| ID |
Question |
Name |
| Questions for From SPARQL to Rules Ankesh |
In section 5.3- SPARQL as a Rules Language- I am able to see how we could use the translation to datalog under ASP semantics as a rule language, but am having some difficulty in seeing SPARQL itself as a rule language. I can think of a rule base in SPARQL as a set of rules that have a graph pattern as their body and a construct statement in the head. Can you elaborate on what would be increase in complexity of such a language- especially when some kind of rule ordering would have to taken into account, and rules (therefore pattern matching) may have to be applied more than once? |
Ankesh Khandelwal |
| Questions for From SPARQL to Rules Jesse Weaver |
Although I think I know the answer, I would like to ask the presenters for their answers. So SPARQL can be reduced to datalog rules. What is the significance of this reduction? (Put another way, why is this important?) |
Jesse Weaver |
| Questions for From SPARQL to Rules Jesse Weaver 2 |
In section 2.2, the valuation of a FILTER R on substition \theta, denoted R\theta can result in an error if R is of the form R=(R1 ^ R2) and (R1\theta = error or R2\theta = error). Wouldn't it make more sense to redefine this condition as R=(R1 ^ R2) and (R1\theta = error or R2\theta = error) and (R1\theta != false and R2\theta != false)? This would be similar to the condition if R=(R1 v R2). |
Jesse Weaver |
