An interesting problem: distance estimating

["markw-4" <user=5@nabble.blu.org>]

 I've been thinking about an issue: 

How do you estimate the distance between two points? 

We all know that geographical "as the crow flies" calculations are wrong 
with regards to distance of travel. 

We also can not estimate a truly *real* number without knowing the actual 
rout and roads taken. 

So, what do you think would be a practical algorithm for estimating distance? 

I have been sort of thinking along the lines of a decreasing multiplier 
based on actual distance, i.e. the closer two points are, the less likely 
a direct line exists between the two points. So as two points get further 
and further away, the travel line between the two points looks more and 
more direct. 

Think of drawing a line on a map how you'd travel between quincy and 
waltham. Its almost twice the distance. 

Think of the line between your house and the house on the next street 
over, its could be more than for times the distance from the center points 
of each house. 

Now, picture a map of the US and draw a line across the interstates 
between boston and san fransisco, its almost a straight line. 

What do you think? 

-- 
This message has been scanned for viruses and 
dangerous content by MailScanner, and is 
believed to be clean. 

_______________________________________________ 
Discuss mailing list 
[hidden email] 
http://lists.blu.org/mailman/listinfo/discuss