Slashdot Mirror
New Calculations May Lead To a Test For String Theory
dexmachina writes "A team of theoreticians, led by a group from Imperial College London, has released calculations that show string theory makes specific, testable predictions about the behaviour of quantum entangled particles. Professor Mike Duff, lead author of the study from the Department of Theoretical Physics at Imperial College London, commented, 'This will not be proof that string theory is the right "theory of everything" that is being sought by cosmologists and particle physicists. However, it will be very important to theoreticians because it will demonstrate whether or not string theory works, even if its application is in an unexpected and unrelated area of physics.' In other words, string theory may finally have shed its critics' most common complaint: unfalsifiability. However, given the second most common complaint, I can't help but wonder: which string theory?" Update: 09/03 23:34 GMT by S : Columbia University's Peter Woit, author of the Not Even Wrong blog, says these claims are overblown, and adds that a number of string theorists said as much to Wired.
So wait, we have two theories that describe different realms and no data for the intersection of the two realms, but people are trying to come up with a theory for the intersection?
That's like saying, I know how to fly a plane, and I know how to drive a car, but neither skill applies to flying cars, which I've also never seen.
But they must exist, right??
Is that it isn't. What I mean by that is it doesn't seem to make any testable predictions. At this point, it is just a bunch of math wanking. Now there's nothing wrong with purse math. A lot of useful theories start out that way and I like the Bacon quote "If in other sciences we should arrive at certainty without doubt and truth without error, it behooves us to place the foundations of knowledge in mathematics."
However when all you've got is a bunch of neat math with no real testable predictions, it is not a theory and it is not the sort of thing to be crowing about to the general public. XKCD, as usual, did a humorous job summing it up: http://xkcd.com/171/.
If you are going to complain that people complain about the lack of testability then you need to do two things:
1) Read The Logic of Scientific Discovery again and brush up on what a theory is and isn't.
2) Don't go making press releases. I'm not saying you personally have done this but physicists are awful happy to talk to the press about something they can't prove.
Part of it is simply wanting accuracy in the use of the words because let's face it: In science accuracy matters. Being pedantic about terms is important in science. Another part of it is this is the kind of thing that confuses normal people. With evolution, scientists have gone to a lot of trouble to explain that a theory is NOT a guess, NOT a wild idea, etc. They show other theories and how they work, how many things we accept as true are theories.
Well something like this undermines that to an extent, because here is something being called a theory that is not only untested, but that they can't even figure out how to test. It is the kind of thing that can make people say "But wait, if this is a theory then theory doesn't mean what you said."
Actually, it seems to me like we don't call those grand-unified things a proper scientific theory either. As long as there are no testable predictions, and it fails Occam's Razor, it's not a theory, plain and simple. It's a hypothesis.
Yes, there is a name for a theory which hasn't yet been tested: hypothesis.
And really, as someone who's gotten tired of hearing Young Earth Creationists go "well, evolution is just a theory" and having to explain to them "yeah, but theory in science doesn't mean what you think. It means it already made testable predictions and is the best we have"... it's getting annoying to see that a whole bunch of physicists are actually using it exactly as the YECs and conspiracy theorists think: as just an untested and untestable supposition, which may or may not actually hold any water at all.
Yes, I realize that calling it a "theory" is more science-y sounding and good for your funding. But it devalues the whole idea of science for everyone. If we accept that some untested and untestable calculation is just as worthy of being called a "theory" with a straight face as GR or electromagnetism just because it's the pet supposition of some physicist, then basically why wouldn't Behe's pencils-up-the-nose ID idiocies be a "theory" too? I mean Behe _is_ a professor of biochemistry.
Call it the String Hypothesis, and you'd see a lot less complaints, basically.
A polar bear is a cartesian bear after a coordinate transform.
Indeed, it's worse. While I don't know much about string theory, I do work in the field of entanglement. And there's no way you could experimentally test this classification, for the simple reason that it's a classification. It may be a more or less useful classification, but you cannot experimentally test whether a classification is right (apart from that an reasonable entanglement classification has to be SLOCC invariant, which this classification is, but of course the others are as well). Trying to experimentally test if a classification is right is like doing an experiment on whether classifying a fruit on its color or on its size is more correct. What you can do is to evaluate the usefulness of a classification (i.e. does it tell you something interesting about the state, like what you can do with it; in the fruit example, you might find that classifying fruits on nutrition value may generally be more useful than classifying on water content).
The Tao of math: The numbers you can count are not the real numbers.
It seems I may have jumped the gun on this one. My bad for being such an easy mark of sensationalist pop science headlines.
Don't feel bad, I submitted it a day before you did. What really blows my mind is that Not Even Wrong used my submission as evidence that Slashdot was running a story on it:
Update: No press campaign for a “finally string theory is testable” claim is complete without a Slashdot story
Big news for theoretical physicists who are fed up with the inability to test String Theory
(that's from my submission)
My work here is dung.
Hi. I have a very stupid question for someone working in the field of entanglement.
Spin is measured as up or down, but presumably the spin is actually at some angle in between. So the up or down measurement is rounding the actually spin. Is the resulting rounding error of any significance? Is the accumulation of rounding errors on multiple measurements of any significance? Or, as is most likely, is this a nonsensical question?
Thanks
I don't think I can blame your for not reading the paper. They just lied and lied and lied in the press release: "Researchers describe how to carry out the first experimental test of string theory"
Come on! I know that press releases are about PR not science but they should at least bear any relation to the paper.
Look in the paper. The word experiment only appears once, and they don't claim to describe an experiment; not even that it is possible or conclusive.
That said, the question is: who is the lier? The scientists or the marketers? If I can take as true the quotes at TFA, Mike Duff.
Well, at least you showed us whose research we should ignore.
entropy happens