W3C home > Mailing lists > Public > public-lod@w3.org > December 2010

Re: Looking for metalex ontology

From: Alan Ruttenberg <alanruttenberg@gmail.com>
Date: Fri, 3 Dec 2010 15:01:30 -0500
Message-ID: <AANLkTimvE159Tv+2JOSPephQ6_4U18=RXicVj7E1HHkv@mail.gmail.com>
To: Tim Berners-Lee <timbl@w3.org>, Architecture of the World Wide Web <public-awwsw@w3.org>
Cc: Bernard Vatant <bernard.vatant@mondeca.com>, Linking Open Data <public-lod@w3.org>, John Sheridan <johnlsheridan@yahoo.com>, Rinke Hoekstra <r.j.hoekstra@vu.nl>
Tim, I don't understand what you are saying here. My understanding of
http-range14 is that one can always use a 303 response for any entity.
Using a 200 promises the result is an IR. Using 303 doesn't promise
anything. In particular it seems perfectly consistent to return a 303
on the way to returning an information resource.

Some more comments in line.

On Tue, Nov 30, 2010 at 4:26 AM, Rinke Hoekstra <r.j.hoekstra@vu.nl> wrote:
>> 2010/11/30 Tim Berners-Lee <timbl@w3.org>
>> Bernard,
>> You have been tripped up by abuse of content negotiation.
>> Their document says they do conneg.

Which document?

>> cwm http://www.metalex.eu/metalex/1.0
>> gives you data, as cwm only asks for RDF.
>> Following it by hand
>> $ curl -H Accept:application/rdf+xml http://www.metalex.eu/metalex/1.0
>> <html><head>
>> <title>303 See Other</title>
>> </head><body>
>> <h1>See Other</h1>
>> <p>The answer to your request is located <a href="http://svn.metalex.eu/svn/MetaLexWS/branches/latest/metalex-cen.owl">here</a>.</p>
>> <hr>
>> <address>Apache/2.2.11 (Ubuntu) DAV/2 SVN/1.5.4 mod_jk/1.2.26 PHP/5.2.6-3ubuntu4.6 with Suhosin-Patch mod_python/3.3.1 Python/2.6.2 mod_ruby/1.2.6 Ruby/1.8.7(2008-08-11) mod_ssl/2.2.11 OpenSSL/0.9.8g mod_perl/2.0.4 Perl/v5.10.0 Server at www.metalex.eu Port 80</address>
>> </body></html>
>> $
>> *** Note here we are a 303 which means that what we are being redirected to is NOT the
>> ontology, but may be relevant.  The chain of custody is broken, te site does not assert that what follows
>> is the ontology.  But let us follow it anyway:

How would it? I don't believe I've seen any guidance on how one would
do that in a 303 response.
Moreover, it seems reasonable that one might consider an ontology not
to be an information resource, but an ontology *document* to be one.
(because Information resource is not adequately defined)

>> Following the 303
>> $ curl -H Accept:application/rdf+xml http://svn.metalex.eu/svn/MetaLexWS/branches/latest/metalex-cen.owl
>> <?xml version="1.0"?>
>> <!DOCTYPE rdf:RDF [
>>     <!ENTITY owl "http://www.w3.org/2002/07/owl#" >
>>     <!ENTITY owl11 "http://www.w3.org/2006/12/owl11#" >
>> [...]
>>      xmlns:metalex="http://www.metalex.eu/metalex/2008-05-02#"
>>      xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#"
>>      xmlns:owl="http://www.w3.org/2002/07/owl#">
>>     <owl:Ontology rdf:about="">
>> [...]
>> *** Note that here you do get some RDF.
>> Tabulator can read that.  Each term is as you explore marked by a red dot to
>> indicate that it could not be looked up on the web.  Because:
>> ** You do not get information about the namespace
>>   xmlns:metalex="http://www.metalex.eu/metalex/2008-05-02#"
>> instead of the one you originally asked about!

The namespace is irrelevant - it is a syntactic entity. The ontology
URI is relevant, but in that it doesn't say the ontology URI is
<http://www.metalex.eu/metalex/1.0> or otherwise make some assertion
relating that to the ontology I would agree this to be misleading. In
OWL 1 they might have at least included a backwardCompatible
assertion, but better make the ontologyURI be
http://www.metalex.eu/metalex/1.0. In OWL 2 they could have made
either the ontologyIRI or versionIRI be that.

>> So after all that you can see what they are getting  at and how they are thinking.
>> But their linked data is seriously and needlessly broken.

>> To fix it, they should just serve the ontology with 200 from http://www.metalex.eu/metalex/1.0
>> and fix the namespace in it to be that. No conneg.

Again, I don't see how the namespace is at all relevant. And it seems
also reasonable that they would have 303ed and then had the
ontologyURI be http://www.metalex.eu/metalex/1.0

Received on Friday, 3 December 2010 20:02:24 UTC

This archive was generated by hypermail 2.3.1 : Wednesday, 7 January 2015 15:16:10 UTC