RE: An open plea to the W3C (was Re: XInclude vs SAX vs

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

Yet by the mid 90s, optimizers for ADA had been shown 
to create code that ran as fast and as reliably as C. 
Just as with SGML and now XML++, sharp people carved 
off the pieces they needed for their projects and 
made them work locally.  What is different in our situation 
is the size of the locale in which these things must 
work.

Complexity is manageable given willing resources.  My 
sense of the web is that anything very complex will 
soon encounter and very unwilling set of resources, 
so the first thing to accept is that nothing universal 
is achievable for very long.

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

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


-----Original Message-----
From: Steven R. Newcomb [mailto:srn@coolheads.com]


About ten years later, another Turing Award Lecturer (don't have time
to track down the reference) complained about the complexity of Ada.
According to what I recall from reading about it in _Computerworld_ 20
years ago, the complaint was different from Dijkstra's complaint about
PL/I (below).  The complaint was not so much that programmers couldn't
handle Ada's complexity, but that reliably implementing Ada itself was
simply out of reach.  The speaker worried that missile guidance
systems and other weapon systems running real-time Ada programs would
misbehave in various unforeseeable ways.  He was horrified that the
U.S. military was trying to standardize on Ada.