On 15 jan, 04:42, glen herrmannsfeldt <
g...@ugcs.caltech.edu> wrote:
> > The conservation of momentum is the consequence of the homogeneity of
> > space. It requires more than simple topology.
> > What I mean is, for example, a photon come from another galaxy. How
> > does it know that it is near an electron on earth, so as to be
> > absorbed by it? Of course, we can say there is an em field at a
> > point, but how does it know that the electron is in the vicinity of
> > that point?
>
> What does 'in the vicinity' mean? Close enough, relative to the wavelength.
It means that, in a near point in time, the wave will propagate to the
electron. (a trajectory or path is a continuous map.)
It is not about quantum mechanical babble, it is about basic space
time. In this context, quantum mechanics is merely wave mechanics,
and classical mechanics also applies.
> My always favorite example is the neutron cross-section, which for
> some nuclides is much larger than the nuclear diameter for slow
> (long wavelength) neutrons. If the wave function has enough amplitude
> at the nucleus, it can get absorbed.
That's much more than topology. For those who aren't familiar with
it, here a short reminding:
If X is a set, a topology of X is a collection of its subsets called
open sets, and that have these properties:
- X and the empty set are open sets.
- The union of any number, including infinite, of open sets is an open
set.
- The intersection of a finite number of open sets is an open set.
A neighborhood of an element x of X is a subset of X that contains at
least one open set to which x belongs.
An example is R with the collection of all its open intervals and
their unions. It is called the usual topology.
A map X -> Y is continuous iff the inverse image of any open set in Y
is an open set in X.
If the open set in Y get smaller and smaller, the open set in X do
too. Thus a continuous map, as it were, conserves the vicinity, which
itself is rather loosely defined.
A homeomorphism X -> Y is a continuous map that has an inverse which
is continuous. It is sort of an isomorphism of topologies.
> Otherwise, if you use path integrals, the photon wave function has to
> explore all the possible places where an electron might be available
> to absorb it. It then has to choose, with the appropriate amplitude,
> which of the possible electrons will absorb that photon.
We don't need paths integral, which is a much elaborated, and
correlatively useless concept. To think about topology, the Huygens
principle is way enough, and the paths integral can be derived from
it.
--
X-Phy