The earliest hope in the development of RST was that it would turn out to be useful in Text Generation, getting computers to produce text. It has now been used in a variety of text generation projects, and the literature is extensive. Text generation has been subject to the ebb and flow of sponsor interest, but in some ways the work in Europe has complemented that in the US, and so development on this subject has continued.

There is a Special Interest Group, part of the Association for Computational Linguistics (ACL), called SIGGEN, which stands for Special Interest Group on Generation. They have a number of online resources.

For a general overview of Text Generation, go to the SIGGEN page, and read the introduction (under "About").

One development project that is extensively represented on the web is the ILEX system at the University of Edinburgh. It includes these options for the user:
• a demonstration of computer generated text.
• an extensive description of ILEX 2, the one that the demonstration is based on.
• a description of ILEX 3, covering both underlying theory and system structure.
• a description of the designers' views of text structure.
• a description of how RST was represented in programs.
• the project's Final Report

To see it all, click here: The ILEX Text Generation System .
Visitors to ILEX should note that the computer program code for ILEX is not available, but the program code for the WAG sentence generator found inside ILEX is available, from Mick O'Donnel.