On Wed, 09 May 2012 08:54:30 +0200 (CEST) Werner LEMBERG <w...@gnu.org> wrote:
> > The problem is fundamental - a good conversion needs structural > > analysis as good as doclifter's, which is *hard*. > > Yes. However, grohtml would be able to do different things, in a > different way, and with proper markup the results could be excellent > also. Sigh. It's interesting you say that, Werner, because I think grohtml is broken by design, for the simple reason that HTML in no way resembles a printer. Because troff affords the user finer control -- dot-addressable control -- over the (intended) output, any conversion to HTML is subject to gross loss of fidelity. Aren't the advantages of ditroff completely lost? If the browser offered something more printer-like, then we'd have something more useful to troff. AFAIK that doesn't exist. (Convert xdvi into a Firefox plugin. Just a small matter of programming.) --jkl
