
This directory contains all the code for a parser designed to
run rapidly over large portions of relatively unrestricted text.
All code and documents in this directory are Copyright 1990, 1991
by Carl de Marcken.

Neither the parser nor any of its affiliated files and data may be
copied or in any way distributed without the express written
permission of Carl de Marcken.  The system may be used for strictly
experimental purposes so long as no results derived from its use are
published or used for any commercial purposes.  For permission to copy
the system or its component files and data, or to use the system for
purposes which violate the above conditions, contact Carl de Marcken.

The latex file "doc.tex" is a short users manual to the parsing
system.  The directory "dictionary" contains the parser dictionary,
"source" contains source lisp code, and machine directories contain
versions of the system compiled (under Lucid except Symbolics).  These
versions may not reflect updates in the lisp code in the "source"
directory.

The file "archives/system.tar" is a tar file containing all the lisp code
in the directory "source", the dictionary in compressed form, and the
documentation.  It is a convenient form to ftp the system in
(in binary mode).  Remember, however, that nothing may be copied
without permission.  The archives directory also contains shell
files for creating tar files of compiled code.

Please do not modify the code in this directory.

Carl de Marcken
NE43-800b
545 Technology Square
Cambridge, MA 02139
(617) 253-5659
cgdemarc@ai.mit.edu
