There are essentially three kinds of users for KPML, each with a different set of use requirements. Those users are:
The first two groups can often make use of the "standalone images" described on the KPML home page. These images require no extra licenses and no extra software. They can be downloaded and used as they come and allow grammar development and searching the linguistic resources available for KPML. To use these images it is recommended that a PC running some variant of Windows (95 up) with at least 128Mb of RAM and a 200+ Mhz processor is used. The images may run on small machines, but will be slower. Some display windows may also fail to start if the machine used is not fast enough.
The last group usually requires more control over the system. Here knowledge will be represented in some knowledge representation language describing the domain in which texts are to be generated, and various flexible interfaces will need to be provided between application system and generator. For this kind of usage, it is best to obtain the source code of KPML and to install it properly: i.e., compiling the code and then loading it into the complete application. Some variations on this are imaginable, but that is the basic setup.
For compiling and installing KPML from the source code the following additional software is necessary:
Instructions for installation are part of the code distribution and the online documentation.