RE: Arch Doc: 18 Sep: abstract

In general, I like where this is going.  I think that the characterization
of "Web services" is incorrect though.  There are a huge number of arbitrary
interactions that occur on the web through POST and other interactions that
are not just retrieval of representations.  So I think that we should not
talk about "retrieval of representations" and rather talk about interacting
with the resources.  I also suggest removing "information" from "information
resources".

> > +The World Wide Web is a networked information space consisting
>
> of information resources which
> have global identifiers. This allows descriptive metadata to describe
> resources, and hypertext systems to be built with link between
> resources.

+The World Wide Web is a networked information space consisting of resources
which have global identifiers.

>
> >   Web architecture encompasses both
> > +protocols that define the information space by way of
> > +identification
>
> the retrieval of representations, and also
> "Web Services" protocols which allow arbitrary interactions between
> agents.

I'd rather stick with Roy's original text.

Cheers,
Dave