Re: a blank node issue

On 2011-03-02, at 16:21, William Waites wrote:

> * [2011-03-02 15:54:15 +0000] Richard Cyganiak <richard@cyganiak.de> écrit:
> 
> ] On 2 Mar 2011, at 14:18, William Waites wrote:
> ] > maybe some convention or
> ] > standard for skolemising blank nodes so they can be 
> ] > referred to might be a good thing?
> ] 
> ] There is already a convention/standard for skolemising blank
> ] nodes: just use a URI instead.
> 
> And if I am not the author of the statements but I 
> want to consume and manipulate them? Surely you aren't
> suggesting getting rid of bnodes completely?
> 
> The vendor extensions that I was talking about do 
> exactly this. The problem is they do it in different
> ways.

Yup, it would be nice to have some sort of consensus on how to do this.

There's a getout in SPARQL 1.1, as you can do URI(_:somebnode) to try and force a skolemisation, but I don't know how many systems support it, and it's awkward compared to a simple lexical transform.

- Steve

-- 
Steve Harris, CTO, Garlik Limited
1-3 Halford Road, Richmond, TW10 6AW, UK
+44 20 8439 8203  http://www.garlik.com/
Registered in England and Wales 535 7233 VAT # 849 0517 11
Registered office: Thames House, Portsmouth Road, Esher, Surrey, KT10 9AD

Received on Wednesday, 2 March 2011 17:16:40 UTC