One comment here: language *is* different from other purely cosmetic tags
such as font, emphasis, etc. The former affects the meaning and
interpretation of the text, whereas the latter do not. For instance, a
good full text indexer has a list of empty words, which are excluded from
the indexing to decrease bulk. This list, of course, depends on the
language, but not on any cosmetic quality of the text. Any attempt at
machine translation, however crude and limited, will also need to know the
source language.
Not that this means that language should be treated at a different level,
though, but it should not be dismissed as just another presentation hint.
>Doing it at a low level adds complexity and complicates clients
>and servers that use Unicode. There has to be a compelling reason to add
>this complexity. There has to be a problem that it solves.
Perhaps this should be "...a problem that it, and only it, solves".
-- Francois Yergeau <yergeau@alis.ca> Alis Technologies Inc. +1 514 738-9171