Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Posting-Version: version B 2.10.2 9/18/84; site brl-tgr.ARPA
Path: utzoo!linus!philabs!cmcl2!seismo!brl-tgr!tgr!gwyn@BRL.ARPA
From: gwyn@BRL.ARPA (VLD/VMB)
Newsgroups: net.lang.c
Subject: Re:  Random Numbers
Message-ID: <43@brl-tgr.ARPA>
Date: Sat, 20-Jul-85 23:49:12 EDT
Article-I.D.: brl-tgr.43
Posted: Sat Jul 20 23:49:12 1985
Date-Received: Mon, 22-Jul-85 03:58:46 EDT
Sender: news@brl-tgr.ARPA
Lines: 12

You are right -- there is no way to generate "really" random
numbers in a deterministic manner.  Indeed, defining what is
meant by "random number" is not easy.  The classic reference
for these matters is Donald Kunth's "The Art of Computer
Programming, Vol. 2: Seminumerical Algorithms", 2nd. Ed.
(Addison-Wesley, 1981).

Most people in practice use a library function like rand()
and "seed" it (via srand()) with a number unique to the
process invocation, such as the XOR of the time-of-day and
the process ID.  Be warned that most implementations of
rand() do not generate very good random sequences!