RE: Comments on new draft of 3023bis (fragments, xpointer)

> From: Chris Lilley [mailto:[email protected]] 
> Sent: Friday, 25 February, 2005 7:15
> To: Paul Grosso
> Cc: Martin Duerst; Henry S. Thompson; [email protected]; 
> MURATA Makoto; Philippe Le Hegaret
> Subject: Re: Comments on new draft of 3023bis (fragments, xpointer)
> 
> On Friday, February 25, 2005, 5:20:01 AM, Paul wrote:
> 
> PG> That's why I think it is important to say that the
> PG> fragment identifier syntax for the XML type is just
> PG> the XPointer Framework and Element() scheme, and
> PG> nothing else--all other schemes must be ignored.  
> 
> To be clear - you are saying that
> 
> - for application/xml
> - for unknown +xml media types being processed as xml
> 
> the fragment identifier syntax for the XML type is just the XPointer
> Framework and Element() scheme, and nothing else--all other 
> schemes must be ignored.

Correct.

> "Ignored" needs to be further defined (ie, it does not cause 
> an error).

I mean "ignored" in the sense that is defined in the XPointer Framework 
spec which explains how to handle an xpointer with schemes that are
not recognized/supported.  (I agree we need to get precise wording,
but I hope what I'm trying to say is clear by reference to the
XPointer Framework spec.)

> And to be clear, you are *not* saying that
> 
> - any +xml media type is forbidden from defining its own, richer,
> fragment identifier syntax in addition to that defined fro +xml as a
> whole.
> 
> Right?

Correct--another media type can say whatever it wants in this regard.

paul