home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!think.com!news!columbus
- From: columbus@strident.think.com (Michael Weiss)
- Newsgroups: sci.physics
- Subject: Re: Gravity & Rubber Sheet Analogy Problem
- Date: 12 Jan 93 15:38:39
- Organization: Thinking Machines Corporation, Cambridge MA, USA
- Lines: 48
- Distribution: usa
- Message-ID: <COLUMBUS.93Jan12153839@strident.think.com>
- References: <79814@hydra.gatech.EDU>
- <Jan.11.21.40.04.1993.10394@ruhets.rutgers.edu>
- <mcirvin.726858610@husc.harvard.edu>
- NNTP-Posting-Host: strident.think.com
- In-reply-to: mcirvin@husc8.harvard.edu's message of 12 Jan 93 17:10:10 GMT
-
- Someone writes:
-
- >>In almost all explanations of gravity as warped space, the analogy
- >>used is two-dimensional. A mass deforms a "rubber sheet", causing
- >>a depression in the two dimensional space . ...
-
- Matt McIrvin replies:
-
- Which really leads to the thing that frustrates me about the rubber-
- sheet analogy: it completely ignores the most important contribution
- to gravitational attraction, which is the curvature you see when moving
- in the *time* direction.
-
- Which brings up the basic problem with using Minkowski diagrams for
- visualizing relativity--- the geometry of spacetime is fundamentally
- different from Euclidean geometry, in that we are comparing an indefinite
- with a definite metric. That means that any naive appeals to ordinary
- geometric intuition are bound to mislead. The "locally shortest path"
- definition of geodesic is almost ingrained in our muscles--- who doesn't
- think of stretching a string taut in the surface?
-
- There is no perfect solution, but an alternative diagrammatic model has
- been developed by Epstein, in his book "Relativity Visualized". First we
- drop down to one spatial dimension, as usual. Then Epstein rewrites the
- formula ds^2 = dt^2 - dx^2 as ds^2 + dx^2 = dt^2, which looks just like
- Pythagoras' formula. (Actually Epstein doesn't tell you in so many words
- that this is what he's doing--- but it is.) Now we have the problem that
- the exact differential dt^2 is "on the wrong side of the formula", compared
- to dx^2 + dy^2 = ds^2, so we can't set up a one-one correspondence between
- spacetime points and points of the Epstein diagram. Still and all, Epstein
- gets a lot of milage out of his diagram. Once you get used to the unusual
- approach, it's very enlightening-- for certain kinds of problems.
-
- His treatment of general relativity is in one way the high point of the
- book. The Newtonian limit emerges the way one wants (the "curvature of
- time with respect to space" that Matt was talking about), and some other
- phenomena come out naturally (e.g., check out his version of the "bullet
- and photon through the center of the earth").
-
- On the other hand, Epstein's diagrams for the case of curved spacetime
- don't give the right curvature tensor (if I've done my calculations
- correctly), so they have at best qualitative value. (For flat spacetime,
- his diagrams are exact.)
-
- I once posted a fairly extensive discussion of Epstein's book, but I now
- think that such a visual work (chock full of figures) can't really be
- "projected" into the ASCII medium of usenet. I recommend it to
- those who like popular treatments of relativity.
-