Re: Dirac String Trick

["John S. Denker" <jsd@MONMOUTH.COM>]

> > (This works with three strings, or any larger number.)

At 08:41 PM 5/12/01 -0700, Michael Bowen wrote:

> Three strings on each side (left and right, the way you drew it), or
> three strings total?

I meant three strings, total.  But even that's not the whole story:

*) If you want to be particular about it, the physics works for ONE string
(grand total) if you count the twist in the string.  Imagine flattening the
string into a broad ribbon to see what I mean.

Of course this isn't the optimal demo, because the audience has a hard time
seeing the twist.

*) If you put a swivel into each string, so that each string can relieve
its own twist, then the minimum number is two strings (grand total).

*) Three strings (grand total) is much more impressive than two.  When
people see the two-string version, it looks easy, and they think that a
third string would block the move that untangles the string.

*) I recommend four strings.  That is nice and symmetric.  With this setup,
it goes without saying that the choice of axis of rotation doesn't
matter.  (In fact it never matters, for any number of strings, but people
are more skeptical with fewer strings.)

*) Misner, Thorne, Wheeler figure 41.6 shows an eight-string version.

This section of MTW has a long discussion of spinors, their geometric
interpretation, their relationship to vectors, and their behavior when
subjected to rotations and boosts.