W3C home > Mailing lists > Public > w3c-rdfcore-wg@w3.org > October 2001

Re: Comments on ioctl (was: Re: big issue (2001-09-28#13))

From: Jan Grant <Jan.Grant@bristol.ac.uk>
Date: Wed, 10 Oct 2001 08:26:59 +0100 (BST)
To: Pat Hayes <phayes@ai.uwf.edu>
cc: w3c-rdfcore-wg <w3c-rdfcore-wg@w3.org>
Message-ID: <Pine.GSO.4.31.0110100819490.16853-100000@mail.ilrt.bris.ac.uk>
On Mon, 8 Oct 2001, Pat Hayes wrote:

> >On Fri, 5 Oct 2001, Pat Hayes wrote:
> >
> >[jan wrote]
> >Actually, I had a stronger notion, which was that conceptually, the
> >number (not a numeral representation) should be the "label" in a graph.
>
> Ah, that is a nice idea. It has some odd consequences, though. Graphs
> with number labels cannot be stored inside computers, send over
> packet-switched networks, printed, etc... They have to be Platonic
> graphs, not data structures. And why stop at numbers? ;-)
>

Why indeed. The second half to my thinking is "what to do about it"...

> >However, any concrete realisation of such an RDF graph must, of course,
> >use a representation (be it numeral or otherwise); this is true whether
> >it's an "RDF database" or an "RDF/XML serialisation". If it makes any
> >sense, I'd like to separate the two notions.
>
> Hmm. That gives us an extra layer of representation, which I find
> clunky. There is the lexicalisation which encodes the
> graph-in-the-machine which represents the abstract graph which
> describes the interpretation....

_All_ RDF implementations must represent the graph somehow, yes, and
that usually means an extra layer of representation. I'm not convinced
that this distinction needs to be made apparent in a MT though.

In practice: literals are stored in a computer using some
representation, together with a description of the "type" of literal so
represented.

So that gives us the potential problem, of course, as to determining
which literal types (of a potentially open-ended collection) RDF
implementations should support. I'd say that, at a bare minimum, unicode
strings and langstrings ought to be understood. A nice next step might
be (and I'm not suggesting we adopt this in the WG) the datatypes of
XSD.

Then RDF implementations could offer "best-effort" support for
unrecognised literal types by recording the type indicator of the
literal (as a URI?) together with its lexical representation, and
offering a dumbed-down equality check.

Incoherently yours,
jan

-- 
jan grant, ILRT, University of Bristol. http://www.ilrt.bris.ac.uk/
Tel +44(0)117 9287088 Fax +44 (0)117 9287112 RFC822 jan.grant@bris.ac.uk
(ECHOY GRUNTING) (EERIE WHISPERS) aren't subtitles great?
Received on Wednesday, 10 October 2001 03:28:15 EDT

This archive was generated by hypermail pre-2.1.9 : Wednesday, 3 September 2003 09:40:59 EDT