- From: Graham Klyne <GK@NineByNine.org>
- Date: Fri, 01 Feb 2002 13:45:02 +0000
- To: Jeremy Carroll <jjc@hplb.hpl.hp.com>, Sergey Melnik <melnik@db.stanford.edu>, Dan Connolly <connolly@w3.org>, Pat Hayes <phayes@ai.uwf.edu>, Brian McBride <bwm@hplb.hpl.hp.com>
- Cc: www-archive <www-archive@w3.org>
With reference to:
http://lists.w3.org/Archives/Public/www-archive/2002Jan/0160.html
I don't think my proposed handling of ILOBJ and DT-interpretation
adequately reflects the use of multiple typing in RDF.
Example goal: if a property has binaryInteger and decimalInteger as its
range (with the usual datatype mappings), then the only literals that can
be used with that property are "0" and "1".
Thus, I would propose redefining 2a:
[[[
2a. Define a mapping ILOBJ from IP to the powerset of DT. Informally, this
indicates DT values that must all be satisfied when interpreting literal
values used in the object position of the corresponding property.
]]]
The definition of interpretation of a literal then becomes:
[[[
4. The interpretation of a statement of the form aaa bbb "foo" .
is defined thus:
If there exists v in IR such that:
forall d in ILOBJ(I(bbb)), <"foo",v> in DTEXT(d) , AND
<i(aaa),v> in IEXT(I(bbb)) THEN True, otherwise False.
]]]
And the additional DT-interpretation condition:
[[[
<x,y> is in IEXT(I(rdfs:range)) and y in DT then y in ILOBJ(x).
]]]
Thus, to be a valid DT-interpretation, all the the members of DT that are
in the range of property x must be satisfied by any interpretion of
literals as objects of x.
#g
--------------------------
__
/\ \ Graham Klyne
/ \ \ (GK@ACM.ORG)
/ /\ \ \
/ / /\ \ \
/ / /__\_\ \
/ / /________\
\/___________/
Received on Friday, 1 February 2002 08:49:19 UTC