() Jason Massey <janixs...@gmail.com> () Thu, 13 Sep 2012 09:48:58 -0400
[the horror the horror] Yeah, XML is suboptimal. SXML is slightly better, but still rotten. *So, I guess what I am saying is that I hope that Texinfo does not meander into XML-verbosity.* Actually, I'm begging that it does not. I agree completely; the suggestion was made w/ a strict anti-meandering mindset. The idea is to allow "pass-through specification", reserving the fewest special-handling specifications (w/ semantic perturbations) possible. Essentially, makeinfo says, "You want to express your style? Fine, no skin off my nose. I'll let the downstream renderer know. If your specs confuse the downstream renderer, don't complain to me! Don't GI if you don't want GO!" (See <http://foldoc.org/GIGO>.) Re verbosity, you can hardly get simpler than a Lisp plist. But that's besides the point (which is explained above). The document model already includes nodes w/ certain attributes (name, depth, relation to other nodes in the graph). The suggestion was to add pass-through attribute specification to environments, but the same idea could also be applied to the nodes (w/ strict anti-meandering mindset) themselves. The sooner makeinfo supports uniform pass-through specification, the sooner it can remove itself from the style-wishlist treadmill. -- Thien-Thi Nguyen ..................................... GPG key: 4C807502
pgpR3zliCYtQK.pgp
Description: PGP signature
