Re: [Phys-l] The behavior of light

[John Denker <jsd@av8n.com>]

On 05/03/2008 05:15 PM, Fink Trevor M wrote:
> After all of this discussion I think that I finally understand the
> idea.  In my thought experiment I had a single photon traveling
> between the mirrors of my light clock.  It would seem that instead it
> is an omnidirectional light source shining through a pinhole.  As the
> light is diffracted it travels along both the x and y axis.  If this
> is correct please let me know so I can put this to rest.

Diffraction is not the key idea.  Ray optics suffices.  The
key idea is that the mechanics of the light source -- the 
physical structure of the light source -- aims the blue light
so that it is vertical in the blue reference frame.  This is 
super-obvious to observers in the blue reference frame.  The 
observers in the green reference frame:
 a) see the light as non-vertical in their frame, and
 b) need to work harder to see how this comes about ...

... but it is the *same* mechanics, the same physical structure.

Also keep in mind that the relative motion between the blue car
and the green observers is assumed to be /uniform/ motion.  It
is a completely different question (with completely different
answers) if you imagine that the blue car is stationary when
it emits the light, and then starts moving.

> The only other question I have is: Is this light clock actually
> measuring time?  When we set it up we had 2 constants and 1 variable.
> We couldn't the change speed of light and we knew what we wanted to
> call one second.  Because it is easier for me to think about, I am
> going to say that every time the light makes one trip up and back
> down it is one second. 

This is leaving the domain of physics and approaching the realm
of opinion and the realm of word-games.  But sometimes words are
important.

The modern (post-1908) view is that yes, a clock measures time.
Proper time (not any sort of dilated time) is what we should
consider "the" time for that clock.

> .... I guess what I am getting
> at is, if a clock reads that 10 seconds ha ve passed, but thousands
> of other clocks say 11 seconds have passed wouldn't it make more
> sense to say that the first clock is broken rather than the first
> clock traveled to a place where time changes? 

More sense than what?

If two cars start out at point A and end up at point B, and rack
up different mileage on their odometers, do you start by assuming
one of the odometers is broken?  I don't.  The simplest hypothesis
is that one of them took a longer route.

So it is with clocks.  If two clocks start out the same and end up
different, do you assume one of them is broken?  I don't  The simplest 
hypothesis is that one of them took a more time-consuming path.

This is discussed in much greater detail, with diagrams and all, at
  http://www.av8n.com/physics/odometer.pdf
See also
  http://www.av8n.com/physics/twins.htm

================================

Yes, I am quite aware that it is possible to "explain" special relativity
in terms of rulers that don't tell distance and clocks that don't tell
time.  But for 100 years now, since 1908, there's been a better way:
  Proper time.  Proper length.  4-vectors.  Spacetime diagrams.