Re: First order logic and SPARQL

On Sep 6, 2010, at 12:13 AM, Bob MacGregor wrote:

> Hi Pat,
> 
> On Sun, Sep 5, 2010 at 6:31 PM, Pat Hayes <phayes@ihmc.us> wrote:
> 
> On Sep 5, 2010, at 4:48 PM, Bob MacGregor wrote:
> 
> > My personal interest is in a query language for RDF that's easy to use, and, among other things,
> > has a negation operator that is intuitive.
> 
> Id be interested to know what you consider to be intuitive here. Is negation by failure intuitive for most Web sources? Do you routinely conclude, from a failure to find a sentence asserted on a website, that it is false?
> 
> Fundamental to your argument seems to be "sentence asserted on a website".   If I grabbed
> triples from some random Website, I might not be confident in using NAF.  But I don't do that.  I work
> with graphs that I've built from sources I trust, and I know which parts of the graph are expected to
> be complete, and NAF is perfect for those parts.  

Well, bully for you, but SPARQL is supposed to be a standard for use with RDF on the Web. These nice assumptions of completeness just where you expect it cannot be sustained in the wider world of RDF data, and there is no way to transmit them (the assumptions) even when they are correct. So NAF is not appropriate for SPARQL. 

Pat

> 
> - Bob

------------------------------------------------------------
IHMC                                     (850)434 8903 or (650)494 3973   
40 South Alcaniz St.           (850)202 4416   office
Pensacola                            (850)202 4440   fax
FL 32502                              (850)291 0667   mobile
phayesAT-SIGNihmc.us       http://www.ihmc.us/users/phayes