W3C home > Mailing lists > Public > public-rif-wg@w3.org > May 2009

ACTION-779: Review SWC

From: Gary Hallmark <gary.hallmark@gmail.com>
Date: Mon, 11 May 2009 12:34:22 -0700
Message-ID: <499257640905111234w34a7b0bfl96494e6f50228f13@mail.gmail.com>
To: RIF WG <public-rif-wg@w3.org>
No major issues to report. Good work, Jos!

One overall minor issue: rule vs rules - e.g. rule language or rules
language, rule engine or rules engine, rule system or rules system,
rule publisher or rules publisher, etc. Probably we should pick 'rule'
or 'rules' everywhere (across all our docs)

Minor issues in document order:

Section 1. 1st sentence - either remove the word "logical" or change
to "logical rules and production rules"

last 2 bullets in Section 1, just before Section 2 are very dense and
hard to  follow. Suggestion:

RDF triples are written using the Turtle syntax [Turtle]: triples are
written as s p o ., where s, p, and o are one of the following:
 - a blank node, written as _:x,
 - an IRI, delimited with '<' and '>',
 - a compact IRI, written as prefix:localname,
 - a plain literal without language tag, written as "literal",
 - a plain literal with language tag, written as "literal"@lang, or
 - a typed literal, written as "literal"^^datatype-IRI.

The following namespace prefixes are used throughout this document:
 - ex refers to the example namespace http://example.org/example#,
 - xs refers to the XML schema namespace http://www.w3.org/2001/XMLSchema#,
 - rdf refers to the RDF namespace http://www.w3.org/1999/02/22-rdf-syntax-ns#,
 - rdfs refers to the RDFS namespace http://www.w3.org/2000/01/rdf-schema#,
 - owl refers to the OWL namespace http://www.w3.org/2002/07/owl#, and
 - rif refers to the RIF namespace http://www.w3.org/2007/rif#.

Section 2. Again, suggest bullets for long lists (1st paragraph):
Where RDF/OWL has four kinds of constants:
 - URI references (i.e., IRIs),
 - plain literals without language tags,
 - plain literals with language tags, and
 - typed literals (i.e., Unicode sequences with datatype IRIs) [RDF-Concepts],
RIF has one kind of constant: Unicode sequences with symbol space IRIs
[RIF-DTB].

Near the end of Section 2, the term "rule body and query pattern"
occurs twice. These aren't defined. Maybe it should be "condition
formula" and link to BLD? Also, "rule heads" should be "rule
conclusions".

Section 3.

change: One consequence of the difference of the alphabets of RDF in RIF is
to: One consequence of the difference between the RDF alphabet and the
RIF alphabet is

3.1.3. I am a bit confused about whether rdf:XMLLiteral is
consistently used as a datatype or an IRI of a datatype. Normally I
wouldn't care :-), but this section is about mapping  IRIs to
datatypes.

Definition of datatype map conformance, point #2: delete 2nd "in"

Section 4.

The paragraph beginning "Since the OWL 2 Full Syntax" is hard to read
because of the use of "respectively". Suggestion:

Since the OWL 2 Full syntax is the same as the RDF syntax and the OWL
2 Full semantics is an extension of the RDF semantics, the definition
of RIF-OWL 2 Full compatibility is an extension of RIF-RDF
compatibility. However, defining RIF-OWL DL compatibility in the same
way would entail losing certain semantic properties of OWL 2 DL. One
of the main reasons for this is the difference in the way classes and
properties are interpreted in OWL 2 Full and OWL 2 DL.

In the Full variant, classes and properties are first interpreted as
objects in the domain of interpretation.  Class objects are associated
with instance objects using the rdf:type property and the extension
function IEXT, as in RDF. Property objects are associated with binary
relations over the domain using IEXT.

In the DL variant, classes and properties are not first mapped to the
domain. Classes are directly interpreted as subsets of the domain.
Properties are directly interpreted as binary relations over the
domain. This is a key property of the first-order logic nature of
Description Logic semantics and enables the use of Description Logic
reasoning techniques for processing OWL 2 DL descriptions.

4.1.1 - the Definition in this section defines 2 things.

4.1.2 - change "the RIF" to "RIF"

4.2.1 - the Definition in this section defines 4 things.  Suggestion:
use the BLD style of parenthesizing the term you are defining, and
define just one thing per Definition. E.g.

Definition (OWL Full-model). Let (I, I) be a ...

4.2.2.1 - "we use the symbol I to denote both the multi-structure and
the common part of the individual structures I1, ..., In"

I didn't notice a case where I is used to denote the common part.

"I(), for any other formula or symbol , " - other than what? (The
antecedent of "other" is probably too far back, before the definition)

4.2.2.2 - Definition of common-RIF-OWL DL-interpretation, in the conditions,
- Iframe' should be Iframe
- (true) should be deleted twice

Reword the Example to avoid the awkward "respectively":

Example. In OWL 2 DL, the domains for interpreting individuals
respectively, literals (data values), are disjoint.
=>
Example. In OWL 2 DL, the domain for interpreting individuals is
disjoint from the domain for interpreting literals (data values).

Also suggest these changes:

Consider the datatype xs:string and a RIF-OWL DL combination
consisting of the set containing only the OWL DL ontology
=>
Consider a RIF-OWL DL combination consisting of the set containing the
following OWL DL ontology:

Consider a RIF-OWL DL combination consisting of the set containing
only the OWL 2 DL ontology
=>
Consider a RIF-OWL DL combination consisting of the set containing the
following OWL 2 DL ontology:

9. 0-ary should be nullary. Predicate symbol rif:error must be written
as rif:error() when used in rules (as it is in several places).

5.

change two-ary to binary

7.

change ORACLE to Oracle

9.

change "for which RIF-OWL combinations that can be embedded." to
for which RIF-OWL combinations can be embedded.

Suggestion: use a different font for the translation function "tr" so
it stands out.

change 0-ary to nullary


9.1.3 -
It may be worth pointing out that RIF-BLD must be used for the
embedding because RIF-Core (and RIF-PRD) do not support ## and do not
support # in the head.

Theorem - C is a triple but a RIF-RDF combination is defined as a pair.

9.1.4 - R<super>RDF</super> references pred:isLiteralOfType

9.2.2.1 - Statement #29 - formatting problem with <tt> tag


-- 
Cheers,

Gary Hallmark
Received on Monday, 11 May 2009 19:34:58 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 2 June 2009 18:34:08 GMT