                                         28 Aprch 2010


The software has been built on a few Linux platforms.

To build the software follow the steps:

 > ./configure
 > make


numbers supports three flavors of floating-point arithmetic:

  1) multiple precision using the GNU gmp library (available
from http://gmplib.org/), i.e., significantly more precision and
range than long double,

  2) The C type long double which is supported by most
compilers and may provide more precision and range than the type double,

  3) The C type double precision

The gmp library supports the widest range of floating-point values and
precision and will be selected by configure if available.

Support for automatically reading the contents of compressed and
archived files is provided by libarchive (available from
http://libarchive.googlecode.com/).
  
The ndb directory contains the 'interesting' numbers database and
currently needs to reside in the same directory as the numbers
executable program.


                        REPORTING BUGS
    
If you find a bug in the program or the numbers database please
report it to:  numbers@knosof.co.uk


