W3C home > Mailing lists > Public > w3c-rdfcore-wg@w3.org > September 2002

Re: XML Base - should be used

From: Dave Beckett <dave.beckett@bristol.ac.uk>
Date: Tue, 03 Sep 2002 11:35:52 +0100
To: Jeremy Carroll <jjc@hpl.hp.com>
cc: w3c-rdfcore-wg@w3.org
Message-ID: <26983.1031049352@hoth.ilrt.bris.ac.uk>

>>>Jeremy Carroll said:
> 
> 
> I would like to suggest that:
> - any test cases with relative URIs should use XML Base
> - any examples in any of our documents that use relative URIs should use XML 
> Base

I'd say, any test cases or examples with relative URIs should be
replace with absolute ones first (http://example.org/ etc)

> - that the syntax and primer documents should suggest that the use
> of XML Base is preferred.

For what cases?  I'd not introduce it unless necessary.  It means
that it is required to have a working (RDF compatible; passes the rdf
core xml-base tests) xml:base support before any other test can be
performed.

> 
> Note: rdf:ID introduces relative URIs.

OK, only when that is present.  

[I thought we were preferring rdf:about="#foo" over rdf:ID="foo" - a
style issue, but the latter does provide some uniqueness checks.]

I counter-propose suggesting xml:base in test cases or examples when
rdf:ID is present and just recommending it when people feel like it
is useful to make all those long URLs shorter - not prefering it
to be present overall.


> The sort of use that I envisage is that where a document on the web includes 
> an xml:base declaration which gives the documents usual base URL.
> 
> Not including such base uris makes teh document ambiguous.
> 
> For example if I GET
> 
> http://www.w3.org/2000/10/rdf-tests/rdfcore/rdfms-not-id-and-resource-attr/test001.rdf
> 
> then the triples that I find are those in
> 
> http://www.w3.org/2000/10/rdf-tests/rdfcore/rdfms-not-id-and-resource-attr/test001.nt
> 
> However, if I get
> 
> http://www.w3.org/2000/10/rdf-tests/rdfcore/rdfms-not-id-and-resource-attr/test001
> 
> then I get different triples, and these are not the ones in any nt
> file in the test area.
> 
> Similarly I might get
> 
> http://www.w3.org/2000/10/rdf-tests/rdfcore/rdfms-not-id-and-resource-attr/test001.rdf
> 
> or
> 
> http://www.w3.org/2000/10/rdf-tests/rdfcore/rdfms-not-id-and-resource-%61ttr/test001.rdf
> 
> and get different results.

Er - don't do that :)

The manifest.rdf files list the absolute URLs to use.  We don't need
to engineer around using the wrong URLs for testing.

Dave
Received on Tuesday, 3 September 2002 06:39:19 EDT

This archive was generated by hypermail pre-2.1.9 : Wednesday, 3 September 2003 09:50:55 EDT