RE: RFC2518 (WebDAV) / RFC2396 (URI) inconsistency

> From: uri-request@w3.org [mailto:uri-request@w3.org]On Behalf Of James
> Clark
> Sent: Wednesday, November 21, 2001 5:03 AM
> To: Jim Whitehead; w3c-dist-auth@w3.org; uri@w3.org
> Subject: RE: RFC2518 (WebDAV) / RFC2396 (URI) inconsistency
>
> ...
>
> I don't think the issue is anything to do with whether URIs are
> hierarchical. From reading your message, the requirements appear to be as
> follows:
>
> - DAV wants to identify elements by a URI not a URI/local name pair

I think this should read: historically, in the development of WebDAV, there
was a phase where the spec assumed the existence of a mapping like this. I
don't think that RFC2518 as published does this.

> - DAV wants the generated URI for an element <D:creationdate/> to be
> dav:creationdate
>
> - DAV to generate the URI that identifies an element from the namespace
> identifier and local name
>
> You might think that these requirements imply that the namespace

I don't think that these actually *are* requirements. At least not according
to RFC2518.

> identifier
> for DAV must be "dav:", but in fact they don't, because there is no
> requirement that the method that you use to generate the URI that
> identifies an element is to concatenate the namespace identifier and the
> local name or the element.  You can solve the problem simply by choosing
> some other method of generating the URI, which allows the namespace
> identifier to be a legal URI, as required by the XML Namespaces Rec.

Is there any (published one) that works well with arbitrary namespace names
and element names?

Received on Wednesday, 21 November 2001 05:20:38 UTC