The problem you pose is notoriously difficult, I think.
 
I was actually thinking about another problem, specifically, to show
that for any n >= 0, there exists a circle whose interior includes
precisely n integer points.
 
An outline of my proof:
 
Let O be a point in the plane which is not equidistant from two
integer points.  Then any circle centered at O passes through at
most one integer point.  Take the sequence {C_n | n = 0, 1, 2...}
of circles centered at O and passing through an integer point,
ordered by increasing radius.   We can show that C_n has exactly
n points in its interior.
 
Now we need only show that O exists.  Certainly any O = (x, y)
with mutually transcendental coordinates works, but I was hoping
to find point O with explicit coordinates.  This got me wondering
exactly which points O were candidates, specifically, exactly which
points of the plane are not equidistant from two integer points.
 
 
----- Original Message -----
From: asimovd@aol.com
To: math-fun
Sent: Thursday, January 22, 2004 1:54 AM
Subject: Re: [math-fun] Simple Question I think

David Wilson asks:

<<
Is every rational point on the plane equidistant from two integer points?
>>

Here's a question in this vein that I think is unsolved:

  In the plane, given a unit square, is there any
  point whose distance to each corner is rational?

Dan

--
NOTE: Please direct any responses to <asimov@msri.org>, since I may not keep this AOL e-address much longer -- thanks.


_______________________________________________
math-fun mailing list
math-fun@mailman.xmission.com
http://mailman.xmission.com/cgi-bin/mailman/listinfo/math-fun