Re: Agenda for June 14 Telcon - Revision 1

* Enrico Franconi <franconi@inf.unibz.it> [2011-06-13 16:06+0200]
> 
> On 13 Jun 2011, at 15:15, ashok malhotra wrote:
> 
> > - Issue 42 http://www.w3.org/2001/sw/rdb2rdf/track/issues/42  NULL values in the DM
> >  consensus seems to be to not create triples where the value is NULL with an explanatory note
> >  in the DM spec or as a separate W3C Note
> 
> I am sorry, but as the chair you should be impartial: where from is this consensus coming? 
> Publicly, I heard only your voice and the one by Richard in favour of this proposal.

I have seen juansequeda, Souri, dmcneil and betehess asserting that missing arcs, complemented by the relational schema, give you the information you need to meet use cases. Examining use case like:

┌┤Conctacts├─────┐      Direct Graph:
│ name │ company │      <Conctacts/name=Bob> <Conctacts#name> "Bob" ;
├──────┼─────────┤                           <Conctacts#company> "BobCo" .
│  Bob │   BobCo │      <Conctacts/name=Sue> <Conctacts#name> "Sue" .
│  Sue │    NULL │
└──────┴─────────┘

What companies does Sue represent:
SELECT company          SELECT ?company        
  FROM Conctacts         WHERE { ?sue <Conctacts#name> "Sue" ;     
 WHERE name="Sue"                     <Conctacts#company> ?company }

How many people represent BobCo? (we don't know if Sue does)
SELECT COUNT(*)         SELECT (COUNT(*) AS ?count)
  FROM Conctacts         WHERE { ?who <Conctacts#company> "BobCo" }
 WHERE company="BobCo"

, I'm inclined to agree. Anyone disagree, or want to provide screw cases?


> If you have evidence that there is a consensus, then on which kind of note is there a consensus?
> AFAIK, the only one which makes sense so far would be saying: "This mapping for NULL values is arbitrary since the WG has left unexplored its relationship with the original meaning and behaviour of NULL values in the source RDB. Moreover, the WG did not consider use cases where preservation of NULL values did play a role, and non-backward-compatible changes may happen in the future when the WG will consider these cases."?
> 
> I will not let that happen, in particular after the message by Juan: "(...) because most databases will always have nulls". So, if this is so important, why should the WG be happy to choose a mapping for which it is unknown the relationship (and the complexity of reconstructing the potential relationship) with the NULLs appearing in the original RDB?
> 
> So: either the WG is going to explore the consequences of this choice before approving it, or the choice should be another one.
> 
> cheers
> --e.

-- 
-ericP

Received on Monday, 13 June 2011 18:19:31 UTC