AWWSW homework for 2007-12-11

Our starting point remains this document: http://esw.w3.org/topic/ 
AwwswTopicsBrainstormPage . It appears it didn't get much attention  
prior to the last meeting, so I hope everyone gets a chance to review  
it this time.

The last meeting began at the top of the file by considering the  
question of what might one infer from a 200 response. Of course we're  
not at a point where we can even ask this meaningfully; we  
immediately got onto the question of whether by "permitting" any  
inferences at all we're interpreting or extending HTTP 1.1, or doing  
something else. I've expanded on the result of this discussion a bit  
in the wiki page.

I remember that on the call Pat said something of the form "but the  
real problem to be solved here is ...". Unfortunately this didn't  
find its way into the meeting record and I don't remember the rest of  
the sentence. Pat, could you give your ideas on where a group like  
this might best put its efforts? For background, the assumption is  
that formalizing HTTP (or rather some "best practices" extension/ 
restriction/fragment of it) would benefit semantic web agents such as  
Tabulator, applications that want to be extra careful about  
provenance (where did something get said - in a resource? in a  
particular representation? in a response? in an "essence"?), and many  
other kinds of applications. I was also personally of the opinion  
that formalization could help force answers to many of the thorny  
questions that keep arising as a result of vagueness and ambiguity in  
AWWW and other informal specifications, and that such clarification  
would make everyone happier; but I don't know whether anyone agrees  
with me on that. So we are not starting with a crisp problem  
statement here, and maybe that's a bad thing.

Jonathan

Received on Thursday, 6 December 2007 15:27:31 UTC