In the mid-1990s, a Statistics professor at UNC reported an IT problem: he couldn't send email more than 500 miles away. He'd had geostatisticians plot the sites where emails could be sent successfully and where they bounced, and found they described a circle a radius slightly over 500 miles. Some sites within the circle could sporadically receive email, but definitely none outside them. The solution to the problem, which involves old sendmail.cf formats and the speed of light, is well worth a read.
Comments