Re: Review of GRDDL Documents and Issues (intro)

At 03:42 PM 9/28/2006 -0500, Dan Connolly wrote:

[...]
> >         There are many ways to look at the content of XML documents that
> >         exist on the web.
>
>I think it's worth including XHTML in the 1st para.

Agreed. But I also thinks that it's worth saying that "There are many ways..."

Also, please add "from orders to invoices," as a nod to UBL.

> >         How, for example, does software discover the author of a poem, a
> >         spreadsheet and an ontology? And how can software determine 
> whether
> >         authors of each are in fact the same person.
>
>I particularly liked that bit; it stresses RDF's strengths.

Thanks. That's what I was going for. Please add a ? at the end of the 
paragraph.

>[...]
>I softened it to..
>
>| Software tools that use the Resource Description Framework
>| naturally work with documents whose data is encoded using RDF/XML.

Better.

>This next para is too strong, to the point that it's incorrect;
>I'd like to find words that are correct and capture the
>gist, but for now, I just deleted it:
>
> >         There are essentially three parts to using GRDDL. Firstly, an
> > XML document
> >         must identify itself as a candidate for use by a GRDDL-aware
> > processor.
>
>Not really; it can just refer to a namespace document; the namespace
>document can be made GRDDL-happy after the fact, even.

My point is that before a document will be acted upon it must somehow
be recognized as being a candidate. That may be by explicit use of the
GRDDL profile, or semi-indirectly by reference to the GRDDL namespace
and use of the TRANSFORMATION attribute, or indirectly by reference
to a namespace or profile that contains appropriate GRDDL stuff.

> >         Secondly, the candidate document must provide a link to one or
> > more decoding
> >         algorithms.
>
>Either directly or indirectly.

Agreed.


> >  Thirdly, the GRDDL-aware processor must traverse the link and
> >         execute the target in order to yield the resulting RDF.
>
>That refers to "GRDDL-aware processor" as if it had been introduced
>earlier. It hasn't. I have in mind to write an appendix about
>a sample implementation of GRDDL, but until then, I prefer
>to stay silent about processors.

OK, we don't necessarily have to mention GRDDL-aware agents at this point.


Anyway, if not the paragraph as I wrote it, then at least something that frames
the discussion that follows will be helpful. I find it hard to follow the 
documents
because of their lack of a unifying model.

Regards,

Murray