Slashdot Mirror
350-Year-Old Newton's Puzzle Solved By 16-Year-Old
First time accepted submitter johnsnails writes "A German 16-year-old, Shouryya Ray, solved two fundamental particle dynamic theories posed by Sir Isaac Newton, which until recently required the use of powerful computers. He worked out how to calculate exactly the path of a projectile under gravity and subject to air resistance. Shouryya solved the problem while working on a school project. From the article: 'Mr Ray won a research award for his efforts and has been labeled a genius by the German media, but he put it down to "curiosity and schoolboy naivety." "When it was explained to us that the problems had no solutions, I thought to myself, 'well, there's no harm in trying,'" he said.'"
You forgot a lot of things:
-gravity is not a constant vector force downward. It is a radial force inward toward the center of the Earth, and its intensity varies with altitude.
-air resistance is not constant either. It depends on air pressure which varies with altitude as well.
-air resistance is not perfectly proportional to v^2, especially at transonic and supersonic speeds.
-if the projectile is spinning, it may cause a net aerodymamic force in a direction other than -v. Like a curveball.
-the earth is a spinning frame of reference, which results in various annoying effects.
-the air is not necessarily stationary. Wind exists.
and so on.
But we don't know whether this dude accounted for any of this stuff or not, because the goddamn article doesn't tell us.
That helps a little, but still doesn't really clarify completely what he did. I'll explain a little about what I know about the projectile problem and what I can figure out about what he might have accomplished here.
In the Principia, Newton poses three closely related problems. One is projectile motion under the influence of a frictional force that's proportional to velocity (book II, section I). Next he considers the case where the friction is proportional to the square of the velocity (book II, section II), and finally the case where it's of the form av+bv^2, where a and b are constants (book II, section III). Let's call these cases 1, 2, and 3.
Case 1 is pretty straightforward. The x and y motions are decoupled, and each of the motions is governed by a first-order, linear, inhomogeneous equation.
Case 2 is actually of more physical interest than case 1 for most real-world projectiles. For example, when you toss a baseball in air, its Reynolds number is about 10^4 or 10^5, and in that regime, a force proportional to v^2 is a pretty decent approximation. There is a well known closed-form solution for the one-dimensional subcase (I actually had a student a few years back who figured it out for herself, which was impressive), which is y=A ln[cosh(t sqrt(g/A))].
A hint is that this page has a photo of him holding up a large sheet of paper with his closed-form solution on it. The equation is clearly visible, and reads g^2/(2u^2)+(alpha g/2)[v sqrt(u^2+v^2) / u^2 + arsinh |v/u|] = const. The notation isn't explained, but clearly u and v are the components of some vector, probably the velocity vector. If so, then the constant alpha has to have units of inverse meters.
This makes me think that what he's solved is the full two-dimensional version of case 2. It can't be case 3, because besides g there is only the one constant alpha appearing in his equation. If you write down the equation of motion, a=F/m=(mg-bv^2)/m=g-(b/m)v^2, the constant that naturally occurs is b/m, which has units of inverse meters. It also makes sense that his solution has a hyperbolic trig function in it, since the y(t) for the one-dimensional version of case 2 has a hyperbolic trig function in it.
If my interpretation is right, then you should get a correct one-dimensional result from his equation when u=0. Unfortunately his equation blows up to infinity in that case, so I'm not sure how to extract any sane interpretation from it. By setting alpha=0, you should also get the case with zero friction. That does sort of make sense, since it says u is a constant, which it should be in that case.
It would be interesting to see if my interpretation is right by doing a numerical simulation and seeing if his expression really does seem to be a constant of the motion.
One thing to point out is that he may not have actually solved the full problem as set by Newton. He hasn't found the equation of the trajectory in closed form (which I think was what Newton was most interested in), and he also hasn't found the position in closed form as a function of time. (This is all assuming my interpretation is right.)
Find free books.
...it does publish great papers, but does require something of a personal connection to get into... Same for The Proceedings of the National Academy of Sciences
Actually, this isn't so true of PNAS any more. One of the previous editors decided in the late 1990s to raise the quality prestige of the journal by accepting more papers through a traditional peer-review route, as opposed to NAS members "communicating" or "contributing" articles (which would often have minimal peer review). This was very successful, and now most articles in PNAS get in through the front door, and they're slowly eliminating the back doors. The overall quality is pretty good - not as high-impact as Science or Nature or some of the top specialty journals, but it's definitely a journal that researchers are excited about publishing in if they can't get into the top tier. The fact that they're not part of Elsevier or one of the other big commercial publishers, and their open-access fee is very reasonable, is an added bonus. (Disclaimer: I've published there, so I'm not entirely unbiased.)
Now, as with any journal, knowing the right people always helps - sadly, this is true at any level.