RE: Banning relative - No real damage?

The specification of x-schema is that it takes a URL reference that locates
the schema. As such the absolute URI of the form

x-schema:http://myserver/myschema.xdr

is both absolute as URI and for resolution,

x-schema:#schema

is not absolute for resolution.

How this differs from file:///foo is not clear to me. The second case above
is also designed to be used for resolution on a single system. We (not I but
the people designing XDR) defined the semantics of x-schema to be a
reference to a schema without implying a schema at the end of a HTTP URL
reference.

Regards
Michael


> -----Original Message-----
> From: Tim Berners-Lee [mailto:timbl@w3.org]
> Sent: Wednesday, June 14, 2000 2:03 PM
> To: Michael Rys; keshlam@us.ibm.com; xml-uri@w3.org
> Subject: Re: Banning relative - No real damage?
> 
> 
> 
> -----Original Message-----
> From: Michael Rys <mrys@microsoft.com>
> Date: Thursday, June 08, 2000 8:38 PM
> Subject: RE: Banning relative - No real damage?
> 
> 
> >Please note that the relative schema references (XDR and not XSD) are
> >expressed using an absolute URI that conveys relative 
> resolution semantics
> >(using the x-schema "protocol").
> 
> 
> That is broken.  You can't do that.  A[n absolute] URI cannot 
> be relative -
> it
> is all you need to quote to refer to something.  It seems to 
> be an attempt
> to
> wriggle around the absolute/relative question but it isn't a 
> URI.  (What
> happens
> when I bookmark such a thing?!).  Could you fix it in future releases?
> It does to the URI spec what <b><i>foo</b>bar</i> does to XML.
> 
> (Don't be confused with file:///foo which is a URI which is 
> designed for use
> on a sungle system, and not designed for interchange across systems
> Some have refered to this as relative, but that was not the intent.
> There is no assumption that a file:///foo URI has any meaning 
> on any other
> system
> or should ever be transmitted to another file system)
> 
> Tim BL
>

Received on Wednesday, 14 June 2000 19:14:01 UTC