W3C home > Mailing lists > Public > public-rdfa-wg@w3.org > October 2010

Re: [Fwd: ACTION-487 Assess potential impact of IRI draft on RDF/XML, OWL, and Turtle]

From: Toby Inkster <tai@g5n.co.uk>
Date: Thu, 28 Oct 2010 19:03:02 +0100
To: Ivan Herman <ivan@w3.org>
Cc: nathan@webr3.org, RDFA Working Group <public-rdfa-wg@w3.org>
Message-ID: <20101028190302.6d758f34@miranda.g5n.co.uk>
On Thu, 28 Oct 2010 14:23:51 +0200
Ivan Herman <ivan@w3.org> wrote:

> Guys, help me out please: what is the difference between 3986 and
> 3987?

RFC 3986 is URI; RFC 3987 is IRI. URIs are US-ASCII only; IRIs are
Unicode and allow characters beyond U+007F in many places. Many
protocols and formats are not Unicode aware, so the IRI RFC defines a
mapping from IRIs to URIs. (A mapping in the reverse direction is
unnecessary as all URIs are automatically IRIs.)

All things being equal, we probably want to use IRIs - they allow
people to use non-Latin characters in identifiers which is likely to
be a boon for RDFa's acceptability in cultures where the usual
alphabets are not derived from the Latin alphabet (e.g. Chinese,
Greek, Japanese, Thai, Iranian, etc).

The problem is that RDF itself uses URIs as it was defined prior to to
existence of IRIs, so this would be an inconsistency between RDF and
RDFa. However, this doesn't seem to have proved a practical problem for
SPARQL which uses IRIs. We should get advice from TAG as they may be
able to provide us with information on what direction RDF is likely to
go (stick with URIs or switch to IRIs).

Toby A Inkster

Received on Thursday, 28 October 2010 18:03:40 UTC

This archive was generated by hypermail 2.4.0 : Friday, 17 January 2020 17:05:22 UTC