastructure will be beyond rescuscitation.
A final problem is that the collapse of the soviet union displaced a tremendous amount of talent trained in the hard sciences. The research associate/visiting professor market is completely flooded with tens of thousands of PhD's from eastern europe who are trying to make ends meet. Inevitably, hiring decisions have to be made between young, unproven U.S. scientists and a senior, experienced expatriate scientists who need to feed their children. All from the same piece of ever-shrinking pie. Everyone looses.
I'm not saying I demand my sci-fi to be nonfiction, mind you, but once you loose the impact of the "what if" causing a flash of thoughtfulness to the audience you have to resort to things like changing nose appendages or poorly scripted love interests to keep them watching. I'm not a soothsayer and can't say for certain that we won't have things like this someday, but I guarantee if I told any of my physics professors I was going to do a thesis investigating the plausibility of warp drives I would find myself in the loony bin thaaaaat fast..
I guess my problem is that physicists, especially in academia, are encouraged to pursue things with no obvious practical use. Excessive attention to practicality is viewed as pandering for funding. Engineers, on the other hand, are taught the complete opposite. . .practicality is king.
Some of the worst examples of this come from NASA. Engineers trying to design space probes to specs written by scientists who have no idea how to write specs. . the engineers do their job and optimize something that completely erases the scientific value of the data. Of course, add 3 layers of management separating every scientist from an engineer and you've got absolute bedlam. And people wonder why they go over budget.
As for "teaching only as much as a major needs", that's the exact problem right there. How does a junior physics professor who studies general relativity have any idea at all what type of physics a neurobiologist is going to find useful? He doesn't, so they learn how balls bounce instead of how an MRI works. That's a big disconnect. I also never understood why future doctors need to know organic synthesis. But that's what they get taught. And it perpetuates itself, the pigeonholing of academic interests because then you can have a whole field "all to yourself" so you can get tenure. Great.
I once heard Richard Ernst (nobel laureate) speak about how it is a crime when academia sees their only real educational duties as the nurturing of future academics to someday replace them. I couldn't agree more.
There's the defeated clan whose only hope lies in smuggling their princess to friendly territory. She is guarded by an old general dressed incognito (help me obi wan!). The comic relief is slapstick from a duo of clueless peasants. . who agree to help smuggle the strangers for money. ..one tall and skinny, the other short and fat . . .always bickering.
At one point, they get captured but the enemy leader recognizes the general and says he would have liked to face him on the battlefield. So instead he calls off his troops and they have a one on one spearfight. (the fight choreography is strikingly similar to a light-saber fight. FYI many of the actors were trained by real samurai, the very last of their kind).
Darth vader's helmet is very similar to a samurai helmet, especially when they wear the facemask and slatted armor.
I'm not saying Lucas didn't add alot of his own. But the influence is striking. Not quite as blatant as "magnificent seven" vs "the seven samurai" but still there.
I've digressed really far. I was trying to assert that given Lucas' fondness for Kurosawa, any perceived "cowboy movie" influence is more likely to be samurai movie influence. That's all. As for what is called "science" in science fiction, that completely depends on your goal (star wars is practically fantasy but we still like it fine). A. Clarke always maintained that suffiently advanced technology is indistinguishable from magic. You could take the smartest neanderthal ever and probably never be able to convince it that a VCR and TV are anything other than magic. So while there is value in making the science believable for stories set in the near-future, it's pretty ridiculous to even pretend we can imagine what the distant future will be like. So the choice is to choose "hard" science fiction so that we can consider society today and how slight changes might affect it, or we can erase all the rules and just make it plain old fiction and try to stand on the quality of the writing without worrying about "scientific" merit (Dune qualifies as a good example here - the author was genuinely trying to do more than just entertain the reader but without "hard" explanations).
"social life. . heh. . .women. ..heh. . .a physicist craves not these things.. ."
I think he went on to say econ majors were the dark side of the force but he was usually drowned out by hysteric laughter at this point.
I have to tell you, though, it doesn't help that most of the women in physics I knew were completely asexual. Except for the one diamond in the rough that was an avid video gamer, RPG freak, complete sci-fi dork, and I have no doubt she making some lucky geek very happy right now :-)
Centrifugal "force" and the Coriolis "force" are not real forces because in both cases you would see it's just inertia and nothing more if you were observing from the proper inertial frame . It just looks like an extra force when we make the assumption that our rotating Earth is an inertial frame or that the merry-go-round we are on is an inertial frame (both cases are rotating which means they are being accellerated and are therefore improper frames). So the short answer is, we understand those "forces" as much as we understand inertia. . the "pseudo" part is just an artifact of its a real pain in the ass to have to deal with the earth's rotation in every little computation we do - so like the lazy bums we are, we just ignore it if we can get away with it.
I was floored on my first day of intro physics when the professor told us it's just an assumption that intertial mass was the same as gravitational mass. It's been shown to 10 decimal places or so but not formally proven (I think it stems from the same inability to describe gravity that the original post was about). My college profs were always very good about pointing out what we do and don't know. my only beef there is that only the physics majors get the good profs who tell us this stuff. . someone who was taking physics for engineers or physics for biologists probably ended up doing alot of arithmetic about bouncing balls without ever being told the parts that are actually interesting.
I do realize, however, that's is much harder to teach a subject to an auditorium full of people taking the subject for a requirement than a small group of nerds who already have proven commitment given that of their own free will they signed up for masochistic science course 666. I wish schools would try and find professors who are good at teaching non-majors instead of relegating that duty to whichever disgruntled professor is on the department's naughty list. There actually was a chemistry 666 in my school, it was advanced synthetic organic chemistry I think :-)
But what's eating all the theorists is that they have absolutely no idea why. The venerable laws of gravitation are empirical, in the sense that noone knows where it comes from other than the fact that it is associated with mass. All the other forces of nature have a quantum explanation, and have a particle that transmits them (most notably electromagnetism and photons). Noone has been able to satisfactorily reconcile gravity with any fundamental (quantum mechanical) nature of a particle.
It's almost scary that we know more about what binds subatomic particles together than what keeps the moon orbiting the earth. It's also ironic that most people's only introduction to physics is newtonian physics which is presented in textbooks as complete and understood. It's true we have the math to predict the effect of gravity to arbitrary precision, but I'm sure engineers can back me up that just because something has a robust empirical law doesn't mean anyone really understands how it works.
That being said it is strange that they are so vehemently against an NIH database which is primary geared towards biological compounds (i.e. proteins and nucleic acids and derivatives) which is pretty orthogonal to most of the chemical research world. But it would be a gross oversimplification to paint the ACS as an evil money grubbing organization.
Besides, chemists are rarely evil. Science fiction proves it's always an overweight doctor come-geneticist played by marlon brando that's evil.
There is plenty of written science-fiction with as much or as little science in it as you want - I think your generalization about modern sci-fi not having science only applies to TV series and big-budget movies. There is always something wrong with being arrogant, agreed, but there's nothing wrong with liking science fiction better than any other genre.
But I do think you are missing out on a crucial observation: real scientists love science fiction. Especially physicists, we crave that stuff, the good , the bad, and sometimes the ugly (I've also met more than a few chemists and engineers who feel the same way). Why? Because if we didn't believe that ass-boring and ass-hard equations could somehow have a big effect in the long run then we are just a bunch of masochistic idealists or incredibly underpaid technicians :-)
I think the focus on the "hardness" of sci-fi is completely misplaced. Granted, there is a divide between science fantasy (say, star wars) and there's science fiction (say 2001). But ultimately, other than being fun escapism (which I wholeheartedly endorse BTW) the function of science fiction has always been to explore the societal impacts of technology. It's not good science fiction just because the author is a PhD and spends an entire chapter lecturing me on a spaceship propulsion system that adds nothing to the plot (although that is ever so slightly better than an author who in real life couldn't build a model rocket from a kit who cribs something he read in scientific american). It's good science fiction if it makes you think hard about something you never thought about before. A good science fiction author immerses you in a world that you, by the end of the book, could actually believe exisited in some future history book. Technology is not a plot gimmick or a crutch but something that brings a new aspect of humanity under the magnifying glass, and none of us would be even reading this news site if we weren't fascinated by how society is shaped by technology, right? My 2 cents (and don't get me wrong, I do like hard-science fiction if its done well, but being a graduate student in science, I have to read enough hokey articles obfuscated by irrelevant technical details in modern journals as it is!)
That was one of the original motivations for the ISS. (to server as a refueling station/staging point). In fact, it's pretty much the only reason we signed up to build it. Buti t's quickly become a moot point because the ISS has been scaled back so many times due to budget cuts that the 2 crewmen there have to spend their entire efforts focused on not dying. I'm afraid the ISS is all show and no substance, they had to cut out any potential for innovative science when they were cutting it's budget. And yet it continues to drain valuable $$ that NASA could really use right now.
I worked at a NASA center briefly as a computer technician. The high level bureaucrats had plush carpeting, receptionists, potted palms trees, and $20,000 sparc workstations to check their email with. While the engineers in the basement actually designing a *&(* satellites would come and asked me what type of computer I could acquire for them with their discretionary fund of 50$. I scrounged around and they were the new, proud owners of a 386sx with a yellow screen (monochrome!) that noone else wanted (I'm dating myself here).
There was plenty of lower middle management that had engineering backgrounds, which you'd think would be a good thing (i.e. being able to communicate with subordinates and translate to superiors, etc). Thing was, ones I met were ex-engineers who were lucky enough to be civil servants. If their project failed miserably, but not so miserably to cause an inquiry, they got reassigned to lower management. Civil servants have tenure and are hard to fire, so obviously contractors are preferred. Which means all your talent leaves the moment a project ends and then the next project doing the exact same thing has to re-learn everything the last guys figured out.
It also doesn't help that between the space station and the space shuttle (neither of which can be lowered any more due to politics), and salaries for 10 levels of bureaucracy, thats already 90% of the entire budget. AFAIK the only thing keeping any science alive is the Jet Propulsion Lab. Caltech is still attracting young talent away from industry instead of vice versa.
Well, there's a big difference between "useful for limited sets of problems" and useful to biology in general. I'll point out you don't even NEED molecular dynamics if all you wanted was a static h-bonding pattern, steric clashes, and distance contraints. Rasmol and a structure file will do fine. The force fields are getting better but with the current machinery in place, we can never even hope in our wildest dreams that MD is going to tell us the mechanistic details of catalysis or protein-protein recognition if you just load a structure and hit "go". We still have to do a boatload of biochemistry, alot of mutational analysis (with hypothesis as deep as "this part looks important so lets kill it"). The "promise" of accurate simulations on the same timescale as the actual biochemical and (gasp) biological events the macromolecules perform is still a pipe dream.