85
edits
Changes
Jump to navigation
Jump to search
Line 115:
Line 115: + + + + + + + + + +
Talk:Essay:Quantifying Order (view source)
Revision as of 05:51, November 15, 2009
, 05:51, November 15, 2009no edit summary
::::I don't know anything about the Mercury data, but the most recent data I've seen here says general relativity might work; it's at the border of the range of error, and it's too precise for any of us to extrapolate. General relativity could be spot on. Or, a new theory being needed - if you're trying to construct a [[theory of everything]], a new theory definitely is needed to deal with [[quantum mechanics]]! But, general relativity does predict a lot of things better than Newton. Newton didn't explain everything, nor does general relativity - but I think you can find better stuff to attack it with than an extrapolation of Mercury orbital data. --[[User:EvanW|EvanW]] 00:37, 15 November 2009 (EST)
::::I don't know anything about the Mercury data, but the most recent data I've seen here says general relativity might work; it's at the border of the range of error, and it's too precise for any of us to extrapolate. General relativity could be spot on. Or, a new theory being needed - if you're trying to construct a [[theory of everything]], a new theory definitely is needed to deal with [[quantum mechanics]]! But, general relativity does predict a lot of things better than Newton. Newton didn't explain everything, nor does general relativity - but I think you can find better stuff to attack it with than an extrapolation of Mercury orbital data. --[[User:EvanW|EvanW]] 00:37, 15 November 2009 (EST)
:::::Ugh, I really should be sleeping, but this is just such an engaging conversation I can't seem to break away.
:::::If you make a measurement to test a theory and the measurement doesn't match the prediction, there are three possibilities. Either the measurement is wrong ("That's not Mercury, that's Saturn!") or the theory is wrong ("Turns out gravity ''isn't'' really caused by leprechauns!") or both are fundamentally right but you failed to take something into account ("We really shouldn't have assumed the sun is a sphere.")
:::::The thing about the Mercury anomaly is that the observed and predicted numbers are ''insanely'' close together. So close as to be declared equal within the margin of error. So either general relativity is ''absolutely'' right and we accounted for ''everything'' — no physicist believes this, by the way — or general relativity is at least ''incredibly close'' to being right, so close that the factors we failed to account for are negligible at the scale of the Mercury anomaly.
:::::But the thing is, Mercury is really thin sauce, as gravity goes. The fields are weak, and the prediction we're testing doesn't say anything terribly interesting about the theory. If you want to get a real feel for how general relativity holds up as a ''physical'' theory, and not just a mathematical one, you really need to look at the Gravity Probe B data. (Conflict of interest alert: I worked on that project.) That experiment didn't measure gravitation indirectly by observing the motion of a particle; it measured it ''directly'' by parallel transporting a vector in a closed loop around a region of curvature. We ''empirically'' measured the curvature of spacetime around a gravitating object (the Earth) and found it to be non-zero. Spacetime is ''not'' flat, massive bodies ''do'' curve spacetime, and the fundamental ''idea'' of general relativity reflects nature.
:::::General relativity doesn't just say that a falling body moves like such-n-such. It tells us ''why.'' And to see it dismissed out of hand because Andy doesn't care for the size of the error bars on the results of a radar study? That, I confess, rubs me the wrong way. --[[User:KSorenson|KSorenson]] 00:51, 15 November 2009 (EST)