Henry Story wrote:
> Ok. Forget the entailment bit. I don't see what harm it can do as a
> hint though.
>
> That's how CWM interprets "=" by the way.
I regard that as a bug in cwm. Yes, it was useful. But we should
make the syntax different.
-------------------
On 2007-10 -03, at 07:39, Richard Cyganiak wrote:
I agree with what others said: This isn't a good idea as it mixes
syntax and entailment.
My vote would be for doing either of those things:
a) allow "=" as pure syntactic sugar for owl:sameAs, or
+1
> b) allow "=" only immediately after "[", as a syntax for creating a
> named node instead of a blank node; owl:sameAs would still need to
> be written explicitly.
>
Friendly amendment: Introduce a new syntactic token 'is' which can
be used syntactically where a predicate is allowed, constrained that
it can only link a symbol to a bnode.
:Joe :aunt [ is :Sally; age 78 ].
:Joe :aunt [age 78; is Sally ].
[
fund:Name "Associated benevolent";
fund:symbol "ABCDE";
== funds:ABCDE ;
fund: number 1234;
].
The last is interesting when you are generating n3 from a script and
at some point you realize you have a URI you can attribute it to.
Or this 'is' keyword for ==, which is already in fact used:
Joe is father of Fred. # Conventional use reversing things
Fred father Joe.
father of fred brother father of bill. # Use of 'of' by itself
father of fred means [is father of fred]
Joe is [aunt Sally; age 39]. # Use 'is' by itself.
Joe aunt [is Sally; age 78]. # Is in the use above.
Clearly the precedence would have to be figured out carefully.
Tim
Received on Friday, 5 October 2007 19:06:06 UTC