W3C home > Mailing lists > Public > w3c-dist-auth@w3.org > October to December 2003

RE: Ignoring Versus Not Validating <!DOCTYPE ...>

From: Julian Reschke <julian.reschke@gmx.de>
Date: Sun, 12 Oct 2003 22:18:25 +0200
To: <dennis.hamilton@acm.org>, "Julian Reschke" <julian.reschke@gmx.de>, <w3c-dist-auth@w3.org>
Message-ID: <JIEGINCHMLABHJBIGKBCIEKFIMAA.julian.reschke@gmx.de>

> -----Original Message-----
> From: w3c-dist-auth-request@w3.org
> [mailto:w3c-dist-auth-request@w3.org]On Behalf Of Dennis E. Hamilton
> Sent: Saturday, October 11, 2003 11:31 PM
> To: Julian Reschke; w3c-dist-auth@w3.org
> Subject: Ignoring Versus Not Validating <!DOCTYPE ...>
>
>
>
> Hi Julian,
>
> I am responding to your last comment first, since I think this is
> an issue to have be clear right away.  This is one of the places
> where I see missing precision in definition of the DAV reliance
> on XML 1.0:
>
> 1.	Accepting that the WebDAV specification says that WebDAV
> XML is not to be validated, that is not the same as saying that
> any <!DOCTYPE ...> Document Type Declaration can be ignored.
>
> 2.	XML 1.0 gives specific instructions about what must be done
> with a Document Type Declaration even for a non-validating
> processor. The key statement is in XML 1.0 Recommendation section
> 5.1, Validating and Non-Validating Processors
> <http://www.w3.org/TR/REC-xml#proc-types>.

True. I was oversimplifying.

> 3.	The use of Document Type Declarations in non-validating
> situations is illustrated in the RDF Primer.   See
> <http://www.w3.org/TR/2003/WD-rdf-primer-20031010/#example48> and
> the discussion immediately preceding the example.  The official
> (candidate) RDF Schema for OWL uses this very technique
> (<http://www.w3.org/TR/2003/CR-owl-ref-20030818/#appB>).
>
> (These illustrate the RDF hack, by the way.  In some parts of
> RDF, (adlib) QNames *are* mapped to URIs by concatenation of the
> namespace URI that the prefix stands for, plus the local-name
> part, into a new URI.  The <!DOCTYPE ...> is used to set up
> entity declarations that are used in xmlns:... attributes and in
> directly writing the URI form in literals and attribute values.)
>
> 4.	Rather than have so much sensitivity to out-of-band
> nuances, I think it would be cleaner and more interoperable (for
> DAV, not arbitrary XML) to have the DAV application of XML 1.0
> specify that a Document Type Declaration must not be present.
> Then (1) the XML can't be presumed to be validatable, and (2)
> there is no confusion about the validating versus non-validating
> use as there is when one is provided.
> 	Since, as you say, it doesn't seem to be used, it might be
> a good idea to simplify here and say that it is not meant to be.

I think RFC2518bis is saying that (see
<http://greenbytes.de/tech/webdav/draft-ietf-webdav-rfc2518bis-04.txt>,
section 4.4).

> ...

Julian

--
<green/>bytes GmbH -- http://www.greenbytes.de -- tel:+492512807760
Received on Sunday, 12 October 2003 16:18:59 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 2 June 2009 18:44:05 GMT