I am undertaking a less ambitious but wider-scope project:
to serve all of the roughly 45,000 e-texts available
at Project Gutenberg in text/plain format, plus a simple
search engine that lets you search PG's metadata.
For example, if you ask the search engine for "Dickens friend",
you will get a link to Charles Dickens's novel _Our Mutual Friend_;
if you look for "Ancient Greek", you get about ten books by Plato in
translation, and so on.

I've done some preliminary work on this, but not written any server
code yet.  I expect to use Python, and to host the service somewhere,
maybe on a tiny AWS server.  Of course the server will be open source.

If anyone is interested in this effort, let me know, and I'll be glad
to discuss the issues.  Reasonable extensions would be to search other
Gutenberg-like sites such as PG Canada, PG Australia, Project Runeberg,
etc.



John Cowan          http://vrici.lojban.org/~cowan        cowan@ccil.org
Being understandable rather than obscurantist poses certain
risks, in that one's opinions are clear and therefore falsifiable
in the light of new data, but it has the advantage of encouraging
feedback from others.  --James A. Matisoff