W3C home > Mailing lists > Public > www-rdf-interest@w3.org > June 2002

RE: N3 and N-Triples (was: RDF in HTML: Approaches)

From: Andy Seaborne <andy.seaborne@hp.com>
Date: Wed, 5 Jun 2002 09:55:54 +0100
To: "'Bill de hÓra'" <dehora@eircom.net>, "'RDF Interest'" <www-rdf-interest@w3.org>
Message-ID: <5E13A1874524D411A876006008CD059F038D3940@0-mail-1.hpl.hp.com>

Bill de hÓra:
> Having RDF/XML wuth shortcomings doesn't seem to help
> get standardized, global interchange of knowledge between
> disparate systems, it seems to get in the way of that.

I agree.  As Patrick noted, the RDF graph is the important concept - and
currently, some RDF graphs can’t be written in the RDF/XML syntax.  This
does not help interchange.

The current RDF/XML syntax can be extended, indeed this would not
necessarily need invalidate current RDF data.  The same mechanism that
N-Triples uses (giving bNodes a label solely for the purpose of
syntactically recording a graph in a separate label space from URIs) can
be transferred into RDF/XML

(1) have attribute rdf:bnode='a' (c.f. rdf:resource) or
or 
(2) use a parseType='bNode'.

The first is more compact because there is no trailing tag.  Both put
the onus on the RDF parser and are not using XML features but then
bNodes aren't in XML in the first place.

	Andy

-----Original Message-----
From: Bill de hÓra [mailto:dehora@eircom.net] 
Sent: 4 June 2002 16:37
To: 'RDF Interest'
Subject: RE: N3 and N-Triples (was: RDF in HTML: Approaches)



 
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1


> -----Original Message-----
> From: www-rdf-interest-request@w3.org
> [mailto:www-rdf-interest-request@w3.org] On Behalf Of Patrick
> Stickler 
>
> I'm very sympathetic to the shortcomings of RDF/XML and the
> need for easier/better serializations of RDF knowledge -- but 
> *NOT* at the expense of standardized, global interchange of 
> knowledge between disparate systems.

Patrick,

I suspect the end effect is the same. Having RDF/XML wuth shortcomings
doesn't seem to help get standardized, global interchange of knowledge
between disparate systems, it seems to get in the way of that.

 
> That's the whole point of standards. You seem to be
> suggesting that we just ignore the standards and do things as 
> we see best.

The whole point of RDF standards is to create standard RDF machinery
whose behaviour is consistent across implementations. We're looking to
commoditize RDF software, make no mistake. If an RDF standard is too
hard to implement consistently, it is economically and politically
useless in that respect. Having a standard so we can say we have one is
folly. 


> If N3 or NTriples are adopted as *official*, *standard*
> serializations for RDF interchange, great!
> 
> But until they are, I am strongly opposed to seeing them
> (mis)used in place of RDF/XML for global interchange.

A good analytic argument can be made to use what works, not what is
supposed to work. 

Bill de hÓra

-----BEGIN PGP SIGNATURE-----
Version: PGP 7.0.4

iQA/AwUBPPyfqOaWiFwg2CH4EQKDzQCfapKi7npu7J/A+CJiWIR2iEdmwsUAnjVc
3DX2Cjen5kPQthckw5xfcL8y
=P41E
-----END PGP SIGNATURE-----
Received on Wednesday, 5 June 2002 04:57:38 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 7 December 2009 10:51:54 GMT