W3C home > Mailing lists > Public > w3c-rdfcore-wg@w3.org > October 2002

more comments Re: top-level Comment on lBase

From: Jeremy Carroll <jjc@hpl.hp.com>
Date: Fri, 25 Oct 2002 09:15:40 +0200
To: w3c-rdfcore-wg@w3.org
Message-id: <BHEGLCKMOHGLGNOKPGHDOEHHCAAA.jjc@hpl.hp.com>


A quick reply to one of pat's questions and then a list of review comments.

> ?? Can you say why XML Schema is nonmonotonic??

default attributes in the syntax

===

I mark my comments as "MUST" "PREFER" "EDIT" to indicate my level of
commitment to them.


Status
======

[[[
These include requirements that...

The RDF Core group must take into account the various formalizations of RDF
that have been proposed since the publication of the RDF Model and Syntax
Recommendation. The group is encouraged to make use both of formal
techniques and implementation-led test cases throughout their work.
The RDF schema system must provide an extensibility mechanism to allow
future work (for example on Web Ontology and logic-based Rule languages) to
provide richer facilities.

This document addresses these two requirements. It does not present an RDF
Core WG design for Semantic Web layering. Rather, it documents a technique
we are exploring to describe the semantics of the RDF Core specifications.
The RDF Core WG solicit feedback from other Working Groups and from the RDF
implementor community on the wider applicability of this technique.
]]]

MUST:
replace "This document address these two requirements." with
<<<
These two requirements  are primarily addressed by the RDF Model Theory
(ref) and the RDF Test Cases (ref).

This document describes an alternative complimentry approach.
>>>


PREFER:
(cf: Peter's comments)
add to Status
<<<
This is a first draft which does not include any treatment of the semantics
of any particular SWEL. A second draft will show how OWL can be layered on
OWL
within the lBase framework.
>>>

2.0 Outline
PREFER/MUST:
replace:
[[[
all currently proposed web langauges
]]]
with
<<<
most features of languages such as RDF, RDFS, DAML+OIL and OWL
>>>
(note owl:imports cannot be handled by lBase).

EDIT:
raison d'Ėtre
sp: the accent should be circumflex, personally I would use lower case e.

2.1 Names and variables

EDIT:
I found the level of detail on the syntax of strings and names surprising;
and somewhat lobsided. I know how to write whitespace in lBase but (without
any examples or a formal EBNF grammar) don't feel that I know how to write
any actual stuff.

EDIT:
I think you could omit all reference to XML structures with no loss.

EDIT:
I think a section clarifying precisely what can be done with special names
may be useful.
I would like
http://www.w3.org/2002/06/lbase/#special-name
to give me useful definitional text.

EDIT:
do special names correspond to dark triples?
do the authors believe that this functionality is useful or confusing?
I don't think the dark triple escapade contributed much to either WG.

2.2 Lbase Syntax

EDIT:
[[[
R(t1,...tn) where R is a relation name
]]]
? insert
<<<
, a special name
>>>

2.4 Axiom schemes
I like it.
(I am sorry that damns the rest with faint praise, but I guess I am being
pretty upfront about not supporting lbase).

3.0 Using Lbase

PREFER:
This section really could do with an extended example, but I can wait til
the next version.

4.0 Inadequacies

PREFER:
I get the impression that the current doc was shooting at where the WG was
in June.
Since then we have progressed with datatyping, and I think coming clean
about this in this section may be appropriate.

e.g.
<<<

This document was primarily based on

http://www.w3.org/TR/2002/WD-rdf-syntax-grammar-20020325/
http://www.w3.org/TR/2002/WD-rdf-mt-20020429/
http://www.w3.org/TR/2002/WD-rdf-schema-20020430/
http://www.w3.org/TR/2002/WD-rdf-testcases-20020429/

and early drafts of
http://www.w3.org/TR/2002/WD-rdf-concepts-20020829/
>>>

As these drafts are revised, this document will need to be updated. In
particular this document does not addequately address datatypes.


Sorry to be a party pooper.

Jeremy
Received on Friday, 25 October 2002 03:16:36 EDT

This archive was generated by hypermail pre-2.1.9 : Wednesday, 3 September 2003 09:52:30 EDT