On Modularity and the History of Markup

["Bullard, Claude L (Len)" <clbullar@ingr.com>]

I'm probably not the only one to notice that 
modularity based on the pattern

driver ->collects modules ->collects components

is basically a C main function struct sort of 
affair.   Is it a common pattern or just that 
we only do variations on what we know?

When SGML was beset by the minimal victory of HTML, 
it was entrenched in modularity discussions, still waiting for 
DSSSL and so on.  It all collapsed to a one 
size fits all language (aka, genCodes) then 
began the relentless climb back up to complexity.

The pattern of events keeps repeating itself in a nicely 
spiral as one would want it to.  I am fascinated 
to see if a new simpler easy any one can learn 
it paradigm will emerge to take advantage of the f
fact that WML/SMIL 1.0/ComicML etc are all variants 
on frame architectures, that most head/body systems 
are scrollers, and so on.  What will be the next 
trendy simplification?

Architecturally, most markup languages are variants 
on common tropes of markup design (a trope is stronger 
than a pattern).  Can we simplify our jobs with tropes?

Architectures?  Where have I heard that before?

Len 
http://www.mp3.com/LenBullard

Ekam sat.h, Vipraah bahudhaa vadanti.
Daamyata. Datta. Dayadhvam.h