Re: Review of F&R doc

Steve Harris wrote:
> Sorry, a bit last minute, but here it is:
> 
> 2.3.1 Motivations
> 
> Perhaps it should say something like "Negation by failure is possible 
> in many cases"? I'm not sure that anyone has shown that it can be done 
> in all cases.

yes, that was shown, e.g.

Renzo Angles, Claudio Gutierrez: The Expressive Power of SPARQL. 
International Semantic Web Conference 2008: 114-129

has it implicitly, I also have some proof in some email.
essentially: when you don't have any shared vars between
P1 and P2 in P1 NOT EXISTS P2, you still can do the OPTIONAL !BOUND 
trick by making a "copy" of P1, renaming variables
with and adding some ?v = ?v' FILTER in to the OPTIONAL pattern...

Axel


> 2.3.3 Existing implementation
> 
> Missing closing parenthesis in ARQ description.
> 
> 2.4: Project expressions
> 
> More mention should be made of subqueries, as the two can be used 
> together to answer many usecases.
> 
> 2.4.3 Existing implementation
> 
> Example uses a CONSTRUCT expression, which is not necessarily the same 
> kind of thing as a Projected Expression.
> I don't feel that this example adds anything other the subselect form. 
> Both are required features, so it's fine (in-fact desirable) to 
> combine them IMHO.
> 
> 4.1.1 Motivations
> 
> Should mention bNodes/roundtripping perhaps? There's no stable and 
> standard way to refer to exported bNodes once they leave the SPARQL 
> environment.
> 
> 4.2.3 Existing implementation
> 
> Garlik's JXT supports HTTP PUT and DELETE.
> 
> - Steve
> 
> --
> Steve Harris
> Garlik Limited, 2 Sheen Road, Richmond, TW9 1AE, UK
> +44(0)20 8973 2465  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
> 


-- 
Dr. Axel Polleres
Digital Enterprise Research Institute, National University of Ireland, 
Galway
email: axel.polleres@deri.org  url: http://www.polleres.net/

Received on Tuesday, 23 June 2009 10:35:21 UTC