Re: Arch Doc: 26 September 2003 Editor's Draft

On Tue, 2003-09-30 at 10:56, Tim Bray wrote:
> Roy T. Fielding wrote:
> 
> > Oh, I see what you are getting at, though now we are also facing
> > confusion over "networked" (meaning connected like a graph) versus
> > spanning the Internet network.  Let's go back to the base concepts:
> > 
> >    The World Wide Web is an information space consisting of resources
> >    that are interconnected by links defined within that space.
> > 
> > Maybe that is sufficient to describe the scope of the Web?  
> 
> Hmm, it seems to me that the fact that everything works over the network 
> is a key defining component.  The description above could apply to 
> Hypercard or even the 1993 version of Microsoft help.   I'm OK with the 
> language above but think we should re-insert "that spans networks" after 
> "information space".  Or "network-spanning" before it.

How about this (mostly copying RF's text):

   The World Wide Web is an information space consisting of
   resources that are interconnected by links defined within that
   space.  This information space is the basis of, and is shared
   by, a number of information systems, including the
   "traditional" Web and emerging technologies such as Web
   Services and the Semantic Web.  Within each of these systems,
   agents (e.g., browsers, servers, spiders, and proxies) acting
   within diverse networks provide, retrieve, create, analyze,
   and reason about resources.

   [Rest of abstract is RF text, though I may split into paras]

This captures the "over the Internet/other networks" part in
"acting within diverse networks". Later, to capture the "network/graph"
aspect, we have in section 2 already:
 
 When a representation of one resource refers to another resource with 
 a URI, a link is formed between the two resources. The
 networked information space is built of linked resources, and the
 large-scale effect is a shared information space. The value of the Web
 grows exponentially as a function of the number of linked resources
 (the "network effect").

Is that enough to capture Roy's point?

> >    A link defines a relationship that can be considered active or
> >    passive, depending upon the type of information system in use.
> >    For example, hypertext browsers consider anchors and in-line image
> >    references to be active links (hyperlinks), whereas a reasoning
> >    system might focus activity on namespace references, a
> >    messaging agent might traverse service descriptions, or a
> >    subscriber might describe "callback" control-points.
> 
> Interesting.  Probably a few words on the taxonomy of linkage is 
> appropriate to this document, it probably lives most naturally in the 
> "identification" section.  Also need to make the point that while an <a 
> href="" is nominally one-way, the relationship is two-way; plus some 
> other things that were written up in the XLink preamble. 

We currently have a section on Hyperlinks in the Representations
section. Some of that TB is referring to (I think) is discussed there.
For now, I will try to introduce RF's text on links into that section;
we can decide to move it at the ftf meeting.

 _ Ian

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