The documentation files in the GNU source distributions were converted from TeXinfo to HTML by

texi2html -split_chapter -glossary -menu absolute-file-name

You can find the texi2html program in the Comprehensive TeX Archive Network archives ( finger ctan@pip.shsu.edu to find a site near you), and at the home of its author.

texi2html is written in the perl language, so you must have a recent version of that already installed; if your site has a Web server, perl is probably already available, since it is heavily used by Webware.

Absolute path names are used in the texi2html command to ensure that each HTML file that is created carries a trace of its origin in the local file system, and importantly, the software version number which is included in the directory name.

Each document is split into multiple small files (by the texi2html -split_chapter option), to reduce net traffic, and to improve browser responsiveness. The average HTML file size is about 22KB, although some are as large as 300KB.

To help automate the conversion, a small UNIX shell script was written to run texi2html on either specified files, or else all the TeXinfo files in the current directory, and then to move the output HTML files to the HTTP tree.

In each GNUware distribution, it was necessary to manually examine the TeXinfo files to determine which were top-level files needing conversion, and which were included files that are automatically handled by texi2html.

Although the absolute directory location of the output HTML files will vary from site to site, that should be the only change you need to make in the script to adapt it to your system; feel free to do so!