Am 03.10.2010 um 10:14 schrieb Philip Taylor (Webmaster, Ret'd): > > > Philipp Stephani wrote: > >> In TeX you cannot state the structure because TeX is a low-level > > typesetting system that offers only a few low-level primitives > > and a macro language. > > If "TeX offered only a few low-level primitives", I would have > been willing to accept that your argument might have merit, but > you yourself then go on to say (quite correctly) "and a macro > language". And it is precisely /because/ of TeX's macro language > that one can express the exact structure of any document, using > either a mutually interchangeable set of conventions such as LaTeX > or Context, or using a private macro package of one's own. Many > choose one of the first two options, I prefer the third, but all > of us are able to express document structure in TeX.
Yes, but is that really "structure"? Of course it's basically a question of definition, but if you look at other technologies that are supposed to be able to express structure (e.g. XML), then you'll find data modeling, schema, transformation and querying languages, all of which are nonexistent in the TeX world. What I want to say is that macros can give a pretty good *simulation* of structure, but that simulation is leaky. In the middle of a LaTeX document you can say \let\chapter\section and all subsequent sections turn into chapters. This lack of referential transparency makes LaTeX documents pretty complex and hard to process compared to XML languages. -------------------------------------------------- Subscriptions, Archive, and List information, etc.: http://tug.org/mailman/listinfo/xetex
