Re: Semantic of the language

On 3/14/2010 5:19 PM, Juan Sequeda wrote:
> Lee,
>
> Good question! I think Marcelo would be the best to answer that
> question.I'm not aware of all the new additions to SPARQL 1.1.

Well, I can help with this part :)

The main additions to the query language are:

+ subselects
+ the ability to project expressions from queries rather than variables only
+ aggregate queries & solution grouping
+ a negation construct (should be similar semantics to optional/!bound, 
so shouldn't change the expressivity)
+ property paths - query arbitrary length paths (?me foaf:knows+ ?known)

>We need
> to prove that all the new additions can be represented in non recursive
> safe datalog. If so.. we are good to go :)

...I can't help so much with this part :)

Lee

> Juan Sequeda
> +1-575-SEQ-UEDA
> www.juansequeda.com <http://www.juansequeda.com>
>
>
> On Sun, Mar 14, 2010 at 3:14 PM, Lee Feigenbaum <lee@thefigtrees.net
> <mailto:lee@thefigtrees.net>> wrote:
>
>     Hi Juan,
>
>     Thanks for the references; I'll try to read through the papers. Do
>     you know if the new additions in SPARQL 1.1 change the expressivity
>     of the language in any significant way for this discussion?
>
>     Lee
>
>
>     On 3/14/2010 2:46 PM, Juan Sequeda wrote:
>
>         Hi everybody,
>
>         Last telcon we said that we were going to talk about the
>         semantics of
>         the language.
>
>         Hopefully you all have the chance to look at these two papers:
>
>         [1] The Expressive Power of SPARQL
>         http://www.dcc.uchile.cl/~cgutierr/papers/expPowSPARQL.pdf
>
>         [2] Translating SQL Applications to the Semantic Web
>         http://userweb.cs.utexas.edu/~jsequeda/sql2sw_dexa2008.pdf
>
>         [1] states that SPARQL and non-recursive safe Datalog have the same
>         expressive power. [2] shows FOL rules, which can be implemented in
>         Datalog in order to automatically generate a rdf schema or owl
>         ontology
>         from a rdb schema. There have been several concerns that the mapping
>         language shouldn't be more expressive than SPARQL. Given this
>         previous
>         work, if we define the mapping language based on non-recursive safe
>         datalog (as shown in [2]), then we can assure that it will have
>         the same
>         expressive power of SPARQL [1]. Once we define the semantics,
>         organizing
>         the syntax should be straightforward.
>
>         Looking forward to our conversation on tuesday.
>
>         Juan Sequeda
>         +1-575-SEQ-UEDA
>         www.juansequeda.com <http://www.juansequeda.com>
>         <http://www.juansequeda.com>
>
>

Received on Sunday, 14 March 2010 21:54:34 UTC