"URI reference" "checking" [was: Minutes for XML Core WG telcon of 2006 February 1]

I guess this means this thread is not, in fact, closed
as I had hoped.

While I could live with such a note, I'm not eager to
add such.  I could imagine an application using XLink
for which interaction is not appropriate, and the right
thing to do is for the app to ignore any bad link.  I
just don't see the point of telling an application what
to do.  XLink is a tool--let the application using it
figure out how to use it.  

Philosophically, I think that XLink "connects" different 
points in web space in some fashion.  How those points are 
determined isn't XLink's business.  XLink should just point
to XPointer and/or URI RFC's for that.  XLink should let
those specs say what to do about bad URIs. 

paul

> -----Original Message-----
> From: Henry S. Thompson [mailto:ht@inf.ed.ac.uk] 
> Sent: Friday, 2006 February 03 9:49
> To: Grosso, Paul
> Cc: public-xml-core-wg@w3.org
> Subject: Re: Minutes for XML Core WG telcon of 2006 February 1
> 
> > "URI reference" "checking"
> > . . .
> > Henry's penultimate email asked if this should generate
> > an official objection (which question was never answered),
> > so we can handle this thread the same as the others in
> > this category (which I assume will be to generate an
> > official objection).
> >
> > Henry, do you agree that this thread should now
> > be considered closed? 
> 
> The draft spec currently reads:
> 
> 
>  *Note*: XLink 1.0 explicitly did not require applications to check
>  that the value of the xlink:href attribute conformed to the syntactic
>  rules of a URI. While [RFC 3986] has clarified the syntactic rules,
>  this specification follows the lead of XLink 1.0 (and many other
>  specifications) and does not impose any new conformance testing
>  requirements on XLink applications in this area.
> 
> I'd be happy to add a sentence to address Bjoern's concerns [1] along
> the following lines:
> 
>  "Applications which detect, either directly or via standard
>   libraries, violations of the syntactic rules of [RFC 3896], SHOULD
>   NOT recover silently."
> 
> What do others think?
> 
> ht
> 
> [1] 
>
http://www.w3.org/mid/ljkft1t4k57k8gnr6c7t5enhmn5gg4g69n@hive.bjoern.hoe
hrmann.de

Received on Monday, 6 February 2006 15:01:31 UTC