Re: Idea on conformance

On Thu, 2003-12-11 at 12:29, Tim Bray wrote:
> I think that there's general agreement that it would be nice if webarch 
> had a conformance section, but general puzzlement about what could 
> usefully go in there.  Here's an idea: suppose we defined a document 
> template called a Web Architecture Conformance Statement.  It would be 
> organized along the lines of our principles, in which people could 
> answer some questions yes/no, e.g., The important objects in this 
> specification/program/whatever are identified by URIs: Y/N.  For those 
> principles that don't lend themselves to a crisp yes or no, e.g. 
> "carefully consider textual vs. binary", the document would include a 
> place in which the person filling it out could put a paragraph 
> outlining the reasons they made the choices they did.  Then, while we 
> need not define a condition called "conformance", we can encourage 
> authors of specifications and software modules to publish a Web 
> Architecture Conformance Statement.  It need not be long or complex. 

My experience with the WAI Guidelines suggests that conformance
to requirements that are as general and broad as those in
Webarch will not be as useful as conformance to domain-specific
requirements derived from Webarch. 

For instance, suppose I am a format designer. I would find it useful to
think about a statement such as "A resource owner SHOULD assign a URI 
to each resource that others will expect to refer to." I would
find the following more useful:

"If you are designing a format specification, assign a URI to the
 following components:

  [  ] Elements
  [  ] Attributes
  [  ] Functions
  [  ] Properties
  [  ] Values of properties or attributes
  [  ] Exceptions
  etc.

"If you are designing an API ..."
"If you are designing a Schema ..."

And so forth. 

 _ Ian

-- 
Ian Jacobs (ij@w3.org)   http://www.w3.org/People/Jacobs
Tel:                     +1 718 260-9447