I think you're right. Arguably it's a good idea, whether or not the organisms themselves are ever grown, because the DNA may have interesting genes in it that future biotechnologists might want to study and use, when we get to the point where we're able to not only "read" a genome easily but with full comprehension.
It's for this reason that the actual viability of the seeds isn't maybe that much of an issue. So long as the DNA remains intact and can be sequenced, it will be useful.
Although...I wonder if they might not be better off spending the money on sequencing the genomes now. That data can then be stored in many different places, and probably far more compactly and easily than the seeds. Furthermore, I think the mol bio field generally agrees that in the not too distant future it should be relatively straightforward to understand gene function from sequence, and that means only the sequence is really needed anyway. We won't need the actual DNA itself, because we can always reconstruct it, or the part of it we need.
Basically I'm saying maybe preserve all these plant species virtually, in cyberspace, instead of actually, in the frozen tundra. Cheaper. As well as more cyberpunk.
Wait...what are you disagreeing with? I didn't say 0.08 is fairly sloshed, I said 0.10 (which is 25% more alcohol in your blood) is. I've had a BAC of about 0.10, if you believe the usual estimates of how much you have to drink to get there. For me, they say I'd have to drink a whole sixpack of beer or an entire bottle of wine over a 2-hour dinner. Hop behind the wheel after that? Noooo thank you, that would be madness.
That said, the 0.08 limit for me -- four beers or 3/4 bottle of wine over dinner, they say -- is still a little dicy. I could maybe see driving on two beers, or 1/2 a bottle of wine, which probably puts me about 0.03.
But in practise, I'm even more conservative than that, and pretty much don't drive if I've had anything to drink at all in the last 3-4 hours. This isn't out of some kind of moral statement or anything, and it's not because I don't think I can cope with the road, bad weather, sleet or snow, whatever. It's not the natural hazards on the road that scare me, it's the human. That is, it's just that I know a lot of other fools out there are probably half-sloshed, or yakking on the cell phone, or having a fight with the GF/BF in the passenger seat, and I want all my wits about me to deal with them.
And, I do agree that the DUI laws are almost inhuman these days in the way the punishments are way out of proportion to the crime. I mean, if you actually do any harm while driving drunk -- e.g. kill someone -- then you get toasted in the usual criminal way, e.g. convicted of voluntary manslaughter and sent to the state pen for 5 years. So the DUI laws really only come into play when you have done no actual harm, you're just seen as capable of harm. Well, it's a little weird to be punishing people so harshly for doing things that might turn out to be dangerous, but which haven't, yet. It doesn't sit well with me. Seems kinda' Brave New World like, where Society decides what's best for you. I'm all for individual freedom, letting people do pretty much what their individual judgment says they can do -- with the catch that if your individual judgment screws up and you do someone harm, you catch all the hell we can devise, and no mercy, no whining your way out of it because society let you down by letting you screw up.
Give me a break. There are two possibilities here:
(a) by "public transporation" you mean something that picks up lots of people at point A at x o'clock and delivers them to point B at y o'clock. Problem is, of course, I neither live at point A, nor work at point B. Nor do I typically want to leave my house at x or arrive at y. The inevitable result is an absurd waste of my time in transportation that is a far more convoluted than it needs to be. And for what purpose? What is the benefit for which I should bear this invisible tax on my labor? Because there is some theoretical increase in energy efficiency when people are moved in large lots, instead of one at a time, on an as-needed, just-in-time basis. (And at that, this usually presumes the mass-mover is operated at its most efficient, e.g. the bus is always full and the train always leaves on time, whereas these assumptions are never justified in practise.)
That is to say, massive public-transporation gains fuel efficiency at the cost of time efficiency, the time it takes to buy tickets, hoof it down to the station, wait in line, wait for the train, re-organize your life around the bus routes, live close to a station instead of where the schools are good, and so forth. We save fuel, but waste time. Eh, count me out. Time is more precious than fuel, I think.
Why not start with a transporation system that is as efficient in its consumption of peoples' time as the automobile, and try to find a way to make it more fuel efficient? Automatic control on highways, for example. You get on and the central computer takes over, drives you inhumanly fast at insane following distances -- four lanes of cars spaced eight inches apart, each doing 90 MPH -- maximizes your fuel efficiency and maximizes the efficiency of the road.
Which brings us to...
(b) By "public transporation" you mean a system designed to be so flexible and reponsive to peoples' actual needs that it isn't much trouble at all to switch from using a car to using the cool snazzy public transportation what's-it, little bubble cars you can call with a clicker and which drop you off within a few hundred feet of the office, et cetera.
But then, you can easily get at this kind of system from the other direction, by evolving the private transporation system we've got. And I think it's better done that way, for the simple reason that people care about systems they own, whereas the Tragedy of the Commons makes sure they neglect systems they own in common. Even poor folks take care of their own cars, and even rich folks tend to abuse the public bus or train. Fact of human nature. So any system held in common ownership is almost guaranteed to be neglected, mismanaged, and -- yep, here it comes: inefficient.
If your blood alcohol concentration is above certain limits set by law, typically 0.08% these days, then you are defined by law to be drunk, whatever your state of reflexes, ability to concentrate, et cetera. Otherwise, whether you are drunk or not is a matter of judgment -- not yours, of course, but typically that of the policeman who stops you and the judge who hears the case.
However, the limit used to be 0.10%, and that is actually fairly sloshed. You would be pretty happy, typically, although people vary. The point is that it used to be the case that you could be definitely drunk, and know it, but still be under the limit at which you would be defined to be drunk by the law. Naive people would imagine, therefore, that you could be drunk but not legally drunk (because you were under the 0.10% limit). This was never the case, of course, since even under the 0.10% limit you could still be determined to be drunk by a policeman and a judge. But it was a popular fiction.
From this beginning I think nowadays "legally drunk" has morphed into a colloquial expression meaning mostly just "pretty definitely drunk" versus just feelin' good -- you know, at that point where friends argue happily with each other -- hey, I'm not drunk, man, just...relaxed...g'wan, ask me anything...look! I can balance a beer bottle on my nose (crash)!
But...why bother training drivers? Why not turn over these mechanical functions to a silicon brain that won't get bored with them?
For my money, driving is an adventure for, say, the first 10 years you do it, mildy amusing for the next 10 years, and then just boring except on rare occasion after that. Mostly it's a drudge. Most days, I'd much rather just slide into the seat, tell the car "To the office, Bud, and don't spare the horses" -- then sit back and read, catch some more Z's, or jack in to some mobile wireless Internet connection and read the headlines while sipping coffee. Weekends, maybe, I'd rent an MG and swoop down the coast. But M-F in eight lanes of 45 MPH traffic? Ugh. Bring on the robot cars!
It doesn't seem that tricky to me. If you have a bigger sensor, I'd think you just need to insert one new lens permanently inside the camera, which takes the image at the location of the old chip and refocusses it on the new one. All your old lenses do the job of bringing the image to that point, just as before. It's only that now there's not a device there but one more lens, that relays the image to the new device. Am I missing something?
I don't think the temperatures is anywhere near "normal" fusion temperatures. Here is a reference suggesting you need 10-100 million degrees. TFA says the the temperature in the bubbles is maybe 15,000 or so at best.
This is not exactly Pons and Fleischmann style cold fusion, because it isn't room-temperature cold. The temperature at the center of the imploding bubbles is at least 5-20 kK. In other words, at least the temperature of the surface of the Sun, probably much higher.
The problem, however, is that this is still nowhere near the temperature of the center of the Sun, which is 10-100 MK. I find it pretty unlikely that it could get that high, too. But who knows? Stranger things have happened. Just not in my lifetime.
Well, the point of their algorithm -- and I haven't read the PRL paper either -- is that it's off-lattice. It's an open question how much off-lattice systems, e.g. real fluids, differ from the lattice models you mention, when it comes to critical behaviour. I think the critical exponents are the same, but other stuff will vary.
So in that sense, if they've come up with a better simulation algorithm for critical fluids, it's a big deal. At least, it is for chemical engineers, although I grant the physicists may be less impressed. The chemical engineers are, as a rule, interested in detailed information about a particular, chemically-distinct critical fluid, like critical CO2. That makes them different from the physicists, who are more interested in universal models of criticality, such as are easily illustrated with an Ising model.
Chemical engineers have been interested in using supercritical CO2 for extraction processes (like decaffeination) for a long time, because the fluid appears to be a powerful solvent, meaning it can dissolve away your impurity of choice, but it has zero work-up and clean-up issues: you just drop the pressure a tad, and your fluid turns into plain gaseous CO2 and puffs away, no mess no fuss.
I completely fail to see any application to fuel cells, however. Nothing supercritical in there that I know about.
You could be right. I tried Windows around 95-98, and kind of gave up on it when Linux on the desktop seemed so clearly superior. Maybe Windows has come a long way since then.
I think the other question is: how long will the system work? My experience with Windows is that, even if everything works fine when you boot up, the system starts to drag and behave weirdly if you keep it running long enough, open and close applications, have multiple users log in and out, et cetera. I'm guessing the memory management isn't so hot.
On the other hand, I've routinely run my Linux machines for 3-6 months without reboot, 'cause the memory management is sound.
But...what did you expect? I'm sure MS actually does do fair tests in which Linux comes out on top, but of course they do them at midnight during the dark of the Moon in a secret underground lab, and they ritually slaughter the engineers afterward so they don't talk.
Seriously, any sensible corporation tests their competitors' products, and keeps the results strictly to themselves. Why give the competition any help? On the other hand, when some random test or other has results that look good for you, however accidental or meaningless that is, then of course the marketing department is sent a memo to slip this into some advertising or other.
Although...if you're finding that you can start (as opposed to kill) conversations at parties with geeky stories like that -- hey, I want to go to those parties. Where are they??
Well, that's the media's fault, isn't it? If people want to grab onto any passing excuse for a Messiah, and read Gospel in random paper napkin jottings, it isn't the fault of the scientists.
There's no hope of reforming the journalists. These are the guys and gals who, like Barbie, think 6th grade math is hard. They're quick-witted and good at turning a clever phrase, but their minds are a mile wide and inch deep, mostly. There are exceptions, of course.
Fortunately the readers and watchers of the media circus often exhibit a healthy skepticism on their own. I daresay plenty of people skim a breathless story in Time asserting millions now living will never die, thanks to this or that nanotech miracle which is Just Around The Corner(TM) -- and think: Yeah, right..
I don't think so. Let me tell you from 15 years of publishing in scientific journals, and reviewing the proposed publications of others, that there is no clear and sharp division between an "honest" mistake and a mistake into which you are led by bias, preconceived notions, or your personal feelings for another scientist whose work you are challenging or confirming. Scientists are human beings as much as the next person. Very few will deliberately and with malice aforethought falsify data. But plenty will talk themselves into believing that a certain dubious "correction" of the data makes sense.
It's a lot like high-school chemistry lab, in which (if you were decently smart), you knew what the results of the lab should be. Does that affect the way in which you write down the data? You bet. You do the experiment once, and you get a result you "know" is crazy. So you say: "That can't be right, something must have gone wrong..." and you do it again. If you get the result you expect, then you tend to just write it down uncritically.
Just expand that typical human behaviour to much more complex experiments, and you'll see what I mean. Grown-up scientists do an experiment, and they get a result that "can't be right," so they do it again until they get a result that "seems right," or they talk themselves into some kind of data analysis that "corrects" the raw data. Have a look here (warning: PDF link) for an interesting discussion of the case or Robert Millikan, who "framed a guilty man", in the phrase made immortal by the LAPD, by falsely presenting experiments that led to a correct scientific conclusion.
The long and short of it is that the question of the "honesty" of the author of a publication is very much a gray area, and anyone who seriously just assumes that all the data from an experiment have been presented, and all the data analysis has been done in completely neutral way, without any influence of preconceived notions, is a fool. You must assume that the personal predispositions of the scientist doing the work had some influence on the experimental data reported. This isn't meant to be pejorative -- I'm not saying you assume other scientists are routinely dishonest. You just assume they're human, and may have fooled themselves or have a bit of an agenda when they present their data, and you take that into account. Healthy skepticism is the order of the day. That's why we like to see even experiments that seem completely unexceptional and from scientists of unimpeachable reputations repeated several times by a broad range of other workers before we accept them.
I certainly agree deliberate fraud is way out of any "gray area" about the motivations of the scientist submitting articles for publication. (And that's why the punishment for doing so is far, far harsher than for simply making an "honest" mistake, or even a mistake into which you are led by bias or incompetence.) But there is no way one can, or should, draw a sharp line between completely unconscious error and semi-conscious half-deliberate fudge, and it would be a great error for anyone to blindly assume that the data in any scientific publication is beyond question.
Well, one can feel for the poor woman. Maybe she feels there's no way she can ever win a rational argument with her husband, since he's got all the goods (the training, the experience, perhaps the bulk of the intelligence). Since people hate to feel completely outclassed, when they're in a situation like this, often enough they'll invoke some kind of mystical Dr. McCoy "there are some things logic can't explain" type of argument, something which by definition isn't subject to rational examination. Game over, I win, ha ha.
Best to recognize this when you see it as fast as possible, and just disengage. Can't do that when you're married to the person, however. Oh well. There are more reasons than mere dorkiness for the proverbial loneliness of the hyperrational male engineer.
First of all, peer review functions in many places besides vetting articles for publication. Indeed, it's much more important in reviewing grant applications and in how and whether your colleagues direct good students and post-docs your way, since while publications are nice, it's successfully attracting research money and recruiting good employees that really counts. This guy is getting that kind of peer review now -- and greatly to his harm. So indeed the system is functioning as designed.
More importantly, if you're saying the system is busted because it must sometimes punish fraud after it's published, instead of preventing its publication entirely -- well, then perhaps something needs to be clarified about the nature of scientific publication. A scientific journal is not a textbook. Stuff published there is current research, not accepted wisdom. It's not meant to be archival quality, things that folks will stake a reputation on. It's meant to be the "bleeding edge" of knowledge, so to speak, the latest and (necessarily) shakiest bit of possible insight. Reasonable people expect much that is published in a journal to turn out to be wrong, or incomplete. They don't ordinarily expect it to be a fraud, but it does happen on occasion, and reasonable people keep that in the back of their minds, too.
In fact, one of the main reasons for scientific publication is to present new ideas and data to the widest possible audience, so that people who don't know, fund, or work for the original researcher have a chance to consider the merits and drawbacks of the idea, test it, challenge it, and prove or disprove it. You might reasonably think of scientific publication as more or less a "debugging" step of a new scientific idea, the process by which you submit some newfangled notion to the rigours of a bunch of "beta testers" (other scientists) who will bang on the idea, make sure it's sound.
You would not, I hope, conclude that because spectacular bugs are sometimes found in software at the "beta" stage this means that the authors were wrong to release it at all. Having a large community of interested expert users cooperate in beta testing your software -- think open-source software -- can speed up the process of producing quality products greatly. That's exactly how scientific publication works.
Hmmm, this sounds a bit like urban legend to me. First off, while indeed the melting point of platinum is much higher than that of silver (2040 K versus 1235), platinum is twice as dense as silver (21.1 g/cm^3 versus 10.5 g/cm^3). Secondly, while platinum does occur mixed with other metals, those are typically palladium, rhodium, iridium, osmium, and ruthenium -- not silver. Silver also occurs with other metals, but usually lead, zinc, and copper, occasionally gold -- but not platinum. This isn't especially surprising, as their chemistry is quite different, Ag preferring a +1 oxidation state but Pt preferring +2 and +4. The big density difference also suggests they would not occur together as native metals (because in the molten state they would separate, like oil and water).
Finally, it seems odd that if medieval silversmiths were familiar with the metal they would not have named it. After all, medieval (or rather Renaissance) miners and smiths did name other "annoying" substances that interfered with their activities: the name of the element nickel comes from kupfernickel ("Old Nick's copper" or "the Devil's copper"), a German miner's term for the worthless mineral niccolite (NiAs), which looks like a valuable copper ore, copper (I) oxide. The element cobalt was named after the term kobold ("evil spirit"), given by German miners to the useless and somewhat poisonous rocks of cobaltite (CoAsS) that occured in their silver mines. Zinc supposedly has its name in the German word for "sharp point" because it formed spiky crystals in certain refineries. And so on.
Anyway, it seems hard to believe if Pt was a common annoyance in early European smithing it wouldn't have been named, and that name reflected in the name of the element. But platinum was known to miners -- it was known as platina del Pinto ("little silver of Pinto" in Spanish) to 18th century Spanish gold miners in the Pinto River basin of Columbia, and it was on that basis that the element was named "platinum" by Ulloa and Sir William Watson in 1748.
Not saying what you've said is definitely wrong, but it seems a little odd.
I've seen an interesting illustration of your excellent point in certain types of theoretical many-body physics. Basically, because it's straightforward to do computer simulations of complex many-body systems, people tend to turn to that tool first. If it gets them the answers they want, that tends to short-circuit the effort that would otherwise go into developing new theoretical insight that would solve the problem without the brute-force number crunching.
Now, all scientists certainly prefer to get their answers from brilliant, simple theoretical insight rather than brute-force numerical computation, but...if the latter is much faster and cheaper, well, you do need to publish or perish...
Fundamentally, I can't argue with you. I don't know enough about what it was like to be a physicist in other ages to know whether what we have now is much more complex than what they had then, or whether it's just different. My instinct is that it's just different. The reason is that on the occasions I have had to study carefully extinct theories, I have been impressed with how complex, ingenious and compelling they are, seen within the framework of the facts known at the time. I don't feel like our ancestors were any less imaginative than we, nor that their world-view was any less complex and clever.
I understand the "standing on the shoulders of giants" argument, but I think that speaks more to the direction of our thinking, not the complexity. It's not, I think, that general relativity is harder in principle to grasp than the cosmological theories of Plato. It's just that GR is correct, and Plato's ideas were not. Furthermore, we usually know Plato's theories, if we know them at all, only as a cartoon, a sketch -- because we don't want to waste time on knowing thoroughly something that is wrong. So we probably routinely underestimate how clever and compelling they were at the time they seemed correct.
But, as I said, I only have a feeling about this. I can't prove you're wrong and I agree you may be right.
To answer your minor questions: my opinion on M-theory would not be worth the paper and ink it took to print it -- it's not my field. As for quantum mechanics -- believe it or not, I find it conceptually simpler than classical mechanics. Classical mechanics requires me to wrap my head around all kinds of funny infinities and infinitismals that lead to such nonsense as Zeno's Paradox. I find the sums and finite differences of QM more "sensible" than the integrals and infinitismals of CM. People tend to think QM is philosophically more complex than CM, with but that's a cat in a box in a superposition of different colors.
I dunno. One of the things that made Einstein great, from the scientific point of view, is that he cleared away a great deal of 19th-century complexity in optics and mechanics and replaced it with a theoretical framework of such dazzling simplicity that anyone could grasp its basic principles, but of such power that understanding its full implications demands high intelligence and decades of sustained study.
It's like the way Copernicus swept away the huge complexity of the Ptolemaic astronomer's theory of planetary orbits, all those cycles and epicycles, with the simple and powerful idea of the elliptical orbit. Or how Mendeleev replaced the 18th century's bewildering lists of correlations between chemical properties of substances with the simple and powerful organizational principles of the Periodic Table.
Even in my own experience as a theoretician I find the truly brilliant ideas are not complex. They're insights that drastically simplify and clarify. They're the kind of things that, when you understand them, make you slap your head in awe and envy.
So, from this point of view, the hideous complexity of modern high-energy physics theories could well be a sign that they lack brilliance, that another Einstein is needed to clear away all the baroque epicycles, so to speak, and replace it all with something beautifully simple and far more powerful.
Of course, this might not be true -- it might instead be the case that the basic structure of the universe is simply too complex for ordinary humans to understand even its principles. But I find this hard to believe (for no logical reason, I admit).
So I personally disagree with Mr. Horgan. I think he's just channeling Albert Michelson in 1896 ("The more important fundamental laws and facts of physical science have all been discovered....Our future discoveries must be looked for in the sixth place of decimals.") Like Michelson, Horgan thinks that because no revolution has happened in 50 years one will never happen. But it was almost 300 years between Newton and Einstein. So I'd give it another century or two before giving up.
Well, I dunno. Have you really considered how much data you are talking about? Roughly taking a continuous movie of every person (or at most small groups of people, so you can resolve individual faces), on every street, all over a city, possibly all over many cities?
576x480 pixels times 16 frames/second is 17.7 megabytes/second for each movie stream. Multiply by at least several million streams if you want a general surveillance of everybody, and we are talking about capturing, transporting and storing tens to hundreds of megabytes per microsecond.
Even if that staggeringly huge data stream could be captured and stored, the bigger question is: how can it be quickly and cheaply searched? Is it really going to be practical for J. Random Asshat to do a "Mission: Impossible" search on exabytes upon exabytes of data to find the 60 or 70 frames in which his poor target is talking to a hooker? I mean, without spending $millions or breaking into the NSA to use their supercomputers (which is bound to attract attention)? It seems kind of doubtful.
I suspect what is much more likely to happen is that as face and gait recognition gets better, it is going to happen that if you have already come to the attention of the police for some reason, and they know what you look like, or how you walk, then they are going to be able to program cameras in certain key spots (e.g. airports or at ATMs) to report when they spot you. This is conceptually similar to putting a large number of tireless, unbribeable, non-coffee-break-taking silicon cops on the street with mugshots of everyone the police want to find in their shiny steely hands.
I'm not sure I agree with you here. In fact, books are exceedingly efficient ways of distributing valuable information that everyone wants. Thing is, the marginal cost for printing and distributing a book is very low, not much more than a few dollars. Plus they're sturdier and more portable and accessible than a laptop can ever be. All you need to boot them up is some decent light source. The only skill you need is to be able to read.
The Internet by contrast is superior at distributing actual applications, like software, or fast-changing information (like news, or the latest tech innovation, or Fedora Core X, or odd bits of information that are only of rare use. (For example, in your example, the reason you go to Google to find out how to purify water is that in your Western life you are very unlikely to need to do it. But you don't go to Google to learn the rules for driving a car, or how to add, because that's a ubiquitous skill that you learned long before you learned about Google. In a country where purifying water is a top and general priority, people are going to learn about it from their parents and neighbors, not Google.)
None of these seem especially relevant to folks in poor countries. What they really need is access to basic information that is already well understood (how to dig a safe well, how to prevent AIDS transmission, reliable accounting and credit practises, basic nutrition). These things are actually very well conveyed by book.
The one exception I can see is that the Internet is also good for two-way communication between people well-seperated, and places like Africa are often short of experts of one stripe or another. I can see how it might do some good if you could have interactive, or semi-interactive software, that might let a back-and-forth go on to teach people stuff better. Say, an adaptive teaching program that could teach a range of students, from the barely getting started to the most sophisticated. A book tends to be a one-size-fits-all solution, which does not serve the spectrum of students in the real world. That's why you need a teacher, too, to customize the learning. But software could, in part, replace the teacher at lower cost. Or serve as some kind of faux triage nurse that could ask some questions about your symptoms and find out whether you should just rest up and drink more fluids -- or whether you really should make that 2 day trip to the clinic. These things would be good.
Slashdot Mirror
I think you're right. Arguably it's a good idea, whether or not the organisms themselves are ever grown, because the DNA may have interesting genes in it that future biotechnologists might want to study and use, when we get to the point where we're able to not only "read" a genome easily but with full comprehension.
It's for this reason that the actual viability of the seeds isn't maybe that much of an issue. So long as the DNA remains intact and can be sequenced, it will be useful.
Although...I wonder if they might not be better off spending the money on sequencing the genomes now. That data can then be stored in many different places, and probably far more compactly and easily than the seeds. Furthermore, I think the mol bio field generally agrees that in the not too distant future it should be relatively straightforward to understand gene function from sequence, and that means only the sequence is really needed anyway. We won't need the actual DNA itself, because we can always reconstruct it, or the part of it we need.
Basically I'm saying maybe preserve all these plant species virtually, in cyberspace, instead of actually, in the frozen tundra. Cheaper. As well as more cyberpunk.
Wait...what are you disagreeing with? I didn't say 0.08 is fairly sloshed, I said 0.10 (which is 25% more alcohol in your blood) is. I've had a BAC of about 0.10, if you believe the usual estimates of how much you have to drink to get there. For me, they say I'd have to drink a whole sixpack of beer or an entire bottle of wine over a 2-hour dinner. Hop behind the wheel after that? Noooo thank you, that would be madness.
That said, the 0.08 limit for me -- four beers or 3/4 bottle of wine over dinner, they say -- is still a little dicy. I could maybe see driving on two beers, or 1/2 a bottle of wine, which probably puts me about 0.03.
But in practise, I'm even more conservative than that, and pretty much don't drive if I've had anything to drink at all in the last 3-4 hours. This isn't out of some kind of moral statement or anything, and it's not because I don't think I can cope with the road, bad weather, sleet or snow, whatever. It's not the natural hazards on the road that scare me, it's the human. That is, it's just that I know a lot of other fools out there are probably half-sloshed, or yakking on the cell phone, or having a fight with the GF/BF in the passenger seat, and I want all my wits about me to deal with them.
And, I do agree that the DUI laws are almost inhuman these days in the way the punishments are way out of proportion to the crime. I mean, if you actually do any harm while driving drunk -- e.g. kill someone -- then you get toasted in the usual criminal way, e.g. convicted of voluntary manslaughter and sent to the state pen for 5 years. So the DUI laws really only come into play when you have done no actual harm, you're just seen as capable of harm. Well, it's a little weird to be punishing people so harshly for doing things that might turn out to be dangerous, but which haven't, yet. It doesn't sit well with me. Seems kinda' Brave New World like, where Society decides what's best for you. I'm all for individual freedom, letting people do pretty much what their individual judgment says they can do -- with the catch that if your individual judgment screws up and you do someone harm, you catch all the hell we can devise, and no mercy, no whining your way out of it because society let you down by letting you screw up.
Give me a break. There are two possibilities here:
(a) by "public transporation" you mean something that picks up lots of people at point A at x o'clock and delivers them to point B at y o'clock. Problem is, of course, I neither live at point A, nor work at point B. Nor do I typically want to leave my house at x or arrive at y. The inevitable result is an absurd waste of my time in transportation that is a far more convoluted than it needs to be. And for what purpose? What is the benefit for which I should bear this invisible tax on my labor? Because there is some theoretical increase in energy efficiency when people are moved in large lots, instead of one at a time, on an as-needed, just-in-time basis. (And at that, this usually presumes the mass-mover is operated at its most efficient, e.g. the bus is always full and the train always leaves on time, whereas these assumptions are never justified in practise.)
That is to say, massive public-transporation gains fuel efficiency at the cost of time efficiency, the time it takes to buy tickets, hoof it down to the station, wait in line, wait for the train, re-organize your life around the bus routes, live close to a station instead of where the schools are good, and so forth. We save fuel, but waste time. Eh, count me out. Time is more precious than fuel, I think.
Why not start with a transporation system that is as efficient in its consumption of peoples' time as the automobile, and try to find a way to make it more fuel efficient? Automatic control on highways, for example. You get on and the central computer takes over, drives you inhumanly fast at insane following distances -- four lanes of cars spaced eight inches apart, each doing 90 MPH -- maximizes your fuel efficiency and maximizes the efficiency of the road.
Which brings us to...
(b) By "public transporation" you mean a system designed to be so flexible and reponsive to peoples' actual needs that it isn't much trouble at all to switch from using a car to using the cool snazzy public transportation what's-it, little bubble cars you can call with a clicker and which drop you off within a few hundred feet of the office, et cetera.
But then, you can easily get at this kind of system from the other direction, by evolving the private transporation system we've got. And I think it's better done that way, for the simple reason that people care about systems they own, whereas the Tragedy of the Commons makes sure they neglect systems they own in common. Even poor folks take care of their own cars, and even rich folks tend to abuse the public bus or train. Fact of human nature. So any system held in common ownership is almost guaranteed to be neglected, mismanaged, and -- yep, here it comes: inefficient.
If your blood alcohol concentration is above certain limits set by law, typically 0.08% these days, then you are defined by law to be drunk, whatever your state of reflexes, ability to concentrate, et cetera. Otherwise, whether you are drunk or not is a matter of judgment -- not yours, of course, but typically that of the policeman who stops you and the judge who hears the case.
However, the limit used to be 0.10%, and that is actually fairly sloshed. You would be pretty happy, typically, although people vary. The point is that it used to be the case that you could be definitely drunk, and know it, but still be under the limit at which you would be defined to be drunk by the law. Naive people would imagine, therefore, that you could be drunk but not legally drunk (because you were under the 0.10% limit). This was never the case, of course, since even under the 0.10% limit you could still be determined to be drunk by a policeman and a judge. But it was a popular fiction.
From this beginning I think nowadays "legally drunk" has morphed into a colloquial expression meaning mostly just "pretty definitely drunk" versus just feelin' good -- you know, at that point where friends argue happily with each other -- hey, I'm not drunk, man, just...relaxed...g'wan, ask me anything...look! I can balance a beer bottle on my nose (crash)!
It has nothing to do with the legal drinking age.
But...why bother training drivers? Why not turn over these mechanical functions to a silicon brain that won't get bored with them?
For my money, driving is an adventure for, say, the first 10 years you do it, mildy amusing for the next 10 years, and then just boring except on rare occasion after that. Mostly it's a drudge. Most days, I'd much rather just slide into the seat, tell the car "To the office, Bud, and don't spare the horses" -- then sit back and read, catch some more Z's, or jack in to some mobile wireless Internet connection and read the headlines while sipping coffee. Weekends, maybe, I'd rent an MG and swoop down the coast. But M-F in eight lanes of 45 MPH traffic? Ugh. Bring on the robot cars!
It doesn't seem that tricky to me. If you have a bigger sensor, I'd think you just need to insert one new lens permanently inside the camera, which takes the image at the location of the old chip and refocusses it on the new one. All your old lenses do the job of bringing the image to that point, just as before. It's only that now there's not a device there but one more lens, that relays the image to the new device. Am I missing something?
I don't think the temperatures is anywhere near "normal" fusion temperatures. Here is a reference suggesting you need 10-100 million degrees. TFA says the the temperature in the bubbles is maybe 15,000 or so at best.
This is not exactly Pons and Fleischmann style cold fusion, because it isn't room-temperature cold. The temperature at the center of the imploding bubbles is at least 5-20 kK. In other words, at least the temperature of the surface of the Sun, probably much higher.
The problem, however, is that this is still nowhere near the temperature of the center of the Sun, which is 10-100 MK. I find it pretty unlikely that it could get that high, too. But who knows? Stranger things have happened. Just not in my lifetime.
Well, the point of their algorithm -- and I haven't read the PRL paper either -- is that it's off-lattice. It's an open question how much off-lattice systems, e.g. real fluids, differ from the lattice models you mention, when it comes to critical behaviour. I think the critical exponents are the same, but other stuff will vary.
So in that sense, if they've come up with a better simulation algorithm for critical fluids, it's a big deal. At least, it is for chemical engineers, although I grant the physicists may be less impressed. The chemical engineers are, as a rule, interested in detailed information about a particular, chemically-distinct critical fluid, like critical CO2. That makes them different from the physicists, who are more interested in universal models of criticality, such as are easily illustrated with an Ising model.
Chemical engineers have been interested in using supercritical CO2 for extraction processes (like decaffeination) for a long time, because the fluid appears to be a powerful solvent, meaning it can dissolve away your impurity of choice, but it has zero work-up and clean-up issues: you just drop the pressure a tad, and your fluid turns into plain gaseous CO2 and puffs away, no mess no fuss.
I completely fail to see any application to fuel cells, however. Nothing supercritical in there that I know about.
You could be right. I tried Windows around 95-98, and kind of gave up on it when Linux on the desktop seemed so clearly superior. Maybe Windows has come a long way since then.
I think the other question is: how long will the system work? My experience with Windows is that, even if everything works fine when you boot up, the system starts to drag and behave weirdly if you keep it running long enough, open and close applications, have multiple users log in and out, et cetera. I'm guessing the memory management isn't so hot.
On the other hand, I've routinely run my Linux machines for 3-6 months without reboot, 'cause the memory management is sound.
But...what did you expect? I'm sure MS actually does do fair tests in which Linux comes out on top, but of course they do them at midnight during the dark of the Moon in a secret underground lab, and they ritually slaughter the engineers afterward so they don't talk.
Seriously, any sensible corporation tests their competitors' products, and keeps the results strictly to themselves. Why give the competition any help? On the other hand, when some random test or other has results that look good for you, however accidental or meaningless that is, then of course the marketing department is sent a memo to slip this into some advertising or other.
Go here. Absolutely amazing stuff.
Although...if you're finding that you can start (as opposed to kill) conversations at parties with geeky stories like that -- hey, I want to go to those parties. Where are they??
Well, that's the media's fault, isn't it? If people want to grab onto any passing excuse for a Messiah, and read Gospel in random paper napkin jottings, it isn't the fault of the scientists.
There's no hope of reforming the journalists. These are the guys and gals who, like Barbie, think 6th grade math is hard. They're quick-witted and good at turning a clever phrase, but their minds are a mile wide and inch deep, mostly. There are exceptions, of course.
Fortunately the readers and watchers of the media circus often exhibit a healthy skepticism on their own. I daresay plenty of people skim a breathless story in Time asserting millions now living will never die, thanks to this or that nanotech miracle which is Just Around The Corner(TM) -- and think: Yeah, right..
I don't think so. Let me tell you from 15 years of publishing in scientific journals, and reviewing the proposed publications of others, that there is no clear and sharp division between an "honest" mistake and a mistake into which you are led by bias, preconceived notions, or your personal feelings for another scientist whose work you are challenging or confirming. Scientists are human beings as much as the next person. Very few will deliberately and with malice aforethought falsify data. But plenty will talk themselves into believing that a certain dubious "correction" of the data makes sense.
It's a lot like high-school chemistry lab, in which (if you were decently smart), you knew what the results of the lab should be. Does that affect the way in which you write down the data? You bet. You do the experiment once, and you get a result you "know" is crazy. So you say: "That can't be right, something must have gone wrong..." and you do it again. If you get the result you expect, then you tend to just write it down uncritically.
Just expand that typical human behaviour to much more complex experiments, and you'll see what I mean. Grown-up scientists do an experiment, and they get a result that "can't be right," so they do it again until they get a result that "seems right," or they talk themselves into some kind of data analysis that "corrects" the raw data. Have a look here (warning: PDF link) for an interesting discussion of the case or Robert Millikan, who "framed a guilty man", in the phrase made immortal by the LAPD, by falsely presenting experiments that led to a correct scientific conclusion.
The long and short of it is that the question of the "honesty" of the author of a publication is very much a gray area, and anyone who seriously just assumes that all the data from an experiment have been presented, and all the data analysis has been done in completely neutral way, without any influence of preconceived notions, is a fool. You must assume that the personal predispositions of the scientist doing the work had some influence on the experimental data reported. This isn't meant to be pejorative -- I'm not saying you assume other scientists are routinely dishonest. You just assume they're human, and may have fooled themselves or have a bit of an agenda when they present their data, and you take that into account. Healthy skepticism is the order of the day. That's why we like to see even experiments that seem completely unexceptional and from scientists of unimpeachable reputations repeated several times by a broad range of other workers before we accept them.
I certainly agree deliberate fraud is way out of any "gray area" about the motivations of the scientist submitting articles for publication. (And that's why the punishment for doing so is far, far harsher than for simply making an "honest" mistake, or even a mistake into which you are led by bias or incompetence.) But there is no way one can, or should, draw a sharp line between completely unconscious error and semi-conscious half-deliberate fudge, and it would be a great error for anyone to blindly assume that the data in any scientific publication is beyond question.
Well, one can feel for the poor woman. Maybe she feels there's no way she can ever win a rational argument with her husband, since he's got all the goods (the training, the experience, perhaps the bulk of the intelligence). Since people hate to feel completely outclassed, when they're in a situation like this, often enough they'll invoke some kind of mystical Dr. McCoy "there are some things logic can't explain" type of argument, something which by definition isn't subject to rational examination. Game over, I win, ha ha.
Best to recognize this when you see it as fast as possible, and just disengage. Can't do that when you're married to the person, however. Oh well. There are more reasons than mere dorkiness for the proverbial loneliness of the hyperrational male engineer.
First of all, peer review functions in many places besides vetting articles for publication. Indeed, it's much more important in reviewing grant applications and in how and whether your colleagues direct good students and post-docs your way, since while publications are nice, it's successfully attracting research money and recruiting good employees that really counts. This guy is getting that kind of peer review now -- and greatly to his harm. So indeed the system is functioning as designed.
More importantly, if you're saying the system is busted because it must sometimes punish fraud after it's published, instead of preventing its publication entirely -- well, then perhaps something needs to be clarified about the nature of scientific publication. A scientific journal is not a textbook. Stuff published there is current research, not accepted wisdom. It's not meant to be archival quality, things that folks will stake a reputation on. It's meant to be the "bleeding edge" of knowledge, so to speak, the latest and (necessarily) shakiest bit of possible insight. Reasonable people expect much that is published in a journal to turn out to be wrong, or incomplete. They don't ordinarily expect it to be a fraud, but it does happen on occasion, and reasonable people keep that in the back of their minds, too.
In fact, one of the main reasons for scientific publication is to present new ideas and data to the widest possible audience, so that people who don't know, fund, or work for the original researcher have a chance to consider the merits and drawbacks of the idea, test it, challenge it, and prove or disprove it. You might reasonably think of scientific publication as more or less a "debugging" step of a new scientific idea, the process by which you submit some newfangled notion to the rigours of a bunch of "beta testers" (other scientists) who will bang on the idea, make sure it's sound.
You would not, I hope, conclude that because spectacular bugs are sometimes found in software at the "beta" stage this means that the authors were wrong to release it at all. Having a large community of interested expert users cooperate in beta testing your software -- think open-source software -- can speed up the process of producing quality products greatly. That's exactly how scientific publication works.
Hmmm, this sounds a bit like urban legend to me. First off, while indeed the melting point of platinum is much higher than that of silver (2040 K versus 1235), platinum is twice as dense as silver (21.1 g/cm^3 versus 10.5 g/cm^3). Secondly, while platinum does occur mixed with other metals, those are typically palladium, rhodium, iridium, osmium, and ruthenium -- not silver. Silver also occurs with other metals, but usually lead, zinc, and copper, occasionally gold -- but not platinum. This isn't especially surprising, as their chemistry is quite different, Ag preferring a +1 oxidation state but Pt preferring +2 and +4. The big density difference also suggests they would not occur together as native metals (because in the molten state they would separate, like oil and water).
Finally, it seems odd that if medieval silversmiths were familiar with the metal they would not have named it. After all, medieval (or rather Renaissance) miners and smiths did name other "annoying" substances that interfered with their activities: the name of the element nickel comes from kupfernickel ("Old Nick's copper" or "the Devil's copper"), a German miner's term for the worthless mineral niccolite (NiAs), which looks like a valuable copper ore, copper (I) oxide. The element cobalt was named after the term kobold ("evil spirit"), given by German miners to the useless and somewhat poisonous rocks of cobaltite (CoAsS) that occured in their silver mines. Zinc supposedly has its name in the German word for "sharp point" because it formed spiky crystals in certain refineries. And so on.
Anyway, it seems hard to believe if Pt was a common annoyance in early European smithing it wouldn't have been named, and that name reflected in the name of the element. But platinum was known to miners -- it was known as platina del Pinto ("little silver of Pinto" in Spanish) to 18th century Spanish gold miners in the Pinto River basin of Columbia, and it was on that basis that the element was named "platinum" by Ulloa and Sir William Watson in 1748.
Not saying what you've said is definitely wrong, but it seems a little odd.
I've seen an interesting illustration of your excellent point in certain types of theoretical many-body physics. Basically, because it's straightforward to do computer simulations of complex many-body systems, people tend to turn to that tool first. If it gets them the answers they want, that tends to short-circuit the effort that would otherwise go into developing new theoretical insight that would solve the problem without the brute-force number crunching.
Now, all scientists certainly prefer to get their answers from brilliant, simple theoretical insight rather than brute-force numerical computation, but...if the latter is much faster and cheaper, well, you do need to publish or perish...
Fundamentally, I can't argue with you. I don't know enough about what it was like to be a physicist in other ages to know whether what we have now is much more complex than what they had then, or whether it's just different. My instinct is that it's just different. The reason is that on the occasions I have had to study carefully extinct theories, I have been impressed with how complex, ingenious and compelling they are, seen within the framework of the facts known at the time. I don't feel like our ancestors were any less imaginative than we, nor that their world-view was any less complex and clever.
I understand the "standing on the shoulders of giants" argument, but I think that speaks more to the direction of our thinking, not the complexity. It's not, I think, that general relativity is harder in principle to grasp than the cosmological theories of Plato. It's just that GR is correct, and Plato's ideas were not. Furthermore, we usually know Plato's theories, if we know them at all, only as a cartoon, a sketch -- because we don't want to waste time on knowing thoroughly something that is wrong. So we probably routinely underestimate how clever and compelling they were at the time they seemed correct.
But, as I said, I only have a feeling about this. I can't prove you're wrong and I agree you may be right.
To answer your minor questions: my opinion on M-theory would not be worth the paper and ink it took to print it -- it's not my field. As for quantum mechanics -- believe it or not, I find it conceptually simpler than classical mechanics. Classical mechanics requires me to wrap my head around all kinds of funny infinities and infinitismals that lead to such nonsense as Zeno's Paradox. I find the sums and finite differences of QM more "sensible" than the integrals and infinitismals of CM. People tend to think QM is philosophically more complex than CM, with but that's a cat in a box in a superposition of different colors.
You're quite right, I was thinking of Kepler, not Copernicus! Thank you for the correction.
I dunno. One of the things that made Einstein great, from the scientific point of view, is that he cleared away a great deal of 19th-century complexity in optics and mechanics and replaced it with a theoretical framework of such dazzling simplicity that anyone could grasp its basic principles, but of such power that understanding its full implications demands high intelligence and decades of sustained study.
It's like the way Copernicus swept away the huge complexity of the Ptolemaic astronomer's theory of planetary orbits, all those cycles and epicycles, with the simple and powerful idea of the elliptical orbit. Or how Mendeleev replaced the 18th century's bewildering lists of correlations between chemical properties of substances with the simple and powerful organizational principles of the Periodic Table.
Even in my own experience as a theoretician I find the truly brilliant ideas are not complex. They're insights that drastically simplify and clarify. They're the kind of things that, when you understand them, make you slap your head in awe and envy.
So, from this point of view, the hideous complexity of modern high-energy physics theories could well be a sign that they lack brilliance, that another Einstein is needed to clear away all the baroque epicycles, so to speak, and replace it all with something beautifully simple and far more powerful.
Of course, this might not be true -- it might instead be the case that the basic structure of the universe is simply too complex for ordinary humans to understand even its principles. But I find this hard to believe (for no logical reason, I admit).
So I personally disagree with Mr. Horgan. I think he's just channeling Albert Michelson in 1896 ("The more important fundamental laws and facts of physical science have all been discovered....Our future discoveries must be looked for in the sixth place of decimals.") Like Michelson, Horgan thinks that because no revolution has happened in 50 years one will never happen. But it was almost 300 years between Newton and Einstein. So I'd give it another century or two before giving up.
Of course the needle is small. It's the haystack that's bogglingly large.
Well, I dunno. Have you really considered how much data you are talking about? Roughly taking a continuous movie of every person (or at most small groups of people, so you can resolve individual faces), on every street, all over a city, possibly all over many cities?
576x480 pixels times 16 frames/second is 17.7 megabytes/second for each movie stream. Multiply by at least several million streams if you want a general surveillance of everybody, and we are talking about capturing, transporting and storing tens to hundreds of megabytes per microsecond.
Even if that staggeringly huge data stream could be captured and stored, the bigger question is: how can it be quickly and cheaply searched? Is it really going to be practical for J. Random Asshat to do a "Mission: Impossible" search on exabytes upon exabytes of data to find the 60 or 70 frames in which his poor target is talking to a hooker? I mean, without spending $millions or breaking into the NSA to use their supercomputers (which is bound to attract attention)? It seems kind of doubtful.
I suspect what is much more likely to happen is that as face and gait recognition gets better, it is going to happen that if you have already come to the attention of the police for some reason, and they know what you look like, or how you walk, then they are going to be able to program cameras in certain key spots (e.g. airports or at ATMs) to report when they spot you. This is conceptually similar to putting a large number of tireless, unbribeable, non-coffee-break-taking silicon cops on the street with mugshots of everyone the police want to find in their shiny steely hands.
I'm not sure I agree with you here. In fact, books are exceedingly efficient ways of distributing valuable information that everyone wants. Thing is, the marginal cost for printing and distributing a book is very low, not much more than a few dollars. Plus they're sturdier and more portable and accessible than a laptop can ever be. All you need to boot them up is some decent light source. The only skill you need is to be able to read.
The Internet by contrast is superior at distributing actual applications, like software, or fast-changing information (like news, or the latest tech innovation, or Fedora Core X, or odd bits of information that are only of rare use. (For example, in your example, the reason you go to Google to find out how to purify water is that in your Western life you are very unlikely to need to do it. But you don't go to Google to learn the rules for driving a car, or how to add, because that's a ubiquitous skill that you learned long before you learned about Google. In a country where purifying water is a top and general priority, people are going to learn about it from their parents and neighbors, not Google.)
None of these seem especially relevant to folks in poor countries. What they really need is access to basic information that is already well understood (how to dig a safe well, how to prevent AIDS transmission, reliable accounting and credit practises, basic nutrition). These things are actually very well conveyed by book.
The one exception I can see is that the Internet is also good for two-way communication between people well-seperated, and places like Africa are often short of experts of one stripe or another. I can see how it might do some good if you could have interactive, or semi-interactive software, that might let a back-and-forth go on to teach people stuff better. Say, an adaptive teaching program that could teach a range of students, from the barely getting started to the most sophisticated. A book tends to be a one-size-fits-all solution, which does not serve the spectrum of students in the real world. That's why you need a teacher, too, to customize the learning. But software could, in part, replace the teacher at lower cost. Or serve as some kind of faux triage nurse that could ask some questions about your symptoms and find out whether you should just rest up and drink more fluids -- or whether you really should make that 2 day trip to the clinic. These things would be good.