- From: Stella Mitchell <cleo@us.ibm.com>
- Date: Sun, 24 Feb 2008 18:18:11 -0500
- To: public-rif-wg@w3.org
- Message-ID: <OFD8688D53.B26777C9-ON852573F9.000843C9-852573F9.00800218@us.ibm.com>
Abstract:
-------------
what about replacing "compatibility" with "interoperation" in the
text of
the abstract (or overview) - to give a sense of what is meant by
compatibility
in this context. (I'm not suggesting replacing it everywhere).
1.0 Overview
-----------------
The main motivating scenario for interchange of RIF-RDF combinations
is given as between two RDF-aware rule systems. In terms of RIF's
purpose, it seems that interchange between an RDF aware and a non
RDF aware system is just as typical/valid of a target? Aware-non
aware
interchange is mentioned later, but it seems like it's presented as
kind of a
special case.
1st para:
of (logical) rules --> of logical rules
4th para:
rule sets explicitly points --> rule set explicitly points
5th para:
it allows to use RDF data --> it supports the use of RDF data
7th para:
The consumers of rules retrieves the OWL -->
A consumer of rules retrieves the OWL
8th para:
With "interaction" --> By "interaction"
9th para:
4 --> four (twice)
2.0 RDF Compatibility
-----------------------------
uncle example, last word in para:
marry --> mary
last para:
Combinations are pairs of RIF rule sets and sets of RDF graphs
-->
A combination contains zero or one RIF rule set and any number
of
RDF graphs.
2.1.2
------
3rd para:
as long as the rdf:XMLLiteral datatype is considered -->
as long as the rdf:XMLLiteral datatype is included in the map.
the following datatypes to be considered -->
the following datatypes to be included
The list of BLD-required xsd datatypes doesn't match BLD's
current list
5th para:
notions of well-formed and ill-formed symbols in RIF -->
notions of well-formed and ill-formed terms and formulas in
RIF ?
6th para:
add an "or" at the end of condition 1
2.2
----
2nd para:
4 --> four
2.2.1
-------
1st para:
definitions satisfaction and entailment -->
definitions of satisfaction and entailment
2nd para:
between RIF semantic structures and RDF interpretations -->
between RIF semantic structures (interpretations) and RDF
interpretations
2.2.1.1
---------
3rd para (before 3rd bulleted list)
We restrict our attention here to DTS, D, Ic, Iv and Islot -->
We restrict our attention here to DTS, D, Ic, Iv, Iframe, Iisa
and Isub
bulleted list:
4th & 5th bullets are slightly out of sync with current BLD
where Iv maps
to Dind and Iframe maps to D X D -> D
add bullets for Iisa and Isub
2.2.1.2
---------
1st para (list of conditions):
condition 1
isn't it redundant because it would have to be true if
condition 3 is true?
condition 4
is a subset of IR and a superset of -->
is a superset of
(because the subset requirement is part of the
definition of RDF interpretation)
condition 5
the set of all pairs (a,b), with a,b in D such that
Islot(k)(a,b)=t for every k in D -->
the set of all pairs (a,b), with a, b, k in D such that
Iframe(k)(a,b)=t
2nd para:
Condition 2 ensures that the set of properties -->
Condition 2 ensures that the set of RDF properties -->
Finally, condition 7 ensures -->
Condition 7 ensures
Condition 8 the ensures -->
Condition 8 ensures
Condition 9 the ensures -->
Finally, condition 9 ensures
4th para:
This combination allows to derive -->
This combination allows the derivation of
2.2.2
------
1st para:
for all 4 entailment regimes -->
for all four entailment regimes
Talks about I satisfying R, but the term "satisfy" (-tion,
ability) is no longer
defined or used in the FLD and BLD documents. Same comment on
the 3rd paragraph in section 2.0.
3.0 OWL compatibility
-----------------------------
Since OWL-DL and OWL-Full have different semantics, why is it
considered undesirable that different RIF constructs, each with
the appropriate semantics, (the uncle example in 3.3.2) be used
to express rules under those two schemes?
3.2
----
1st para:
Since RDF graphs and OWL Full ontologies cannot be distinguished,
we
use the notion of RIF-RDF combinations for the syntax of
combinations
of RIF rule sets with OWL Full ontologies
-->
Since RDF graphs and OWL Full ontologies cannot be distinguished,
we use the same syntax for RIF-OWL Full combinations as for
RIF-RDF combinations
restrictions on the syntax of the rules -->
restrictions on the syntax of the RIF rules
2nd definition:
of a vocabulary V --> with a vocabulary V
3.3
----
2nd para:
cannot straightforwardly extends the -->
cannot straightforwardly extend the
3.3.2
-------
2nd para:
It is now the case that elementary class... --> Elementary class...
depending on whether they are used in OWL DL and OWL Full -->
depending on whether they are used in OWL DL or OWL Full
hasUncle(?x,?y) --> hasUncle(?x,?z)
?x[hasUncle -> ?y] --> ?x[hasUncle -> ?z]
3.3.2.2
---------
It would be good to briefly review OWL-DL interpretations here, like
it
is done for RDF interpretations in section 2.2.1.1
1st definition:
The symbol D used in 2 different ways here? (as the datatype map
and as
the domain of the RIF interpretation.)
condtions 3 & 4: there's no Ir mapping in a RIF interpretation
last definition:
every OWL DL model of C satisfies S -->
every OWL DL model of C satisfies O
3.3.2.3
---------
definition:
Ir isn't part of a RIFinterpretation
5.0 Appendix
-----------------
2nd para:
For the embedding we use the concrete syntax of RIF -->
For the embedding we use the presenation syntax of RIF
5.1
----
table:
2nd row has some formatting problems (<em><td>)
5.2
----
3rd para:
function sk takes as arguments a formula -->
function sk takes as an argument a formula
5.4
----
1st para:
The embeddings of RDF and RDFS entailment -->
The embeddings of RDF and RDFS graphs ?
5.5
----
table:
3rd row, 3rd column has a formatting problem (<tt>)
5.6
----
7th forall is missing "?" off a few variables
All sections
---------------
Since RIF is pronounced beginning with a consonant sound, "an RIF"
should "a RIF."
-Stella
Received on Sunday, 24 February 2008 23:18:36 UTC