$Id: README,v 1.2 1998/10/17 15:20:48 fraserm Exp $

Multiple-word anagram search program (version 1.4).

Unpack all files into an empty directory, edit Makefile to reflect
your system. Unlike previous releases, a word file "words" is
supplied. Please read the notes on how it was created at the bottom of
this file.

If you are on SunOS and Solaris or Digital Unix, a suitable file is
/usr/dict/words, and you may choose not to install the supplied word
file; in this case, edit the Makefile and change WORDDIR. Do *not* do
a "make install-words" :-).

For other systems, the word file is required. See the manual page for
how to use this program to build unique word lists from ordinary text
files.

Output from the first example in the manual page using the supplied
wordfile is in the file "sample_output"; this should allow you to
check the "thoroughness" of any wordlist you generate yourself.

Run your choice of:

	make		make executable 'agm' in current directory
	make install	make 'agm' and install it and the Tcl/Tk front-end
			'xagm' in BINDIR and the manual page in MANDIR
	make install-words
			install the supplied wordfile in WORDDIR
	make doc	if you have "groff", make a PostScript version of
			the manual page
	make clean	remove all but distribution files

If you find you have to change anything to get it to compile (even if
it's just the Makefile, please let me know, to the bug address.

For quick reference usage and options, type 'agm'.

Bug reports to the address below. It would be helpful if you included
all switches and target words you gave, details of which wordlist you
are using (but please don't email it to me unless I ask :) the output
of the program, and the output of the command 'agm -v', together with
details of what hardware and OS release you are using.

		Fraser McCrossan

			fraserm@gtn.net

Word File Notes

The supplied word file is of English words only :-).

The word file was compiled from some of Project Gutenberg's Etexts:

Websters Unabridged Dictionary, 1913 (pgwht04.txt)
The Jargon File version 4.0.0 (jarg400.txt)
A Midsummer Night's Dream, William Shakespeare (1ws1710.txt)
The $30,000 Bequest and Other Stories, Mark Twain (beqst10.txt)
The Adventures of Huckleberry Finn, Mark Twain (hfinn10.txt)
Roget's Thesaurus (all 3 versions) (roget13.txt, roget13a.txt, roget14a.txt)

As well as the Linux words file version 2 from Rik Faith (on Sunsite in
libs/linux.words.2.tar.gz). README* files from this distribution are included
for your reference.

This gives a reasonably comprehensive list of words.

According to the heading information on the Websters dictionary, the
<...> field marks in this dictionary are "Copyright (C) 1996 by MICRA,
Inc. Plainfield, N.J.", whilst the actual dictionary entries are in
the public domain. In creating the word file, all the field marks were
removed, leaving no copyrighted material.

Furthermore, although I state that the word file was compiled from these
etexts, there the resemblance ends. The word file bears no resemblance
to these etexts, and I mention "Project Gutenberg" for information
purposes only. I have deliberately manipulated the list for efficiency
reasons and do not guarantee that all words in any of the etexts
are also in the word file. In other words... if you want either of
these texts, go get them from Project Gutenberg!