Proposed TAG issue on the boundaries of the Web

Discussion here has swirled around under a variety of
subject lines but centering on the vexed question of what
is part of the Web (& hence its architecture) and what isn't.  
For my sins I have been given an action item to try to tease 
out a reasonably crisply stated issue here for adoption into 
the TAG issues list.  Addressing this has required actually 
reading each and every posting and thinking about at least 
some of them, for which I feel the Universe owes me a beer 
or ten.

Related questions include whether the REST framework should 
play a part in defining what is architecturally important, 
whether the Web architecture must, does, or should include
any or all of email, SOAP, DAV, and so on.

Here's my best shot at crystallization:  

 Is the Web Architecture characterized by an enumerated
 list of technologies or by a rule-based decision procedure?
 If the former, what's now on the list and how do things get
 on the list?  If the latter, what are the rules, how
 do rules get on the list, and are exceptions allowed?

At the moment, the TAG seems fairly comfortable with the
very short draft intro to the architecture doc, which leans 
to the architecture-by-enumeration approach.

Cheers, Tim Bray

Received on Monday, 1 April 2002 13:00:58 UTC