> > I don't think it's good to downplay camlp4.
> > we need more documentation and more tests.

> (The new) Camlp4 has been here for several years.
> Documentation and tests are still lagging behind.

Is it possible for Inria or someone else to provide
funding for that?  I would like to help out starting
sometime in the summer.

> the most common uses of camlp4 might be based on a much
> simpler approach ... Coq is one of the few examples of
> a big project that relies on other aspects of camlp4

Why not focus on optimizing new lightweight tools for
small problems and keep camlp4 for big ones?  It wouldn't
be so bad if it were documented better.