Re: Rainbows in 3D?

[P3D john bercovitz <bercov@xxxxxxxxxxx>]

There aren't that many great questions but this is one makes that 
category IMHO.  It's a fun one.  Seems to me that if the center of 
the circle always lies on the line from the sun to the observer, 
and the distance of the sun from the observer is large, those 
lines must be essentially parallel and so the lines from the 
observers in the case of hypers to the arcs must be parallel and 
so the rainbow's image must exist at infinity.  Since the rainbow 
occludes objects behind it, there is a more powerful clue than 
stereopsis to tell us that the rainbow is nearer than infinity.  
Since common rainbows are seen to be at a great distance, it is 
easy to believe they are nearer than the objects they occlude; 
there is no observable parallax to contradict the assumption.  Now 
for an artificial rainbow created with a garden hose, that rainbow 
should also appear at infinity and so should be confusing as to 
its range due to conflicting clues.  I'm not sure what to do with 
the ends of the rainbows in either of these cases.  I don't know 
where they appear to lie.  (If I did, I'd get the gold, right?)  
This is all speculation of course.
 
John B


------------------------------