Re: Minutes of the 2010-01-18 telcon for the RDF2RDF WG

On 19/01/2011 13:22, Harry Halpin wrote:
>> On Jan 18, 2011, at 19:05 , Alexandre Bertails wrote:
>> [snip]
>>> Juan, in order to help you with your action, I would like you to
>>> considerer the following:
>>>
>>> [[
>>> CREATE TABLE Debts (
>>> Name varchar(50),
>>> Amount Integer
>>> );
>>> INSERT INTO Debts (Name, Amount) VALUES("juan", 50);
>>> INSERT INTO Debts (Name, Amount) VALUES("juan", 50);
>>> ]]
>>>
>>> Using this very simple RDB [1] example, can you go through all the
>>> phases that lead to the RDF where I owe you 100?
>>
>> Alex, for my understanding: what I would expect to see in RDF are two
>> pairs of identical triples with different subjects. How would a direct
>> produce anything whereby I owe Juan 100? I will owe 50 twice, but the fact
>> that this is a hundred is a step that the RDF per se cannot say...
> Otherwise known as "RDF can't add, and (up until recently) neither can
> SPARQL". Which is a feature, not a bug, re decidability :)
>
> However, I'm assuming Alex is asking with Datalog semantics, do you get
> with those two insert statements *two* distinct pairs of triples
> (multiset) or just one (set)?
>
> Multisets have repeated membership, while sets don't I think.
Thanks Alex for pointing out this case.
This case was included by Eric within the test cases
http://www.w3.org/2001/sw/rdb2rdf/wiki/R2RML_Test_Cases_v1#DB5

So, as you said the question is what is the default mapping for this case?
how we can face this case with R2RML? and what would be the expected result?


Boris

Received on Wednesday, 19 January 2011 13:04:45 UTC