W3C home > Mailing lists > Public > public-swbp-wg@w3.org > October 2005

IRIs

From: Jeremy Carroll <jjc@hplb.hpl.hp.com>
Date: Mon, 03 Oct 2005 15:51:42 +0100
Message-ID: <4341457E.6080002@hplb.hpl.hp.com>
To: SWBPD <public-swbp-wg@w3.org>


One point about SPARQL that has wider implications is that they use IRIs 
rather than RDF URI References (as used in RDF and OWL).
e.g.
http://www.w3.org/TR/2005/WD-rdf-sparql-query-20050721/#WritingSimpleQueries
[[
The terms delimited by "<>" are IRI references
]]

RDF URI References were intended to be IRIs, but the IRI spec was not 
ready, but is now published as:


http://www.ietf.org/rfc/rfc3987.txt

There are a number of differences between these concepts.

e.g.
    spaces are legal in RDF URI References
    more detailed consideration of specific Unicode characters occurs in 
the concept of IRI
    a treatment of bidirectional control characters is included [e.g. 
IRIs MUST NOT contain bidirectional formatting characters
    (LRM, RLM, LRE, RLE, LRO, RLO, and PDF).]
    IRIs contain a treatment of internationalized domain names, and 
punycode encoding of them

I believe all IRIs are RDF URI references, but the converse is 
definitely false.

Moreover this is an improvement. However, it works within a mindset in 
which RFC 3987 will be the definitive document for W3C specs about 
internationalizeds URIs; somehow it implicitly takes RDF as endorsing 
RFC 3987 and normatively depending on it, with a future reference.


Jeremy
Received on Monday, 3 October 2005 14:51:48 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 21:09:44 UTC