10 Nov 2008

See also: IRC log


Harold, DaveReynolds, MichaelKifer, AxelPolleres




<DaveReynolds> Instructions for rrsagent and script: http://dev.w3.org/2002/scribe/scribedoc.htm

<DaveReynolds> Michael: document is almost as long as BLD but they need to read BLD as well anyway


# 7 RIF-BLD as a Specialization of the RIF Framework [RIF-FLD]

* 7.1 The Presentation Syntax of RIF-BLD as a Specialization of RIF-FLD

* 7.2 The Semantics of RIF-BLD as a Specialization of RIF-FLD

* 7.3 The XML Serialization of RIF-BLD as a Specialization of RIF-FLD

* 7.4 RIF-BLD Conformance as a Specialization of RIF-FLD


<AxelPolleres> sorry, I am late.

<AxelPolleres> call-in details as usual?

<DaveReynolds> Michael: suggest replacing 2.1 to 2.5 by just summary of changes without the full summary


<DaveReynolds> Michael: it is not standalone, too long for something where you need to refer back to BLD anyway

7.2 The Semantics of RIF-BLD as a Specialization of RIF-FLD

This normative section defines the precise relationship between the semantics of RIF-BLD and the semantic framework of RIF-FLD. Specification of the semantics that does not rely on RIF-FLD is given in Section Direct Specification of RIF-BLD Semantics.

The semantics of the RIF Basic Logic Dialect is defined by specialization from the semantics of the semantic framework for logic dialects of RIF. Section Semantics of a RIF Dialect as a Specialization of the RIF Framework in [RIF-FLD] lists the parameters of the semantic framework that can be specialized. Thus, for RIF-BLD, we need to look at the following parameters:

* The effect of the syntax.

RIF-BLD does not support negation. This is the only obvious simplification with respect to RIF-FLD as far as the semantics is concerned. The restrictions on the signatures of symbols in RIF-BLD do not affect the semantics in a significant way.

* Truth values.

The set TV of truth values in RIF-BLD consists of just two values, t and f such that f <t t. The order <t is total.

* Datatypes.

RIF-BLD supports the datatypes listed in Section Datatypes of [RIF-DTB].

* Logical entailment.

Recall that logical entailment in RIF-FLD is defined with respect to an unspecified set of intended semantic structures and that dialects of RIF must make this notion concrete. For RIF-BLD, this set is defined the set of all models.

* Import directive.

The semantics of the two-argument Import directive is given in [RIF-RDF+OWL]. The semantics of the one-argument directive is the same as in RIF-FLD.


<DaveReynolds> Michael: suggests shortening the specialization of 2.1 to 2.5 futher

<DaveReynolds> Michael: suggests not duplicating examples but referring back to BLD

<DaveReynolds> Michael: just focus on comparative analysis

<AxelPolleres> hmmm, my own problem is, I am not 100% sure where I can help at the moment. (I was a bit offline the last two weeks)

<AxelPolleres> I think for a 1st WD, it is not important to have self-containedness.

* As a direct specification, independently of the RIF framework for logic dialects [RIF-FLD], for the benefit of those who desire a direct path to RIF-BLD, e.g., as prospective implementers, and are not interested in extensibility issues. This version of the RIF-BLD specification is given first.

* As a specialization of the RIF framework for logic dialects [RIF-FLD], which is part of the RIF extensibility framework. Building on RIF-FLD, this version of the RIF-BLD specification is comparatively short and is presented in Section RIF-BLD as a Specialization of the RIF Framework at the end of this document. This is intended for the reader who is already familiar with RIF-FLD and does not need to go through the much longer direct specification of RI

<AxelPolleres> That's fact, that

<AxelPolleres> 's not just because I joined ;-)

<AxelPolleres> Can someone sum up the problem with the PRD semantics again? or paste the pointer to the resp. mails?

<AxelPolleres> So, it sounds like BLD needs change, I am not really clear on the implications on Core editing. If taking over a part, I would prefer to have it clearly separated from BLD changes.


(Atom/Expr positional arguments role, with fixed 'ordered' attribute, containing n TERMs)

- args


<AxelPolleres> We do have slots in the head, yes?

<DaveReynolds> Axel - yes frame slots

" As a direct specification"



Definition (Models). A multi-structure I is a model of a formula, f, written as I |= f, iff TValI(f) = t. Here f can be a RIF-Core document or a non-document formula. ?


Maybe Gary can help us next week (in the Core telecon) with understanding the relevant parts of the PRD semantics?

"In order to allow production rule systems and logic programming systems to interchange rulesets via RIF-Core we define a strictly safe subset of RIF-Core and only require minimally conformant RIF-Core consumers to correctly translate this safe subset."


<DaveReynolds> Axel: could replace "strictly safe" by "safe" and so replace "minimally conformant" by "safely conformant"

<AxelPolleres> We can just talk about "safe" instead of "strictly safe" ... then we can just rename "minimally conformant" to "safely conformant"

<AxelPolleres> I can do it now.

Summary of Action Items

[End of minutes]

Minutes formatted by David Booth's scribe.perl version 1.133 (CVS log)
$Date: 2008/11/10 17:23:14 $