Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Posting-Version: version B 2.10.2 9/18/84; site lanl.ARPA
Path: utzoo!linus!philabs!cmcl2!lanl!lhl
From: lhl@lanl.ARPA
Newsgroups: net.puzzle
Subject: Re: PIRATE PROBLEM **SPOILER**
Message-ID: <32562@lanl.ARPA>
Date: Tue, 29-Oct-85 10:50:48 EST
Article-I.D.: lanl.32562
Posted: Tue Oct 29 10:50:48 1985
Date-Received: Fri, 1-Nov-85 00:52:52 EST
References: <874@bbncc5.UUCP>
Distribution: net
Organization: The New Mexico Institute for the Incurably Different
Lines: 17

In article <874@bbncc5.UUCP> cosell@bbn-labs-b writes:
>The problem is pretty simple if you do it analytically instead of
>geometrically.
>
>***NOTE: one thing underspecified in the problem but that is *crucial* is that
>the pirates must turn in *OPPOSITE* directions.  It makes no difference which
>party turns clockwise or counter-clockwise, just that they one goes one way and
>the other goes the other.

Yes, it is crucial that they turn in opposite directions, but it does make a
difference which turns clockwise.

Let the landmarks be at (0,h) and (0,-h).  If the pirate walking to (0,h)
turns clockwise, the treasure will be buried at (h,0); if counterclockwise,
at (-h,0).

An easy way to see this is to have them start at either one of these points.