These live on a continuum. I don't think that there is a shift from one to
another. SGML allows you to decide what is formatting and what is
abstraction. It is decided on an element by element basis and most DTDs
will have different parts at different points along the continuum. For
instance bibliographies typically tend towards data modelling, predictable
body text towards what the middle, and unpredictable body text (e.g. web
pages) towards the formatting side.
Paul Prescod - http://itrc.uwaterloo.ca/~papresco
"Perpetually obsolescing and thus losing all data and programs every 10
years (the current pattern) is no way to run an information economy or
a civilization." - Stewart Brand, founder of the Whole Earth Catalog
http://www.wired.com/news/news/culture/story/10124.html