Books on Computational Semantics by Patrick Blackburn & Johan Bos
Representation and Inference for Natural Language
A First Course in Computational Semantics
Software Requirements and Downloads

Platforms
The software was primarily designed for linux and solaris environments. With a little bit of luck and tweaking, you can get it to work for MS Windows platforms as well (it seems to work well under DOS and CYGWIN using Sicstus).

Prolog
Almost all programs accompanying the lecture notes are written in Prolog. We have tested our programs using two dialects of Prolog:

SWI-Prolog is a Free Software Prolog compiler, whereas running Sicstus Prolog requires a license. Both can be downloaded following the links above. In case you need to fresh up your rusty Prolog programming skills, try Learn Prolog Now!.

The Course Software
The latest version 1.3 (November 2006) of the programs for Representation and Inference are packed in this gzipped tarfile.
To unpack do: gzip -d bb1.tar.gz; tar xvf bb1.tar
Previous versions: BB1 Version 1.2 (August 2005); BB1 Version 1.1; BB1 Version 1.0.

Perl
The Curt family (Chapter 6) uses the scripting language Perl to hook up various theorem provers and model builders with the Prolog code. Therefore, you need to have Perl installed in your working environment.

Inference Engines
Chapter 5 and 6 make use of a series of theorem provers and model builders. These need to be downloaded and installed separately, and can be obtained via the following links:

You need at least one theorem prover and one model builder. Otter and MACE come together in one distribution package, and are also available for MS Windows. Bliksem and Paradox are (for some problems) more powerful alternatives.

If it still doesn't work try the following. Ensure that directories of the executables (of MACE, Otter, Bliksem, and Paradox) are included in the search path. Some provers (such as Bliksem) require textfiles in unix format when ran under MS Windows. Add a call to "dos2unix" in callInference.pl to convert the format of the input files for the inference engines.