- From: Brian McBride <bwm@hplb.hpl.hp.com>
- Date: 30 Jul 2003 11:23:16 +0100
- To: Martin Duerst <duerst@w3.org>
- Cc: Graham Klyne <gk@ninebynine.org>, rdf core <w3c-rdfcore-wg@w3.org>, i18n <w3c-i18n-ig@w3.org>
On Tue, 2003-07-29 at 22:21, Martin Duerst wrote:
> Hello Brian,
>
> I think it is important to stay as close as possible to the
> actual specs, so I appreciate David's corrections,
me too.
> but I think
> they are irrelevant to your main point at hand,
Sorta - the only thing I was really trying to do was to help create a
notation where we could communicate clearly. I appreciate the
improvements you have made to it. Its always such a relief when one
gets to the point of feeling like we can communicate effectively.
[...]
> >C(x) is cannonicalization of x, encoded as a UTF8 octet sequence, e.g.
> >C("&") is the octet sequence corresponding to "&".
>
> This is dangerous, because you have missed one escaping level.
Right - another mistake. Should have been:
C("&") = ...
[...]
> (4)
> Concrete Syntax: <eg:prop pt="L"><br/></eg:prop>
>
> Abstract Syntax: "<br></br>"^^rdf:XML
>
> Denotation:
> sequence(markup('<br>'), markup('</br>'))
This seems like the key notion you are introducing. That we could have
defined/adopted a different "value space" for XML. We did, at one point
have something similar on the table; there was a suggestion that
parseType="Literal" should generate an RDF representation of the
infoset. I can't remember why we dropped it - those brian (sic) cells
have gone.
I think we've got greater clarity now, so I can butt out again.
Brian
Received on Wednesday, 30 July 2003 06:25:31 UTC