I'm happy to see the discussion the review has generated. Here are some scattered replies to questions (sorry if I've missed any -- I am logging in just briefly, busy day.)
* Riding in traffic exhaust: there is no good discussion of bike-car interactions. (I recommend Effective Cycling, John Forrester, pub. MIT Press, for lots of statistical and practical information on this subject.) Unfortunately, neither EC or BS cover exhaust; I remember some studies done before gas (petrol) went unleaded in London that showed elevated lead levels in cyclists. Not sure how the trend to LEVs has helped.
* Effeciency of the bicycle design. It is definitely the case that recumbents are more efficient. This is in large part due to wind resistance. In general, designs that allow for continuous motion (e.g., circular motion of pedals) are far better than pumping (discontinuous change of direction), and that's what we've got. Attempts to make the pedal motion more efficient on the upright have not been too successful -- it seems we adapt well and smooth out the minor troubles.
* Bicycle weight: yes, I was wrong to call it a "new" trend to drill out chainwheels and generally obsess over grams. The new trend is perhaps the use of Ti and exotic compounds to lessen weight without sacrificing strength. In any case, even a one pound change in the weight (very large for the obsessives) has little effect on efficiency compared to, e.g., tire pressure, out of alignment parts, chain loss, &c&c. Lance needs it, but Lance has a team of engineers to keep everything else in check! Check it out!
* My comment on the traffic "rant." I am a huge anti-car person when it comes to urban design, and I generally agree with the conclusions of his rant. However, he makes some unsubstantiated claims about traffic flow in order to support his argument, and I just don't think they hold up. It is in stark contrast to the rest of the book, where he is very careful to cite and discuss the evidence for even the most "obvious" assertion.
* Climbing and cycling are totally the geek sports! No question. They are problem solving sports, where you combine smaller syntactical units to form original solutions to constantly changing conditions. (Martial arts fits this description very well, as well.) In cycling, the problem solving happens "offline" and during training, of course, where as martial arts and rock climbing are live. Compiling C versus interpreting LISP, I guess.
Thanks, all who contributed and will contribute! It is fun to see people's opinions, and to discover the number of/bikes.
OK, it's time for that "what's a spline" question -- how do these things actually work?
I understand how to confuse the computer -- give it a sufficiently large "number of entries" such that (n+2)*16 is larger than (2^m-1).
But how does overwriting the rest of memory allow you to gain control? Surely the "execution" pointer -- where the computer is looking next for an instruction -- is in some unpredictable place relative to the code you've written in to the heap? Is this just a way to crash the machine as I do if I accidentally reference a memory position I haven't allocated?
Do you just wait and hope that the pointer ends up in your patch of overwritten memory, and write your malicious code so that you can make sure it never leaves your space? Or is there some way to trick the machine into sending the execution pointer directly to your chunk of code?
Not to start a/. flame war, but you will note my use of the imperative mood in beginning "imagine." Of course google is not selling your e-mail transcripts. I was demonstrating why communications providers should not be allowed to indiscriminately violate privacy even in an opt-in situation.
Meanwhile, can you try an experiment for us? Mail yourself from a different account a couple of short e-mails containing keywords-bad-to-advertise-on like (sorry for the gloom):
"abortion", "miscarriage", "car accident", "suicide", "funeral"/. is curious. Will gmail give users ads for anti-abortion websites, casket discounts, "Suicide items on ebay" (actual google text ad right now)?
As far as I can tell, Gmail's biggest problem is this: "Dear son, your grandma died suddenly. Details on the funeral ASAP. Call me." On the right hand side, google text ads hawking caskets, flowers, funeral homes. It's tacky, to say the least, and I have little respect for people who are willing to let ads into their private lives to this degree.
Tackiness aside, though, if there are privacy problems, they need to be addressed. Yes, I know that Gmail is the ultimate in "opt-in." Don't like it, don't use it. This should make privacy concerns a moot point: interesting to debate, but nothing to fume about.
But google is a huge service. If Gmail is launched, people will flock to it in droves. Not just geeks, but ordinary people who have no idea how much of their private lives are lived "in plaintext." The privacy of many, many people, even those who do not use Gmail, is at stake.
Imagine, for example, a phone company that halves your rates in exchange for being allowed to sell transcripts of your phone conversations. Don't like it, don't use it -- but what about my rights to privacy when I call you? The simple answer ("don't call people with NoPrivacyPhone") is no solution at all.
So bully to the publishers, I say! I'd much rather wait a week (or more) for a long-form, considered and balanced review.
This is one of the ways publishers maintain a lock on intellectual culture. The last time I reviewed a book (in a small, but influential journal), I walked into my local university bookstore and lots of hell broke loose. I got plenty of compliments as well, but it was interesting to note that the same people who frowned at me for criticising their friend were also in control of either the distribution of review copies, or the management of the book review section.
If I have a choice between trolls and backscratchers, I'll take the trolls with a side of relish please. Slashdot has a great thing going with their book reviews, but if people notice that Slashdot is "scooping" them by publishing insider accounts, or ignoring their labor in favour of linking to an outside newspaper, they will eventually get bored and tired and stop writing.
The weird ickyness kind of pervades your post here, but the ordinary viewer might be interested to know that Turing was hounded by the British government because of his sexuality, and forced to undergo weird pseudoscience injections of estrogen.
If that isn't a creepy tragedy that inspires sympathy (and also a fascinating story), I don't know what is. Scientist saves country, is slightly off-beat, is forced into suicidal depression by same government because of said off-beatness.
Meanwhile, if you want to watch persecuted Christian characters, why not go with the other fifty million people and watch Mel Gibson's film?
I thought the tradition for book reviews was for slashdot contributors to post their own, here. It would be disappointing if we forewent this to just post a link to somewhere else.
Slashdot has a lot of interesting readers who, because they aren't tied up in the mostly non-functional reviewing world, can contribute interesting takes on whatever's come out. Online and off, most of the book reviews are either LCD "here is a book about stuff neither of us understand", or unmitigated love-ins where authors review each other in a mutual backscratch.
I would hate to see people stop writing reviews for first post on slashdot, and I would hate to see slashdot stop supporting its own review culture.
It is interesting to note the differences between 1953 and this year's crop.
Two thing that stick out the most: how electronic media have grown up -- in some ways, overtaken -- the novel for sci fi, and how the "semi-pro" zines have become real powerhouses (I'm thinking in part of the NYRSF, which I'm most familiar with.)
Sci fi in 1953, when taken seriously by adults, was really an offshoot of "ordinary" literature still, and the television and movie stuff was silly. Sci fi in 2004 has grown and splintered and now has its own conventions and genres quite set in stone. Yes, there were the space operas even back in the day, but now it seems sci fi has, for better or for worse, become something quite separate from "mainstream" literature. I think it's true that a lot of the most adventurous, genre-busting work has been in TV and film, which would have been unthinkable in 1953, even though the film medium could have been said to be "mature" by then.
While 451 could be judged in the context of other work coming out at the time, it is only the cyberpunk novels (and the occasional Handmaiden's Tale type) that get examined in that way today.
Good or bad? I like sci fi, and I like hard sci fi as well, although it is clear that the writing suffers quite a great deal. If SF was taken seriously by The Man, would it improve? Or would the attention from the book reviewers in mainstream magazines make it less adventurous?
Final question: what about Ted Chaing? What's he been doing recently?
Re:string theory *not* being tested here...
on
Testing Relativity
·
· Score: 2, Insightful
Yes, there are these discretized spacetime models, which are not string theory (although I have been working on some which can be derived from ST effects.)
However, the fundamental problem is that such effects would show up in other experiments as very large corrections. It would be interesting to see how this competes with the large-distance time delay experiments. They have two advantages over this: very large distances (megaparsec, 10^11 times longer), and much higher energies (the spacing of the lattice is expected to be very small, so in the radio and optical you just don't notice it.) Anyway, a naieve estimate might say "if you can get an order one measurement from GRBs, that is equal to a one part in 10^20 measurement in the solar system" -- taking on a very generous (i.e., small) factor of 10^9 to account for the difference in wavelengths.
Rotation of polarization measurements also come in (the lattice would "mix" the polarizations in strange ways.)
I don't know if the lattice people have done the calculations to see what the limits for solar system type tests will do for us. The problem is made a lot harder because of the presence of the sun, which complicates the models.
Again, I don't mean to dismiss this project at all, it is great that it is being done. I just think that to describe it as a test of string theory is misleading.
In the end, my fundamental issue is that string theory is not really a "theory" -- it is a collection of approaches. The 'essentials' are very abstract, and do not lead directly to phenomenological predictions in the way that Einstein's GR did. The theories that do make predictions are rather jumbled, and there is a huge need for theoretical work to see if they can predict what we already have.
Re:Still not a justification for ISS
on
Testing Relativity
·
· Score: 2, Insightful
Actually, I think doing this on the ISS decreases the ability of the experiment to produce good results in comparison with an unmanned platform.
Even forgetting, for now, the fact that the amount of money you can spend on the apparatus itself is reduced because you have to pay for the astronauts and their safety --
The movement of astronauts on the ISS is going to generate vibrations in the structure, which will be transmitted to the two interferometers. Those vibrations will lead to a loss in your ability to maintain a precise phase between the two detectors.
I'm not sure if the effects are important enough -- it depends on the wavelengths they are using. But it is the case that the presence of astronauts on the station will limit the smallest wavelength you can go down to, and thus limit the effectiveness of this kind of study.
string theory *not* being tested here...
on
Testing Relativity
·
· Score: 4, Insightful
Except in a few cases. The article seems to be more than a little cheerlead-y.
String theory predicts deviations from General Relativity at very high energies and very small distances. I would be very surprised to read of a string theory model -- or class of models -- that predicted solar system scale effects in their basic framework. The importance of string theory effects is suppressed by a huge factor depending on the local energy density in the experiment you are testing. The string energy scale is so far away that it would be a great coincidence if it just barely showed up in the solar system but did not, e.g. rip it all apart. Sort of like crashing your car through the window of a bookstore and having the resultant mess just precisely turn one page of one book.
This is not a bad experiment to do, because there are theories -- mostly cosmological ones -- that predict differences in gravity that would show up in this theory, but they are definitely non-standard modifications to particular theories. I have done work on these kinds of theories, and let me tell you, it is a certain amount of work to actually generate theories that even care about such low energies and large distances that you can test them even with an "ultimate" measurement.
I am disappointed by the rather slipshod understanding of science and the issues that this article represents. "Evicting Einstein" is a sensationalistic headline, and it's just not true -- as anyone will tell you, Newton was not "impeached." A much better angle that this article could have taken was that of exploration of gravity, as opposed to "putting the chalk scribblers in place."
Firstly, your description of a Killing vector/field, as you are perfectly aware, doesn't capture their essential meaning, and certainly isn't acceptable to a mathematician even as a broad overview. Sure, you make a good stab at an introduction, but now try and describe Chapter 13 of Weinberg's "General Theory of Relativity" using your cartoons. It's not possible, regardless of how you try to dress it up.
Actually, with a few extra words (like including subsets of the path), what I wrote is a workable definition of Killing vectors as they are used in General Relativity. A few more words, and I'd cover even pathological cases.
Weinberg's book eschews the whole talk of manifolds and makes it rather hard to see the issues in a cartoony way. I prefer MTW. Ironically, I think your point would be better made for particle physics, which is a lot less tractable in cartoons. I like those books a lot less, and have not seen a satisfactory one, because it takes a mind the order of Feynman's to really get at the heart of the issues without a huge amount of notation.
Furthermore, had you bothered to address the part of my post that came after "Popularizations, sadly, do have a place in our world," you'd have to admit that I wasn't being condescending at all. Rather, it was a roundabout way of lamenting the continuing fall in the numbers of students deciding to pursue careers in the sciences.
I object to the attitude of your posts on this subject. Scientists have had enough trouble in the past for their arrogance, and I think these popularizations are the best possible news. Essential aspects of high-level science can be conveyed with a minimum of mathematics by those with the skill, and that should be praised to the skies, not damned.
By the way, I know many people around here who loved Greene's book but were less keen on the PBS version. I haven't seen it, but I suggest you give Greene a chance, and try to distinguish between gosh-whiz BS and serious efforts at popularization. Look at the regulars on sci.physics.research, and the amount you can accomplish with a minimum of required background.
"Try asking someone to explain what a Killing vector is, as well as its relation to diffeomorphism-invariance, and I can *guarantee* that they won't be able to do it adequately without the use of mathematics."
You're on.
Killing vectors: take a clock with you on your spaceship. Fly from star A to star B, and time it. Now, you get a new mission: fly from star A to star B along a slightly altered path, displaced at each point by a small amount determined by a particular vector field. (Yeah, in my million dollar PBS special, I get to have cartoons to make this easier to see.) To your amazement, your clock measures the same time!
Some spacetimes have the strange property that you can define an entire vector field (cue cartoon) with this property. Some spacetimes are even weirder! They have multiple vector fields like this. Actually, perhaps they are no so weird (cue cartoon of flat space.)
"diffeomorphism-invariance" is a scary-sounding way to express the fact that physical facts are unchanged by a shift in coordinate system. If I were to write a book, I'd start with the idea of gauge invariance and the difference between coordinates and events.
"You [me] wrote: "Plenty of other physicists and astronomers I work with have read it just to get a sense for what is going on in this rather abstracted part of physics they don't have the time to catch up on." That's all well and good. I'm simply saying that people who work as theoretical physicists/mathemagicians wouldn't read it since they'd never stop picking holes in it. Popularizations, sadly, do have a place in our world."
By "sadly," you seem to imply that anybody -- including other scientists -- who is not at the tail end of a Ph. D. in string theory (which is the level we are talking about) -- is in a sad state. I'm sorry, but this is just crazy.
Your attitude is sufficiently condescending as to be absolutely insane.
Oh, man, there's so much wrong with this comment, I don't know where to begin.
First off, this stuff is hard. No, really hard. You need to focus down and study this stuff for years before you can really get up to speed and read the preprint server with any understanding. Oh, and did I mention those years need to be spent hanging around people who already understand it?
Secondly, overviews are hard -- and hard to write well. The value of a good overview is respected by everyone in the field. You need maps with different grades of detail.
Thirdly, first hand experience: at least one of the string theory people I know read the first Greene book when she was starting out and loved it. Plenty of other physicists and astronomers I work with have read it just to get a sense for what is going on in this rather abstracted part of physics they don't have the time to catch up on.
Oh, and fourthly, this kind of book does wonders for scientific literacy and interest in the general public. Selfishly, it helps build the case for continued public funding of this kind of thing. Better put, it is a sort of 'return' to the public that repays them for their support by working hard to generate a story both intellectually respectable and comprehensible to the educated and motivated layman.
Actually, given the recent flurry of well written books on cosmology (Greene is the tip of the iceberg -- more than a few astronomers have written about the recent renaissance [naissance?] of the field), I am really hoping to see something like this.
I've definitely noticed that ideas tend to propagate into literature pretty quickly these days. "Chaos theory" is, well, you can't spit in a bookstore without hitting fiction that mentions it. Unfortunately, chaos theory is a bit of niche (potentially rich, don't want to offend anyone) -- whereas the recent science shelves have been much more wide ranging. The Gribben book came out in (um) 1989?
Hopefully we will get some good Stephenson-esque stuff out of this when authors get around to reading the new crop. I hope we get stuff that actually engages with the analogies and metaphors and the mechanisms of the ideas, as opposed to stuff that just renames the 'warp drive' the 'dilaton drive.' It really is wild.
It is strange to me that science-R&D is so poorly rated for happiness. It probably has something to do with the highly graded status within the occupation. All plumbers are mostly equal; it's not like at the plumbers convention some of the plumbers are considered great geniuses of the plumbing trade while others are the plumbers who 'never made it.'
The relative autonomy of vocational trades probably has a lot to do with it -- you're not continually comparing your performance / rate of professional advancement / success to others.
This thread is most over, but I thought I'd toss my quarter in the hat:
Beginning in a round about way, the biggest mistake most college students make is not taking a year off before going to college. A year off working, travelling (if you can afford it), half-working-half-playing -- etc. The break is an important time to focus the mind temporarily towards the ultra-long term questions of where you want to end up and what you want to be doing when you end up there.
That kind of advice applies, I think, to many points in your life. When work isn't getting it together for you (and if you've been searching on the side, this seems to be the case for you as well) if you have the opportunity for "time away" in a half-fantasy world, I highly recommend it.
If you haven't had the chance to see what the life is like as a street busker / Great American Novel-writer / inspirational high school teacher / iron sculptor / Trans-American hiker / social worker / ___ , you will probably harbour a lot of illusions about it. If you think you have a pretty good idea from theoretical musing, go out and find some of those people, and tell them what you think their lives are like. Don't be surprised if you get flamed.
Maybe that life is for you, but the only way to know is to give it a shot. The danger of not following this advice is that you will find other, more destructive ways to "act out" your unrealized goals. Into this camp fall thousands of bitter people you have probably worked with / run into.
Depending on your career path -- and I think both tech and R&D fall into this category -- you can sometimes feel "streamed." A good amount of talent and hard work can throw up opportunity and advancement much faster than you can really form a distinct desire for them. In contrast to, say, writers, who may spend years before they see themselves making money from their output, tech and R&D people usually can get work in their field with far less effort -- you don't see many starving UNIX admins reading the AOCP in a garret.
The downside to this relative bounty is that it is quite possible to end up in a place that you once wanted, but that your desires have left.
After I didn't take a year off before college, I learned that lesson the hard way (and had a very unfocused freshman year.) Before going to grad school Ph. D. program, I found funding for a one year free-form masters where I did a bunch of the things I thought I might want to do instead. On either side of that, I spent a few months teaching high school and working alongside much older people who made a (not fantastic) income as artists. I came in with a much better attitude and understanding of a wider range of 'possible lives' than I would have otherwise.
Academia is generally much more tolerant of these sorts of things, but I have heard similar stories to mine from the private sector. Even if you do take a small hit in prestige points, you will easily win them back -- which is much better than a long, slow downward spiral.
slammed by more than a few...
on
Everything and More
·
· Score: 4, Informative
I am a huge David Foster Wallace fan, and think Infinite Jest is the greatest book of the 1990s.
However, the reports on Everything and More have not been good. The reviewers who have demonstrated some understanding of the mathematics involved (not particular heavy, but somewhat obscure), have come down pretty hard on DFW for his errors. Here is a representative review (from the LRB), which covers DFW's book and a slew of other "books on infinity" at once:
"As for Wallace's book, the less said, the better. It's a sloppy production, including neither an index nor a table of contents, and after a while his breezy style grates. No one who is unfamiliar with the ideas behind his dense, user-unfriendly mathematical expositions could work their way through them to gain any insight into what he is talking about. Worse, anyone who is already familiar with these ideas will see that his expositions are often riddled with mistakes. The sections on set theory, in particular, are a disaster."
(You might put this down to academic anxiety, since the reviewer, A. W. Moore, is a professional philosopher with an anthology on "infinity" to his name as well.)
It is strange, since DFW did spend part of his youth (not the alcohol and drug-addicted part) in a philosophy and logic Ph.D. program. I'm not sure if I'll read it; on the bright side, he has a new collection of short stories coming out in June.
(for example) is rich enough and (because of mandatory service) widespread enough to make its appearance in more "literary" contexts; Hebrew is also very amenable to nickname and acronym formation (even long-dead Torah scholars are commonly referred to by strange semi-acronymistic [?] names.)
I wonder though... I highly doubt Apple has designed the iPod to spin up the drive, buffer large amounts of data, then spin it back down. That would be the only possible way I could see it designed to allow for "non-harddrive" (as in not spinning) operational periods during use. (It'd also dramatically increase random-access times).
But, that's exactly how it works! It spins up, caches a few songs, and then spins down. If you want to fast forward through more than a few songs, you do sit around waiting for the drive to spin up.
One of the first times I took my ipod out was on a long bike ride. I was heading down a gorgeous hill near the Delaware river with nobody in sight, so I decided to go for it.
Around about the time I broke 35 mph, my ipod lept out of my pocket, hit the asphalt, and started bouncing. I ground to a halt. Still running!
Yeah, I probably got lucky because it wasn't accessing the hard drive when it hit. (And, it was in the little fabric case.) It's been five months, and it still runs fine, however.
I am talking here about the group velocity. All plasma waves have a group velocity slower than the speed of light in vacuum; it is the difference in group velocity between different frequencies that leads to the whistler effect. (In this regieme, the phase and group velocities happen to be roughly proportional.)
You can read all about it in Jackson's Classical Electrodynamics. Whistlers are quite beautiful; I first heard them when I started listening to shortwave radio.
The ionosphere is a plasma: it is made up of electrons and protons that are sufficiently diffuse that they spend most of their time unbound from each other (incident radiation keeps ionizing atoms; once their ionized, the two particles have a great deal of trouble finding each other.)
Because light waves are just propagating electric and magnetic fields, they interact with charged particles. In solids (and most gases, where the electrons are bound), they don't really "see" the charge: the electrons are very close to the protons and the configuration "looks" neutral. It is only at very shortwavelength that the light can "discover" that these atoms aren't neutral, but have, rather, two charged components. In plasmas, where the electrons and protons are very far apart, much longer wavelength -- radio -- waves discover this fact.
The particular density and temperature of the plasma defines a resonant frequency. Near this resonance, the radio waves have difficulty propagating: they start to interact strongly with the medium. Roughly (OK, very roughly) speaking, the stronger the interaction, the faster the radio wave propagates.
But wait! I thought light travelled at a constant speed regardless of wavelength! Well, yes, in vacuo. Here, because of the interactions with the plasma, all frequencies of interest are slowed down. Off-resonant frequencies are slowed more, because the plasma moves about in response to the fields of the light wave to partly cancel the wavefront (imagine trying to send a pressure wave through silly putty.)
The whistler comes about when you pump a large spectrum of radio frequencies into the ionosphere from something like a thunderclap. These waves then start travelling over the Earth. The higher frequency waves are closer to the resonant frequency of the ionosphere plasma, and so get to you sooner. You literally have to wait for the lower frequency waves to arrive.
The plasma is sufficiently ''dispersive'' that, for average ionosphere conditions, the lowest and highest frequency waves will be 10,000 km apart. A pretty large musical instrument!
A final, neat point: while the whistlers are somewhat like a prisoner tapping on a radiator pipe, you can under certain circumstances get enough energy into the ultra-long wavelength modes to get a cavity resonance (make the Earth's ionosphere ring like a bell.) Could it be? Yes, indeed: this cavity resonance is detected after the detonantion of nuclear bomb (the EMP provides the lightning strike.)
After the power plant deregulations plunged us all into darkness, you thought we might have learned something. But no, not the ever liberatarian computer scientists! (The only time their free market dedication wavers is on the subcontinent.)
Companies make money by pushing the envelope. They take calculated gambles on what they produce. This is a good thing: nothing ventured, nothing gained -- especially when you are using and developing techniques and technologies that have never been seen before. We have invented the 'corporation' to allow people to do this sort of thing at less risk: you can gamble millions of dollars (if you can convince people you're worth the risk) and come out the other end more or less OK regardless.
But there are some things you shouldn't be allowed to gamble with. You shouldn't gamble with water quality (how much profit can we make if we have a 10^-4 risk of Hg contamination?) You shouldn't gamble with power line reliability. You should be allowed to gamble on software reliability -- except in life support or military applications. Go crazy with your new distributed quantum computing net, but don't put it in grandma's pacemaker or a GI's helicopter until you can satisfy certain politically defined standards. Who decides what you can and can't gamble on? Amazingly, the voters.
The voters, in their wisdom, decided to make 911 service -- and the E911 extension -- something that you couldn't dispense with. They figured that the social good of being able to track down and solve emergencies at the source was more important than a few months of lower profits for Vonage et al. Disagree if you wish, but to declare all regulation off limits is to ignore the fact that some regulation is a necessary good.
Interstellar scintillation is really neat and interesting. You can use its properties to measure the nature of the interstellar medium indirectly -- for example, from the dispersion of a pulsar signal (a pulsar emits something a little like a square wave pulse in the radio, which is then smudged and smeared out as different wavelengths scintillate differently.)
I studied I.S. a little bit awhile back. Carl Sagan did some work on scintillation; the scintillation effect can pull out a distant radio signal by gathering in rays from a lot of different directions and accidentally throwing them right at you. The famous WOW signal, I believe, was investigated as an example of scintillation from a big cloud much like the ones described in the article.
It is interesting to see this technique used to do radio astronomy. Most of the times when you encounter a natural lens, it is sufficiently weird that you use the observation to analyse the lens itself, and not what it happens to be magnifying. Gravitational lenses are interesting in large part because you can try to figure out the distribution of dark matter in the lens itself -- and not because you can use it to "see into" the object being lensed. These lenses are not exactly perfect optics -- they're more like balls of glass, which distort and differentially magnify something behind.
But I'm not as familiar any more with radio astronomy. It is definitely possible that we understand enough about the properties of the ISM that the more interesting problem of figuring out the properties of the background object is open for work. Very cool!
Slashdot Mirror
* Riding in traffic exhaust: there is no good discussion of bike-car interactions. (I recommend Effective Cycling, John Forrester, pub. MIT Press, for lots of statistical and practical information on this subject.) Unfortunately, neither EC or BS cover exhaust; I remember some studies done before gas (petrol) went unleaded in London that showed elevated lead levels in cyclists. Not sure how the trend to LEVs has helped.
* Effeciency of the bicycle design. It is definitely the case that recumbents are more efficient. This is in large part due to wind resistance. In general, designs that allow for continuous motion (e.g., circular motion of pedals) are far better than pumping (discontinuous change of direction), and that's what we've got. Attempts to make the pedal motion more efficient on the upright have not been too successful -- it seems we adapt well and smooth out the minor troubles.
* Bicycle weight: yes, I was wrong to call it a "new" trend to drill out chainwheels and generally obsess over grams. The new trend is perhaps the use of Ti and exotic compounds to lessen weight without sacrificing strength. In any case, even a one pound change in the weight (very large for the obsessives) has little effect on efficiency compared to, e.g., tire pressure, out of alignment parts, chain loss, &c&c. Lance needs it, but Lance has a team of engineers to keep everything else in check! Check it out!
* My comment on the traffic "rant." I am a huge anti-car person when it comes to urban design, and I generally agree with the conclusions of his rant. However, he makes some unsubstantiated claims about traffic flow in order to support his argument, and I just don't think they hold up. It is in stark contrast to the rest of the book, where he is very careful to cite and discuss the evidence for even the most "obvious" assertion.
* Climbing and cycling are totally the geek sports! No question. They are problem solving sports, where you combine smaller syntactical units to form original solutions to constantly changing conditions. (Martial arts fits this description very well, as well.) In cycling, the problem solving happens "offline" and during training, of course, where as martial arts and rock climbing are live. Compiling C versus interpreting LISP, I guess.
Thanks, all who contributed and will contribute! It is fun to see people's opinions, and to discover the number of /bikes.
I understand how to confuse the computer -- give it a sufficiently large "number of entries" such that (n+2)*16 is larger than (2^m-1).
But how does overwriting the rest of memory allow you to gain control? Surely the "execution" pointer -- where the computer is looking next for an instruction -- is in some unpredictable place relative to the code you've written in to the heap? Is this just a way to crash the machine as I do if I accidentally reference a memory position I haven't allocated?
Do you just wait and hope that the pointer ends up in your patch of overwritten memory, and write your malicious code so that you can make sure it never leaves your space? Or is there some way to trick the machine into sending the execution pointer directly to your chunk of code?
Not to start a /. flame war, but you will note my use of the imperative mood in beginning "imagine." Of course google is not selling your e-mail transcripts. I was demonstrating why communications providers should not be allowed to indiscriminately violate privacy even in an opt-in situation.
Meanwhile, can you try an experiment for us? Mail yourself from a different account a couple of short e-mails containing keywords-bad-to-advertise-on like (sorry for the gloom):
"abortion", "miscarriage", "car accident", "suicide", "funeral" /. is curious. Will gmail give users ads for anti-abortion websites, casket discounts, "Suicide items on ebay" (actual google text ad right now)?
Tackiness aside, though, if there are privacy problems, they need to be addressed. Yes, I know that Gmail is the ultimate in "opt-in." Don't like it, don't use it. This should make privacy concerns a moot point: interesting to debate, but nothing to fume about.
But google is a huge service. If Gmail is launched, people will flock to it in droves. Not just geeks, but ordinary people who have no idea how much of their private lives are lived "in plaintext." The privacy of many, many people, even those who do not use Gmail, is at stake.
Imagine, for example, a phone company that halves your rates in exchange for being allowed to sell transcripts of your phone conversations. Don't like it, don't use it -- but what about my rights to privacy when I call you? The simple answer ("don't call people with NoPrivacyPhone") is no solution at all.
This is one of the ways publishers maintain a lock on intellectual culture. The last time I reviewed a book (in a small, but influential journal), I walked into my local university bookstore and lots of hell broke loose. I got plenty of compliments as well, but it was interesting to note that the same people who frowned at me for criticising their friend were also in control of either the distribution of review copies, or the management of the book review section.
If I have a choice between trolls and backscratchers, I'll take the trolls with a side of relish please. Slashdot has a great thing going with their book reviews, but if people notice that Slashdot is "scooping" them by publishing insider accounts, or ignoring their labor in favour of linking to an outside newspaper, they will eventually get bored and tired and stop writing.
If that isn't a creepy tragedy that inspires sympathy (and also a fascinating story), I don't know what is. Scientist saves country, is slightly off-beat, is forced into suicidal depression by same government because of said off-beatness.
Meanwhile, if you want to watch persecuted Christian characters, why not go with the other fifty million people and watch Mel Gibson's film?
Slashdot has a lot of interesting readers who, because they aren't tied up in the mostly non-functional reviewing world, can contribute interesting takes on whatever's come out. Online and off, most of the book reviews are either LCD "here is a book about stuff neither of us understand", or unmitigated love-ins where authors review each other in a mutual backscratch.
I would hate to see people stop writing reviews for first post on slashdot, and I would hate to see slashdot stop supporting its own review culture.
Two thing that stick out the most: how electronic media have grown up -- in some ways, overtaken -- the novel for sci fi, and how the "semi-pro" zines have become real powerhouses (I'm thinking in part of the NYRSF, which I'm most familiar with.)
Sci fi in 1953, when taken seriously by adults, was really an offshoot of "ordinary" literature still, and the television and movie stuff was silly. Sci fi in 2004 has grown and splintered and now has its own conventions and genres quite set in stone. Yes, there were the space operas even back in the day, but now it seems sci fi has, for better or for worse, become something quite separate from "mainstream" literature. I think it's true that a lot of the most adventurous, genre-busting work has been in TV and film, which would have been unthinkable in 1953, even though the film medium could have been said to be "mature" by then.
While 451 could be judged in the context of other work coming out at the time, it is only the cyberpunk novels (and the occasional Handmaiden's Tale type) that get examined in that way today.
Good or bad? I like sci fi, and I like hard sci fi as well, although it is clear that the writing suffers quite a great deal. If SF was taken seriously by The Man, would it improve? Or would the attention from the book reviewers in mainstream magazines make it less adventurous?
Final question: what about Ted Chaing? What's he been doing recently?
However, the fundamental problem is that such effects would show up in other experiments as very large corrections. It would be interesting to see how this competes with the large-distance time delay experiments. They have two advantages over this: very large distances (megaparsec, 10^11 times longer), and much higher energies (the spacing of the lattice is expected to be very small, so in the radio and optical you just don't notice it.) Anyway, a naieve estimate might say "if you can get an order one measurement from GRBs, that is equal to a one part in 10^20 measurement in the solar system" -- taking on a very generous (i.e., small) factor of 10^9 to account for the difference in wavelengths.
Rotation of polarization measurements also come in (the lattice would "mix" the polarizations in strange ways.)
I don't know if the lattice people have done the calculations to see what the limits for solar system type tests will do for us. The problem is made a lot harder because of the presence of the sun, which complicates the models.
Again, I don't mean to dismiss this project at all, it is great that it is being done. I just think that to describe it as a test of string theory is misleading.
In the end, my fundamental issue is that string theory is not really a "theory" -- it is a collection of approaches. The 'essentials' are very abstract, and do not lead directly to phenomenological predictions in the way that Einstein's GR did. The theories that do make predictions are rather jumbled, and there is a huge need for theoretical work to see if they can predict what we already have.
Even forgetting, for now, the fact that the amount of money you can spend on the apparatus itself is reduced because you have to pay for the astronauts and their safety --
The movement of astronauts on the ISS is going to generate vibrations in the structure, which will be transmitted to the two interferometers. Those vibrations will lead to a loss in your ability to maintain a precise phase between the two detectors.
I'm not sure if the effects are important enough -- it depends on the wavelengths they are using. But it is the case that the presence of astronauts on the station will limit the smallest wavelength you can go down to, and thus limit the effectiveness of this kind of study.
String theory predicts deviations from General Relativity at very high energies and very small distances. I would be very surprised to read of a string theory model -- or class of models -- that predicted solar system scale effects in their basic framework. The importance of string theory effects is suppressed by a huge factor depending on the local energy density in the experiment you are testing. The string energy scale is so far away that it would be a great coincidence if it just barely showed up in the solar system but did not, e.g. rip it all apart. Sort of like crashing your car through the window of a bookstore and having the resultant mess just precisely turn one page of one book.
This is not a bad experiment to do, because there are theories -- mostly cosmological ones -- that predict differences in gravity that would show up in this theory, but they are definitely non-standard modifications to particular theories. I have done work on these kinds of theories, and let me tell you, it is a certain amount of work to actually generate theories that even care about such low energies and large distances that you can test them even with an "ultimate" measurement.
I am disappointed by the rather slipshod understanding of science and the issues that this article represents. "Evicting Einstein" is a sensationalistic headline, and it's just not true -- as anyone will tell you, Newton was not "impeached." A much better angle that this article could have taken was that of exploration of gravity, as opposed to "putting the chalk scribblers in place."
Actually, with a few extra words (like including subsets of the path), what I wrote is a workable definition of Killing vectors as they are used in General Relativity. A few more words, and I'd cover even pathological cases.
Weinberg's book eschews the whole talk of manifolds and makes it rather hard to see the issues in a cartoony way. I prefer MTW. Ironically, I think your point would be better made for particle physics, which is a lot less tractable in cartoons. I like those books a lot less, and have not seen a satisfactory one, because it takes a mind the order of Feynman's to really get at the heart of the issues without a huge amount of notation.
Furthermore, had you bothered to address the part of my post that came after "Popularizations, sadly, do have a place in our world," you'd have to admit that I wasn't being condescending at all. Rather, it was a roundabout way of lamenting the continuing fall in the numbers of students deciding to pursue careers in the sciences.
I object to the attitude of your posts on this subject. Scientists have had enough trouble in the past for their arrogance, and I think these popularizations are the best possible news. Essential aspects of high-level science can be conveyed with a minimum of mathematics by those with the skill, and that should be praised to the skies, not damned.
By the way, I know many people around here who loved Greene's book but were less keen on the PBS version. I haven't seen it, but I suggest you give Greene a chance, and try to distinguish between gosh-whiz BS and serious efforts at popularization. Look at the regulars on sci.physics.research, and the amount you can accomplish with a minimum of required background.
You're on.
Killing vectors: take a clock with you on your spaceship. Fly from star A to star B, and time it. Now, you get a new mission: fly from star A to star B along a slightly altered path, displaced at each point by a small amount determined by a particular vector field. (Yeah, in my million dollar PBS special, I get to have cartoons to make this easier to see.) To your amazement, your clock measures the same time!
Some spacetimes have the strange property that you can define an entire vector field (cue cartoon) with this property. Some spacetimes are even weirder! They have multiple vector fields like this. Actually, perhaps they are no so weird (cue cartoon of flat space.)
"diffeomorphism-invariance" is a scary-sounding way to express the fact that physical facts are unchanged by a shift in coordinate system. If I were to write a book, I'd start with the idea of gauge invariance and the difference between coordinates and events.
"You [me] wrote: "Plenty of other physicists and astronomers I work with have read it just to get a sense for what is going on in this rather abstracted part of physics they don't have the time to catch up on." That's all well and good. I'm simply saying that people who work as theoretical physicists/mathemagicians wouldn't read it since they'd never stop picking holes in it. Popularizations, sadly, do have a place in our world."
By "sadly," you seem to imply that anybody -- including other scientists -- who is not at the tail end of a Ph. D. in string theory (which is the level we are talking about) -- is in a sad state. I'm sorry, but this is just crazy.
Your attitude is sufficiently condescending as to be absolutely insane.
First off, this stuff is hard. No, really hard. You need to focus down and study this stuff for years before you can really get up to speed and read the preprint server with any understanding. Oh, and did I mention those years need to be spent hanging around people who already understand it?
Secondly, overviews are hard -- and hard to write well. The value of a good overview is respected by everyone in the field. You need maps with different grades of detail.
Thirdly, first hand experience: at least one of the string theory people I know read the first Greene book when she was starting out and loved it. Plenty of other physicists and astronomers I work with have read it just to get a sense for what is going on in this rather abstracted part of physics they don't have the time to catch up on.
Oh, and fourthly, this kind of book does wonders for scientific literacy and interest in the general public. Selfishly, it helps build the case for continued public funding of this kind of thing. Better put, it is a sort of 'return' to the public that repays them for their support by working hard to generate a story both intellectually respectable and comprehensible to the educated and motivated layman.
I've definitely noticed that ideas tend to propagate into literature pretty quickly these days. "Chaos theory" is, well, you can't spit in a bookstore without hitting fiction that mentions it. Unfortunately, chaos theory is a bit of niche (potentially rich, don't want to offend anyone) -- whereas the recent science shelves have been much more wide ranging. The Gribben book came out in (um) 1989?
Hopefully we will get some good Stephenson-esque stuff out of this when authors get around to reading the new crop. I hope we get stuff that actually engages with the analogies and metaphors and the mechanisms of the ideas, as opposed to stuff that just renames the 'warp drive' the 'dilaton drive.' It really is wild.
The relative autonomy of vocational trades probably has a lot to do with it -- you're not continually comparing your performance / rate of professional advancement / success to others.
Beginning in a round about way, the biggest mistake most college students make is not taking a year off before going to college. A year off working, travelling (if you can afford it), half-working-half-playing -- etc. The break is an important time to focus the mind temporarily towards the ultra-long term questions of where you want to end up and what you want to be doing when you end up there.
That kind of advice applies, I think, to many points in your life. When work isn't getting it together for you (and if you've been searching on the side, this seems to be the case for you as well) if you have the opportunity for "time away" in a half-fantasy world, I highly recommend it.
If you haven't had the chance to see what the life is like as a street busker / Great American Novel-writer / inspirational high school teacher / iron sculptor / Trans-American hiker / social worker / ___ , you will probably harbour a lot of illusions about it. If you think you have a pretty good idea from theoretical musing, go out and find some of those people, and tell them what you think their lives are like. Don't be surprised if you get flamed.
Maybe that life is for you, but the only way to know is to give it a shot. The danger of not following this advice is that you will find other, more destructive ways to "act out" your unrealized goals. Into this camp fall thousands of bitter people you have probably worked with / run into.
Depending on your career path -- and I think both tech and R&D fall into this category -- you can sometimes feel "streamed." A good amount of talent and hard work can throw up opportunity and advancement much faster than you can really form a distinct desire for them. In contrast to, say, writers, who may spend years before they see themselves making money from their output, tech and R&D people usually can get work in their field with far less effort -- you don't see many starving UNIX admins reading the AOCP in a garret.
The downside to this relative bounty is that it is quite possible to end up in a place that you once wanted, but that your desires have left.
After I didn't take a year off before college, I learned that lesson the hard way (and had a very unfocused freshman year.) Before going to grad school Ph. D. program, I found funding for a one year free-form masters where I did a bunch of the things I thought I might want to do instead. On either side of that, I spent a few months teaching high school and working alongside much older people who made a (not fantastic) income as artists. I came in with a much better attitude and understanding of a wider range of 'possible lives' than I would have otherwise.
Academia is generally much more tolerant of these sorts of things, but I have heard similar stories to mine from the private sector. Even if you do take a small hit in prestige points, you will easily win them back -- which is much better than a long, slow downward spiral.
However, the reports on Everything and More have not been good. The reviewers who have demonstrated some understanding of the mathematics involved (not particular heavy, but somewhat obscure), have come down pretty hard on DFW for his errors. Here is a representative review (from the LRB), which covers DFW's book and a slew of other "books on infinity" at once:
"As for Wallace's book, the less said, the better. It's a sloppy production, including neither an index nor a table of contents, and after a while his breezy style grates. No one who is unfamiliar with the ideas behind his dense, user-unfriendly mathematical expositions could work their way through them to gain any insight into what he is talking about. Worse, anyone who is already familiar with these ideas will see that his expositions are often riddled with mistakes. The sections on set theory, in particular, are a disaster."
(You might put this down to academic anxiety, since the reviewer, A. W. Moore, is a professional philosopher with an anthology on "infinity" to his name as well.)
It is strange, since DFW did spend part of his youth (not the alcohol and drug-addicted part) in a philosophy and logic Ph.D. program. I'm not sure if I'll read it; on the bright side, he has a new collection of short stories coming out in June.
(for example) is rich enough and (because of mandatory service) widespread enough to make its appearance in more "literary" contexts; Hebrew is also very amenable to nickname and acronym formation (even long-dead Torah scholars are commonly referred to by strange semi-acronymistic [?] names.)
But, that's exactly how it works! It spins up, caches a few songs, and then spins down. If you want to fast forward through more than a few songs, you do sit around waiting for the drive to spin up.
Around about the time I broke 35 mph, my ipod lept out of my pocket, hit the asphalt, and started bouncing. I ground to a halt. Still running!
Yeah, I probably got lucky because it wasn't accessing the hard drive when it hit. (And, it was in the little fabric case.) It's been five months, and it still runs fine, however.
I am talking here about the group velocity. All plasma waves have a group velocity slower than the speed of light in vacuum; it is the difference in group velocity between different frequencies that leads to the whistler effect. (In this regieme, the phase and group velocities happen to be roughly proportional.)
The ionosphere is a plasma: it is made up of electrons and protons that are sufficiently diffuse that they spend most of their time unbound from each other (incident radiation keeps ionizing atoms; once their ionized, the two particles have a great deal of trouble finding each other.)
Because light waves are just propagating electric and magnetic fields, they interact with charged particles. In solids (and most gases, where the electrons are bound), they don't really "see" the charge: the electrons are very close to the protons and the configuration "looks" neutral. It is only at very shortwavelength that the light can "discover" that these atoms aren't neutral, but have, rather, two charged components. In plasmas, where the electrons and protons are very far apart, much longer wavelength -- radio -- waves discover this fact.
The particular density and temperature of the plasma defines a resonant frequency. Near this resonance, the radio waves have difficulty propagating: they start to interact strongly with the medium. Roughly (OK, very roughly) speaking, the stronger the interaction, the faster the radio wave propagates.
But wait! I thought light travelled at a constant speed regardless of wavelength! Well, yes, in vacuo. Here, because of the interactions with the plasma, all frequencies of interest are slowed down. Off-resonant frequencies are slowed more, because the plasma moves about in response to the fields of the light wave to partly cancel the wavefront (imagine trying to send a pressure wave through silly putty.)
The whistler comes about when you pump a large spectrum of radio frequencies into the ionosphere from something like a thunderclap. These waves then start travelling over the Earth. The higher frequency waves are closer to the resonant frequency of the ionosphere plasma, and so get to you sooner. You literally have to wait for the lower frequency waves to arrive.
The plasma is sufficiently ''dispersive'' that, for average ionosphere conditions, the lowest and highest frequency waves will be 10,000 km apart. A pretty large musical instrument!
A final, neat point: while the whistlers are somewhat like a prisoner tapping on a radiator pipe, you can under certain circumstances get enough energy into the ultra-long wavelength modes to get a cavity resonance (make the Earth's ionosphere ring like a bell.) Could it be? Yes, indeed: this cavity resonance is detected after the detonantion of nuclear bomb (the EMP provides the lightning strike.)
Companies make money by pushing the envelope. They take calculated gambles on what they produce. This is a good thing: nothing ventured, nothing gained -- especially when you are using and developing techniques and technologies that have never been seen before. We have invented the 'corporation' to allow people to do this sort of thing at less risk: you can gamble millions of dollars (if you can convince people you're worth the risk) and come out the other end more or less OK regardless.
But there are some things you shouldn't be allowed to gamble with. You shouldn't gamble with water quality (how much profit can we make if we have a 10^-4 risk of Hg contamination?) You shouldn't gamble with power line reliability. You should be allowed to gamble on software reliability -- except in life support or military applications. Go crazy with your new distributed quantum computing net, but don't put it in grandma's pacemaker or a GI's helicopter until you can satisfy certain politically defined standards. Who decides what you can and can't gamble on? Amazingly, the voters.
The voters, in their wisdom, decided to make 911 service -- and the E911 extension -- something that you couldn't dispense with. They figured that the social good of being able to track down and solve emergencies at the source was more important than a few months of lower profits for Vonage et al. Disagree if you wish, but to declare all regulation off limits is to ignore the fact that some regulation is a necessary good.
I studied I.S. a little bit awhile back. Carl Sagan did some work on scintillation; the scintillation effect can pull out a distant radio signal by gathering in rays from a lot of different directions and accidentally throwing them right at you. The famous WOW signal, I believe, was investigated as an example of scintillation from a big cloud much like the ones described in the article.
It is interesting to see this technique used to do radio astronomy. Most of the times when you encounter a natural lens, it is sufficiently weird that you use the observation to analyse the lens itself, and not what it happens to be magnifying. Gravitational lenses are interesting in large part because you can try to figure out the distribution of dark matter in the lens itself -- and not because you can use it to "see into" the object being lensed. These lenses are not exactly perfect optics -- they're more like balls of glass, which distort and differentially magnify something behind.
But I'm not as familiar any more with radio astronomy. It is definitely possible that we understand enough about the properties of the ISM that the more interesting problem of figuring out the properties of the background object is open for work. Very cool!