Re: namespace names, URI-references, and system literals

C M Sperberg-McQueen wrote:
> 
> >Date: Wed, 22 Jul 1998 22:01:42 -0500
> >From: Dan Connolly <connolly@w3.org>
> >
> >Er... this is a bug in the XML 1.0 spec too...
> >I'm not sure I every formally reported it.
> >
> >The thing inside the quotes in a system literal
> >is a URI reference not a URI. (i.e. "foo" is
> >allowed, as a relative reference.)
> 
> Let me make sure I understand correctly.  The XML 1.0 spec says "The
> SystemLiteral that follows the keyword SYSTEM is called the entity's
> <term>system identifier</term>. It is a URI, which may be used to
> retrieve the entity."  But the examples make clear that we intend that
> both relative and absolute references (e.g.  '../../dtds/mydtd.dtd'
> and 'http://www.uic.edu/~cmsmcq/stuff/dtds/mydtd.dtd') should be
> legal.  So the spec is inexact, and ought to say "It is a URI
> reference, ..."  and not "It is a URI, ..."
> 
> Is that what you mean?

Yes.

> (Or alternatively do you mean "XML 1.0 makes it a URI reference, when
> we wanted it to be just a URI ..."?)

no.

> I've added it to my list of problems to be fixed.  Gotta get to that
> someday.

 [copy to xml-editor for tracking purposes]

-- 
Dan Connolly
http://www.w3.org/People/Connolly/
phone:+1-512-310-2971 (office)

Received on Thursday, 23 July 1998 11:31:28 UTC