- From: Seaborne, Andy <andy.seaborne@hp.com>
- Date: Wed, 22 Dec 2004 17:06:33 +0000
- To: jos.deroo@agfa.com
- Cc: 'RDF Data Access Working Group' <public-rdf-dawg@w3.org>
jos.deroo@agfa.com wrote:
> Andy Seaborne wrote:
>
>>[...]
>>
>>Turning into a test case:
>>
>>== Data 1:
>>@prefix : <http://example.org/ns#> .
>>
>>:x a :foo .
>>:x a :bar .
>>
>>:y a :bar .
>>
>>== Query 1:
>>PREFIX : <http://example.org/ns#>
>>
>>SELECT ?r
>>WHERE (?r rdf:type ?type) # Get all the things with a type
>> UNSAID (?r rdf:type :foo)
>>
>>== Result Set 1:
>>------
>>| r |
>>======
>>| :y |
>>------
>
>
> not only seems to assume closed world but also assume unique names
> i.e. there is an implicit assumption that
> :x owl:differentFrom :y.
> and unless that is stated explicitly, I can't and want make
> such result
>
> also still think it would be more explicit to say
> UNSAID <in-which-graph> ( triple )
This happens anyway in the untrusted graph proposal:
UNSAID { pattern } would be in the default KB.
SOURCE <uri> { UNSAID { ..pattern.. } }
would be in the graph named <uri>.
There isn't a case of an indeterminant case in the untrsted graph proposal
(I hope!).
Andy
>
>
> hm.. now realize that my implementation of UNSAID using
> log:notIncludes has to be much improved.. not sure if
> it will succeed..
>
>
Received on Wednesday, 22 December 2004 17:07:03 UTC