Re: Review XML Data in RIF

Hi Michael,

Thanx for the quisk feedback.

Michael Kifer <kifer@cs.stonybrook.edu> wrote on 19/10/2010 19:55:34:
> 
> I am ok with the revised "conservative" notion of entailment + the 
> explanations
> per the above.
> 
> [...]
> 
> I think the doc should include the revised definition (the 
> conservative one that
> is currently there as of today). We can add an editor's note with the 
earlier
> definition and state that it is unclear whether that other notion is 
also
> needed.

I changed the doc to say:
Definition (Logical entailment of a RIF BLD+XML data combination). Let φ 
and ψ be (document or non-document) formulas. We say that a RIF BLD+XML 
data combination <φ, E, S> entails the RIF BLD+XML data combination <ψ, E, 
S>, denoted <φ, E, S> |= <ψ, E, S>, if and only if every model of the RIF 
BLD+XML data combination <φ, E, S> is also a model of the RIF BLD+XML data 
combination <ψ, E, S>.  ? 
Notice that, in the case where E and S are empty, that is, if a RIF-BLD 
formula is combined with an empty XML data set and an empty set of XML 
schema definitions, the interpretation of that RIF BLD formula under a RIF 
BLD+XML data combined multi-interpretation (?, ?, ?) is still different 
from its interpretation under the standard RIF BLD semantics as defined in 
[RIF BLD]. 
Example 2.4. Let expr be a well-formed XPath 2.0 expression, and let 
φ = Forall ?x ?y ?z (?x["fn:data(expr)"->?z] :- ?x["expr"->?y]) 
ψ = Forall ?x ?y ?z (?y["fn:data(.)"->?z] :- ?x["expr"->?y]) 
Notice that <φ, ?, ?> |= <ψ, ?, ?>, provided that the pair (fn, 
http://www.w3.org/2005/xpath-functions) belongs to the statically known 
namespaces when the XPath expressions are evaluated; but φ |? ψ. This is 
because the xs:string constants "expr", "fn:data(expr)" and "fn:data(.)" 
are interpreted as simple string constants, if the formulas are 
interpreted under the standard RIF BLD semantics; whereas they are 
interpreted according to the XPath 2.0 semantics of their literals, if the 
formulas are interpreted under the semantics of RIF+XML data combination. 
Editor's Note: An alternative, less conservative definition for logical 
entailment in a RIF BLD+XML data combination would be: Let φ and ψ be 
(document or non-document) formulas. We say that φ entails ψ in a RIF 
BLD+XML data combination if and only if every model (E, ?, S) of any RIF 
BLD+XML data combination <φ, E, S> is also a model of the RIF BLD+XML data 
combination <ψ, E, S>.? The use cases for the broader definition remain 
to be examined.

Also, I added the following statement in the previous section, on the 
combined interpretation of RIF BLD non-doc formulas.
Notice that the semantics of XPath 2.0 expressions may entail that two 
different expressions, evaluated with respect to two different [context 
nodes]XP, must select the same sequence; and, therefore, that the two 
different pairs of [context node]XP and XPath expression must have the 
same RIFValue. This implies, per clause 1, in the definition of a RIF 
BLD+XML data combined interpretation, above, that the RIF BLD+XML data 
combined interpretation of frames with different objects and attributes 
may be constrained by the semantics of XPath 2.0 expression, even in the 
absence of XML data in the combination (see, for instance, example 2.4, 
below). 

And I added in the introduction the following statement:
The semantics of RIF combinations with XML data imports the semantics of 
the XPath2.0 expressions that are used to relate RIF formulas to XML data: 
as a consequence, the interpretation of RIF formulas in a combination with 
XML data is more restrictive than their interpretation outside of a 
combination, even if the XML data in the combination is empty. Notice, 
however, that, in practice, there can never be an ambiguity regarding 
under what semantics a RIF document must be interpreted: a RIF document is 
interpreted under the combined semantics specified in the document if and 
only if it imports XML data or XML schemas explicitly. 

Does that work with you?

If yes, and neither Gary nor Adrian has objections, we can publish as WD.

Please, let me know.

Cheers,

Christian

IBM
9 rue de Verdun
94253 - Gentilly cedex - FRANCE
Tel./Fax: +33 1 49 08 29 81

Sauf indication contraire ci-dessus:/ Unless stated otherwise above:
Compagnie IBM France
Siege Social : 17 avenue de l'Europe, 92275 Bois-Colombes Cedex
RCS Nanterre 552 118 465
Forme Sociale : S.A.S.
Capital Social : 612.509.964 €
SIREN/SIRET : 552 118 465 03644