Slashdot Mirror
Possible Explanation of Unpredictable Sun
Weedstock writes "According to this article on New Scientist, it would be caused by a magnetic field that affect the gas flow in the star. This magnetic field being created in the first place by the gas flow, it results in a loop which cause the star to act unpredictably."
What a horrible, horrible summary.
Within the sun there is a finite number of particles, which are made up of a finite number of subatomic particles... and the movement of each particle can be measured exactly if we take in account of all the number of forces acting upon it (which is... you guessed it... finite). Essentially, you would have to make a great deal of computations (...again, finite), but it is possible. The only problems you would get into would be continual motion... if there is such a thing...
Not unless our understanding of atomic and sub-atomic particles is completely wrong. You can NOT measure "the movement of each particle ... exactly", as explained by the Heisenberg Uncertainty Principle. There is a limit to the precision to which you can measure the particle's 'movement'. If you somehow measured the velocity of the particle exactly, you would have a very uncertain idea of its exact location, which would also be a major factor in determining that particle's effects on its surrounding particles. Which is of course necessary if your goal is to be able to calculate the movements of every particle in the Sun.
It would not be possible, and not necessary or useful either, to build a complete model that calculates the exact movement of every part of the Sun. We will just keep creating slightly more accurate models over time, until we have one that is "good enough".
Sometimes the best solution to morale problems is just to fire all the unhappy people.
Please pick up an introductory book on quantum mechanics. The typical second-year book should be adequate provided it's calculus-based.
Newtonian mechanics has the concepts of "position" and "momentum," but in QM there is only the wave equation. That wave equation can be solved for position or momentum, or both with a bit of "smearing," but the equations themselves incorporate a minimum amount of uncertainty when you try to solve for both simultaneously.
No problem, the equations just describe behavior, not reality, right? Only problem is the "hidden variables" theory was disproved by some clever experiments back in the 30's (iirc).
It's possible that QM will someday be superceded by a new theory... but that new theory has to explain close to a century of experiments that have repeatedly demonstrated that the Uncertainty Principle, or something functionally equivalent to it, is real. In other words, any direct replacement for QM will still show Uncertainty.
Is all hope lost? Not quite - all of these experiments were conducted with tools that themselves incorporated a number of implicit assumptions. It's conceivable that some radically different technology could yield absolute knowledge. But the problem here is that we don't have this technology, we don't know where to start looking for it, and even if we found some instruments in a crashed UFO the concepts may be totally foreign. Perhaps these instruments can precisely measure the snerk and squutl... but there will again be inevitable uncertainty when we convert snerk and squutl to position and momentum.
For every complex problem there is an answer that is clear, simple, and wrong. -- H L Mencken
Actually, homeostatic systems, such as the feedback loop described above, are generally in a condition known as "sensitive dependance on initial conditions"... basically, that it is very difficult to measure the entire system precisely enough to predict long-term behaviour in any sort of way.
In addition, it's so difficult to measure the system, that even if one managed to do it, the energy and precision required to do so would ultimately result in a change of the system as you measure it; meaning that the model that you create would no longer be accurate when you were done.
Basically, in any positive-feedback system, precision on the atomic scale is necessary to predict long-term behaviour. Short-term behaviour can be predicted in a gross way, but the length of the prediction is volatile; the speed of the positive-feedback system (in this case, how long it takes a small magnetic domain to become a large magnetic domain) ultimately determines for how long your model will remain reasonably accurate. And measuring things on the atomic scale ALWAYS results in changes at the same scale; it's like taking a gymnasium full of tennis balls and bouncing basket balls around, trying to determine where the tennis balls are from the angles that the basket balls bounce at; you can get useful information that way, but the tennis balls will be bounced around in unpredictable ways while you do it.
I am disrespectful to dirt! Can you see that I am serious?!