Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Posting-Version: version B 2.10.1 6/24/83; site rabbit.UUCP
Path: utzoo!linus!decvax!harpo!eagle!allegra!alice!rabbit!ark
From: ark@rabbit.UUCP
Newsgroups: net.math
Subject: Re: Largest of ten numbers (answer)
Message-ID: <1886@rabbit.UUCP>
Date: Wed, 31-Aug-83 09:45:28 EDT
Article-I.D.: rabbit.1886
Posted: Wed Aug 31 09:45:28 1983
Date-Received: Thu, 1-Sep-83 04:34:55 EDT
References: <20@princeton.UUCP>
Organization: Bell Labs, Murray Hill
Lines: 5

This problem appeared many years ago in Gardner's column
in Scientific American.  I thought that the solution he
gave was to let the first sqrt(N) numbers go by and then
pick the first that was bigger than any of them.
This was claimed to be optimum.