W3C home > Mailing lists > Public > public-rdf-dawg-comments@w3.org > January 2006

Re: [OK?] Re: SPARQL: isURI poorly named

From: Bjoern Hoehrmann <derhoermi@gmx.net>
Date: Sun, 29 Jan 2006 01:57:12 +0100
To: Eric Prud'hommeaux <eric@w3.org>
Cc: public-rdf-dawg-comments@w3.org
Message-ID: <ra0ot1p3dfkekcm8ur4auig9282h6shs2m@hive.bjoern.hoehrmann.de>

* Eric Prud'hommeaux wrote:
>> >I don't follow. Is it the "IRI" part or the "is" part that you object
>> >to? (Perhaps this is in the context of preferring the spec to refer to
>> >URI refs instead of IRIs.)
>> 
>> Well, CSS has url(), XML DSig has URI="", if SPARQL uses isIRI() it's
>> probably just a matter of time until we agree on yet a different term
>> for resource identifiers in which case the name becomes obsolete and
>> we'd have to deal with yet more confusion. Names that omit reference
>> to the current terminology would avoid this problem, e.g. isRI, isRe-
>> source, etc. would all work for me, but the WG probably knows best
>> what'd be a good replacement.
>
>I'm not sure there's a "right" answer here. The WG wrestled with
>current practice and a few issues like URI vs. IRI, and what do people
>understand by "resource". The RDF precedent is that everything is a
>resource. We decided to offer two spellings for the test, "isIRI" and
>"isIRI". The relevent text is scattered through the document:

Well, I can (without much happiness) accept this if the plan is to
revise the RDF Concepts specification such that RDF URI References
are renamed to RDF IRI References (or RDF IRIs maybe) and if the
SPARQL results format is changed to either use <iri> instead of
<uri> or allow both <iri> and <uri> or something better [1].

A better design might be to change RDF Concepts to call those e.g.
Names Terms and SPARQL refers to them with isNAMED() or something
like that but I didn't think much about that yet...

[1] <uri> is defined as xsd:anyURI which is a superset of IRI
    References which is a superset of URIs, so there is quite
    a mismatch and this is much like using <iso8601> instead
    of <time> or <date> which isn't really good design. But I
    wouldn't know what else to call it, given the unfortunate
    naming in RDF Concepts.
-- 
Björn Höhrmann · mailto:bjoern@hoehrmann.de · http://bjoern.hoehrmann.de
Weinh. Str. 22 · Telefon: +49(0)621/4309674 · http://www.bjoernsworld.de
68309 Mannheim · PGP Pub. KeyID: 0xA4357E78 · http://www.websitedev.de/ 
Received on Sunday, 29 January 2006 00:56:23 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 8 January 2008 14:14:50 GMT