Report on W3C Technical Plenary Session
Gandon (INRIA, Sophia Antipolis, ACACIA Lab.)
Semantic Web Interest Group
Session #1: Eric Miller
summarized the work in semantic web the past 4 years.
Eric Prud'hommeaux gives an update on the activity of DAWG (Data Access
Working Group). In particular, the publications so far:
- SPARQL Query;
- SPARQL variable binding in XML,
- SPARQL Protocol
The work on nested queries was deferred i.e. the use of the result of a
query to create a new query.
For more, see http://www.w3.org/2005/Talks/28-DAWG-SWIG/
Chris Bizer talks of named graphs and gives an update of its use in
SPARQL (with Carroll, Hayes, Stickler). They use the keyword "WITH"
used to give trusted sources. Also, trust weighting and propagation
over the sources.
Use of SWP (?) for signing RDF graphs.
Two syntaxes are mentioned to represent named graphs (TriX, in XML and
TriG in Turtle)
NG4J is mentioned as an implementation for Jena.
Ralph raised a question: Timbl has log:includes in N3. What's the
relation between that and named graphs?
Jeremy Carroll, SW Best practices work on Datatyping
see SWBG session on that.
Bijan Parsia presents the Filmtrust application: trust network, rating
and recommendation of films see : http://trust.mindswap.org/
Second presentation is about Swoop: web browser inspired ontology
A hypermedia-based Featherweight OWL Ontology Editor
Also shows the use of Pellet to validate schemas e.g. in tanbis
ontology, 144 unsatisfiable classes
Third presentation: Econnection: automatic partitioning of Web
ontologies to address problems of ontologies where things are
accumulated as needed and redundant with existing ontology or which
small parts could be reused somewhere else if these ontologies were
modular. Example of ontologies used for testing: GALEN, NCI, etc.
Session #6: Alistair Miles on
Semantic Web and Thesauri
Recall: SKOS( initially Simple Knowledge Organisation Systems) is an
open collaboration developing specifications and standards to support
the use of knowledge organization systems (KOS) on the semantic web.
Expressing concept schemes (thesauri, classification schemes,
controlled vocabularies, structured vocabularies, terminologies,
glossaries, taxonomies ...) in RDF
SKOS interacts with the task force "Porting Thesauri" in the working
Group Semantic Web Best practices.
- SKOS Core Vocabulary Specification, W3C Working Draft in Preparation
- SKOS Core Guide, W3C Editor's Working Draft 15 February 2005
- SWBP Quick Guide to Publishing a Thesaurus on the Semantic Web:
This guide of the Semantic Web Best Practices Working Group makes use
of the RDFS schema of SKOS.
Notes on the talk: http://isegserv.itd.rl.ac.uk/skos/pres/swig-talk-2005.txt
Session #7: Amit Sheth
(Semagix) presents projects Real World applications from his group in
anti-money-laundering system and law enforcement systems.
Online demo: http://www.semagix.com/downloads/product_demo.htm
Session #8: Frank Manola (SW
Business): talks about business applications of the semantic web :"How
do I recognize a SW business app if I meet one?". Underlines that there
is not enough visibility of existing work from industry.
The talk makes a reference to a message provocatively entitled "non
semantic-web uses of RDF / S ";-)
Session #9: Charles
McCathie-Nevile talks about an application of Semantic Web to Human
Session #10: Jeremy Caroll
gives an update on RDF/A, a collection of attributes for layering RDF
on XML languages, as described in the note by Mark Birbeck and Steven
See also SWBP session on this subject.
Session #11: Evan Wallace gives
an update on the "Ontology Definition Metamodel" issued by OMG and
reviewed by SWBP. This document is intended to present foundational
formalisms for ontology-based Model Driven Architecture (MDA) in
software engineering. It is concern with the different representation
formalisms, their interoperability, and their contribution to the
automation of business semantics. One of the objectives stated by the
introduction of this document is to provide the basis for
specifications to marry MDA and Semantic Web technologies "to support
semantic web services, ontology and policy-based agent communications,
and policy-based applications in general".
See Revised submission for the OMG Ontology Definition Metamodel (Jan.
And "Representing UML in RDF": some early previous work on UML in RDF
from Sergey Melnik
Session #12: SIMILE Project by
Ryan Lee, Stefano Mazzocchi, David Huynh.
Semantic Interoperability of Metadata and Information in unLike
SIMILE is a joint project conducted by the W3C, HP, MIT Libraries, and
MIT CSAIL. SIMILE seeks to enhance inter-operability among digital
assets, schemata / vocabularies / ontologies, metadata, and services.
Welkin is a graph-based RDF visualizer http://simile.mit.edu/welkin/index.html
Longwell is a suite of web-based RDF browsers http://simile.mit.edu/longwell/index.html
Piggy-Bank, a Firefox extension to let users collect and browse
"semantic data" linked from ordinary web pages http://simile.mit.edu/piggy-bank/index.html
Session #13: RDF-SIG:
Leveraging XML-SIG and XKMS for digitally signing RDF Graphs
by Jose Kahan
Growing interest in signing RDF graphs but while some work is new,
other is duplicating things done previously by XML-SIG. We need to
leverage existing technologies.
If we sign the XML serialization, we need an algorithm for
canonicalization of the graph serialization.
Session #14: Phil Archer on
quality labels: quality and content description in RDF by Internet
Content Rating Association. He proposes rules with URL patterns to
annotate the quality and content of all the resources at a web site.
RDF Content Labels: Use Cases and Draft Schemas:
Envisage centralized authorities for labeling and distributed labeling.
Session #15: Fumihiro KATO
(Keio) Trust with labeling
Session #16: Dan Applequist -
vodafone perspective on using RDF for content metadata in mobile
applications. Try to achieve device neutral format of content.
Session #17: Bijan Parsia on
Web Service policy checked using the pellet engine.
Session #18: Andreas Harth from
DERI: Linking Semantically-Enabled Online Community Sites - SIOC
Use RDF to link between different community exchange technologies
(email, RSS, etc.).
Query result in RDF to get the info wherever it comes from.
Common data format (RDF), Common ontology (SIOC) or mappings to SIOC,
Common protocol for data exchange (SPARQL), Links/metadata to relate
posts, forums, and sites, Object consolidation/smushing.
Session #19: Marja - Annotea
and Firefox update. Annotea Ubimarks in Mozilla
Inspirited by study of physical bookmarks, tabs in book / scribbles,
Second phase: shared bookmarks and topics
Shows an annotea-extended bookmark view pane in firefox running live
Mapping between SKOS and W3C Bookmarks vocabs
Session #20: Tom Croucher -
compound IFP discussion (Inverse Functional Property)
Problem of a set of Functional Property providing an identity criterion
for a resource.
Especially to uniquely identify things that don't have URIs also called
" unambiguous property constellations", " composite keys in database"
(// Identity Conditions)
Paper on the subject : "Situation and Identity A Generalisation of
Inverse Functional Properties" by Tom Croucher and Joe Geldart
Problem of open-world assumption for designing "unambiguous property
Session #21: Amit Sheth
presents METEOR-S and SWETO.
METEOR-S is a platform for web service development, discovery,
invocation and composition.
SWETO Semantic Web Technology Evaluation Ontology : Large-Scale
Semantic Web Test-bed
Session #22: Sandro - Rule
W3C Workshop on Rule Languages for Interoperability
Workshop on rule language for the web (and not only the semantic web)
Ontaria = searchable browsable directory of semantic web data.
Session #23: Patrick Stickler,
Forum Nokia Meets the Semantic Web
Nokia forum for developers of Nokia applications using RDF metadata for
menus, summaries, etc.
Session #24: Fabien Gandon,
KmP: A Semantic Web Service for the Cartography of Competences in the
Telecom Valley of Sophia Antipolis and Update on CORESE (conceptual
graph based search engine) and its query language
Plenary Session Day
Introduction: Leading the Web to its Full Potential
Steve Bratt, Chief Operating Officer W3C
Panel on extension and versioning
XML and Binary content.
Where XML is Going, and Where it Should (or Shouldn't) Go
Issues in Developing Multiple Specification Test Suites
User Agent Perspective: http://cita.rehab.uiuc.edu/presentations/2005-03-02-w3c/slide1.html
WAI : http://www.w3.org/2005/Talks/0302slh/
Lihgting talks, Multi-modality demos,
Semantic Web Best Practices Working Group
Session #1: Software Engineering Task
Ontology Driven Architectures (ODA) and Potential Uses of the Semantic
Web in Software Engineering
Automated consistency checking get better quality of software,
maintenance costs reduced because of tie between model & software.
It is expected to see more focus on ODA application of RDF/OWL.
It would be surprising if some form of ontological approach wasn't used
in case tools or case research or other form of MDA in the past.
Discussion on the usefulness of the document: some members think the
current version is just a collection of hype statements; some thinks it
is just an early draft that could be turned into something with more
content and that it is just too early now for a review.
Issue on who really contributed.
One idea is that a good note could be a document that focuses on
ontological additions to software engineering practise, building on 30
years of research; a focussed note on how some of previous approaches
in software engineering could benefit from a Semantic Web approach.
Session #2: Vocabulary Management Task
The goal is to produce a short document (8-12 pages) reviewing current
practices for managing vocabularies. Draft of TF is available here:
Link with TAG issues:
What should a "namespace document" look like?
(see also http://www.textuality.com/tag/Issue8.html
What is the range of the HTTP dereference function?
Syntax and semantics for embedding RDF in XHTML
Meaning of URIs in RDF documents
Session #3: Application and Demo Task
Question raised: "What is a good application as convincing application
for semantic web?"
Idea to movie to DOAP descriptions:
Use an online DOAP-a-matic form to produce descriptions for projects.
A better workflow would be that people use DOAP to describe their
projects themselves and the TF links these descriptions. Then have
facetted servers to present these descriptions.
What should be the criteria for inclusion?
RSS as an application of RDF?
Article on "The Role of RSS in Science Publishing"
Session #4: Embedding RDF in HTML
NB; the task force has a separate list
Former charter (2003):
Grouped session with the XHTML working group for XHTML 2.0
Just a reminder for HTML nothing to embed RDF, for XHTML 1.0 one can
use GRDDL, for
XHTML 2.0 one can use RDF/A.
GRDDL (Gleaning Resource Descriptions from Dialects of Languages) a
mechanism for encoding RDF statements in XHTML and XML to be extracted
by programs such as XSLT transformations. They are considering moving
Section in XHTML2.0 W3C Working Draft on Metainformation Attributes
Discussion on real scenario that would use RDF in HTML, I proposed the
case of e-learning:
Session #5: Ontology Engineering and
Patterns Task Force (OEP)
The draft "Representing Specified Values in OWL: 'value partitions' and
'value sets' " is moving to Note:
The draft "Representing Classes As Property Values on the Semantic Web
is moving to Note:
New draft is proposed on "Units and Measures"
Alan to take over the Qualifying Cardinality Restrictions Note from Guus
Session #6: WordNet Task Force
Session #7: RDF and XML Schema
To reference XSCD in OWL, a possible solution is to have an ID in the
Datatype definition to reference the type in RDF.
Also, the idea of using a mime type to type fragIDs as type not
Session #8: Quick intervention
of Bijan Parsia on the way to include RDF descriptions in WSDL:
compares a straight way and a complex way.
Session #9: RDF/Topic Maps
Goal of RDFTM is to provide Guidelines for combining RDF/OWL and Topic
RDFTM: Survey of Interoperability Proposals
Test Cases for RDF/TM Interoperability
Key problem is that TM and RDF have different levels of semantics any
triple could map to 6 different things in a TM.
Discussion on naturalness ("faithfullness" to the paradigm) vs.
fidelity (correctness of the translation)
Is it the case that information expressed naturally in the source
formalism (Topic Maps) and then translated into the target formalism
(RDF) can be naturally integrated with information expressed naturally
in the target formalism (RDF).
Session #10B: Applications and Demos
- How to provide and maintain a list of Applications and Demos?
- Should we go for a DOAP format?
- Criteria for including/excluding a demo in the list?
Use DOAP and leave it to the users to maintain it: we would only
maintain a list of small descriptions (a pointer and a sentence) and
with the pointer can access the DOAP file and harvest.
First priority = getting a number of them in the list.
Summarizing : we stay with web log, we try to set up some tools for
task force administration tasks (accept a description), we provide
support for DOAP and we maintain a simple list of DOAP files in the
Issue with including TM applications: to be included in the repository,
an application should at least import and export RDF(S) / OWL data.
List of projects is available at http://www.aktor.org/technologies
for this working group:
OEP: Defining N-ary Relations on the Semantic Web: Use With
OEP: Classes as values
OEP: Value partitions
OEP: Part-whole relations
PORT: Core Guide
PORT: Core Vocaab Spec
PORT: Quick Guide
XSD: XML Schema Datatypes in RDF/OWL
WNET: WordNet data model:
WNET: ISLE lexical entries
SE: ODA draft