Web Ontology Working Group Request for Candidate Recommendation Status and Preliminary Implementation Report (original) (raw)
Please see OWL Implementations for a more up-to-date OWL implementation report. [This note added 2003-09-17, no other modifications since 2003-07-30]
Revision:1.43Revision: 1.43 Revision:1.43 of Date:2003/09/1703:02:29Date: 2003/09/17 03:02:29 Date:2003/09/1703:02:29
Dan Connolly and Jim Hendler
Dear W3C Director,
WhereAs
- you chartered us to develop an ontology language, with support of the W3C Membership
- we have elaborated on the value of this work to the community by way ofuse cases and derived design requirements
- we have developed specification, test materials, and supplementary documentation for OWL that meet our charter and requirements
- this specification and other materials has received wide review, within the Working Group and the community, and we have addressed the issues raised in this review with consensus on all but 2.
- we have implementation experience to validate the design
the Web Ontology Working group have decided (24 July) to request that you advance this specification to W3C Candidate Recommendation and call for implementation.
Titles, Abstracts, and Proposed Status
- Web Ontology Language (OWL): Overview
W3C Last Call Working Draft 31 March 2003. Deborah L. McGuinness and Frank van Harmelen eds.
formerly Feature Synopsis for OWL Lite and OWL- editor's draft,changes since last call
The OWL Web Ontology Language is designed for use by applications that need to process the content of information instead of just presenting information to humans. OWL facilitates greater machine interpretability of Web content than that supported by XML, RDF, and RDF Schema (RDF-S) by providing additional vocabulary along with a formal semantics. OWL has three increasingly-expressive sublanguages: OWL Lite, OWL DL, and OWL Full.
- editor's draft,changes since last call
- Web Ontology Language (OWL) Guide Version 1.0
W3C Last Call Working Draft 31 March 2003. Smith, Welty, McGuinness, eds. - OWL Web Ontology Language Reference
W3C Last Call Working Draft 31 March 2003, 12 November 2002. Dean, Schreiber, eds. - Web Ontology Language (OWL) Abstract Syntax and Semantics
W3C Last Call Working Draft 31 March 2003. Peter F. Patel-Schneider, Ian Horrocks, Patrick Hayes, Frank van Harmelen, eds. - Web Ontology Language (OWL) Test Cases
W3C Working Draft 31 March 2003. Jeremy J. Carroll, Jos De Roo, eds. - Web Ontology Language (OWL) Use Cases and Requirements
W3C Last Call Working Draft - 31 March 2003. Jeff Heflin, ed.
Status of This Document (proposed)
This section describes the status of this document at the time of its publication. Other documents may supersede this document. A list of current W3C publications and the latest revision of this technical report can be found in the W3C technical reports index at http://www.w3.org/TR/.
Publication as a Candidate Recommendation does not imply endorsement by the W3C Membership. This is a draft document and may be updated, replaced or obsoleted by other documents at any time. It is inappropriate to cite this document as other than "work in progress".
This 18 Aug 2003 (PENDING CONFIRMATION) draft, along with the other working drafts for OWL, are a Candidate Recommendation; it been widely reviewed and satisfies the Working Group's technical requirements; W3C publishes a Candidate Recommendation to gather implementation experience.
The first release of this document was 4 November 2002 (tune to each part) and the Web Ontology Working Group has made its best effort to address comments received since then, releasing several drafts and resolving alist of issuesmeanwhile. The design has stabilized, and once theexit criteria below are met, the Working Group intends to advance this specification to Proposed Recommendation.
Patent disclosures relevant to this specification may be found on the Working Group's public patent disclosure page.
Comments on this document should be sent topublic-webont-comments@w3.org, a mailing list with apublic archive. General discussion of related technology is welcome inwww-rdf-logic@w3.org(archive).
This document has been produced as part of the W3C Semantic Web Activity (Activity Statement) following the procedures set out for theW3C Process. The document has been written by theWeb Ontology Working Group. The goals of the Web Ontology working group are discussed in theWeb Ontology Working Group charter.
Summary of Review
Starting in November 2001, review followed the scope section of the Working Group charter:
The Working Group shall start by evaluating the technical solutions proposed in the DAML+OIL draft. If in this process the Working Group finds solutions that are agreed to be improvements over solutions suggested by DAML+OIL, those improved solutions should be used.
The first Working Drafts were released in July 2002 for review by the community, including groups with identified dependencies.
About 50 issues were identified by the Working Group during the review.
Dependencies were discharged as follows:
- the W3C RDF Core Working Group reviewed Refence, semantics, and test, and the WebOnt WG responded to those comments with clarifications and editorial changes; we await confirmation that this is satisfactory. (see also exit criteria below.)
- The DARPA DAML project endorsed the work (Murray A. Burke 8 May 2003) and stated "We view OWL as a major advancement for the Semantic Web, and are rapidly converting our ontologies, content and tools to use it."
- Among the Current OMG Specification materials is an March 2003 RFP which "seeks a specification of a MOF2 Metamodel, UML2 profile, and any additional information needed to support: * Development of ontologies using UML modeling tools * Implementation of ontologies in the W3C Web Ontology Language OWL.". This confirms that the plans outlined in a report on OMG relationship, 5 Aug 2002 by Evan Wallace continue to advance.
The Working Group responded to commentson a best-effort basis throughout the review, and formally addressed all of the last call-comments. We received 74 comments in all. Of these we have answered and been acknowledged on 63 messages. For 11 more we have not received acknowledgements yet - on 9 of these we agreed or mostly agreed with the comment raiser and made edits to our documents in response. One comment has not been answered to the commentor's satisfaction, and we address this below. See disposition and summary of comments and OWL-generated view for a detailed list.
We achieved consensus on all but two issues:
5.6 daml:imports as magic syntax
This issue, regarding how the imports feature of DAML+OIL fits with the rest of OWL, was raised May 2002 and discussed at length before being resolved in November 2002. Dissenters Hendler/MIND, Connolly/W3C and Welty/IBM argued that the WG should postpone the issue, but the proposal by Hefflin to specify a sufficient level of detail for how imports works gained considerably more support, and no other proposals have become available. Last call comments on this feature were addressed to the satisfaction of the commentors with minor editorial changes to the specifications. A Feb 2003 report from the Jena/HP development team reports their satisfaction with supporting the feature.
Since this design seems to be specified to the satisfaction of a critical mass of the community; we ask that The Director confirm the WG decision despite the outstanding dissent.
Beckett/ILRT commented on section 4.1 Translation to RDF Graphs in the OWL Semantic and Abstract Syntax, arguing that "Providing half of a complex mapping between an abstract syntax to/from a concrete syntax seems insufficient to me." The working group discussed mappings that went both ways (e.g. Carroll 21 Jan) as these two-way mappings involved substantive as well as editorial changes, did not work out the remaining design details before the 31 March Last Call.
The Working Group did accept an exit criterion to verify that the design is implementable by exhibiting two OWL syntax checkers. In light of this criterion and Beckett's comment that "Maybe you do have multiple interoperable implementations of the mapping from OWL's concrete syntax (RDF triples) to OWL's abstract syntax and I am just unaware of them. If that is the case, then I would be more satisfied." We ask the Director to confirm the WG decision despite this dissent.
Feature at risk
Per Section 7.4.3 of the Process Document, the Web Ontology Working Group would like to declare that one feature of our current technical reports are "at risk." In particular, the semantics of OWL DL appear to be simpler to prove if certain restrictions are made on the use of bnodes in the RDF graphs that are allowed in OWL DL. However, the restrictions may be difficult to implement. The Working Group has therefore decidedto place these restrictions on bnodes "at risk" and call for implementation experience to determine the implementability of these restrictions. The details of the document changes below present the details of effected tests and documents.
Implementation Experience
The following implementation experience leads us to believe that once theexit criteria below are met, we will have sufficient implementation experience to validate the design and merit widespread deployment.
- Demos/Portals
These demonstrate deployment related to identified use cases and requirements:- The DARPA DAML project developed the Horus[PDF] project for the intelligence community.
- The AKT Portal at the University of Southampton is largely based on ontologies, and is now using OWL.
- The University of Maryland Baltimore County (Finin) has developed two demos using OWL (Finin 7May)
- BioPax - A Data Exchange Format for Biological Pathways has been using OWL
- The W3C tech reports - related to the "multimedia collections" use case
- ( The MINDSWAP project web site) uses OWL to generate web pages and "custom home pages" for members of the research group. (Hendler 6May)
- The OWL-generated Comment Status List submitted for this CR request was generated by a combination of OWL tools.
- Reasoners
- FaCT complete OWL Lite (bechofer) and nearly complete OWL DL WonderWeb project, Horrocks 12 May.
- Cerebra from Network Inference - nearly complete OWL DL White 15 Jul
- cwm -- useful but incomplete OWL Full
- Euler -- useful but incomplete OWL Full,see De Roo 11 Jul: 51 / 234 tests
- Jena/HP (Reynolds/HP 7 May)will support OWL reasoning.
- Vampire Horrocks 17 Jul - uses a first-order theorem prover to do OWL DL
- PELLET
is a OWL Lite reasoner in Java. - OWLLisaKB a rule-based OWL Lite reasoner written in LISP (using Wilbur)
- SWI-Prolog Semantic Web Library contains owl.pl - an OWL reasoning package.
- F-OWL is an f-logic based Owl tool from UMBC.
- E-wallet is an e-commerce and mobile computing tool based on a rule-based OWL reasoner.
- Parser/validators
- owl species validator/parser, bechofer/volz (almost conforming)
- Jena: almost conforming ( Carroll/HP 6 May )
- BBN OWL Validator
- IC Instance Creator, ConvertToRDF, and SMORE use an OWL parser developed at the University of Maryland MIND Laboratory.
- Editors
- The Protege has developed an OWL plugin.
- Construct from Network Inference Complete OWL-DL (White 15Jul)
- SMORE is a markup tool that produces OWL documents
- cwm .n3 -> .rdf produces conforming OWL Full documents
- Ontologies
- Over 200 DAML ontologies can be converted into OWL using OwlConverter - most of these are in OWL Full
- The NCI cancer ontology,(Paper - pdf link) is an OWL Lite ontology with over 17,000 classes in ontology with close to 500,000 triples.
- UMBC: several ontologies (Finin 7May)
- API
- OWL API (bechover/volz 7 May)
- Jena (Reynolds/HP 7 May)
- Cerebra from network inference
Candidate Recommendation Exit Criteria
- finish resolving dependency on RDF Core specs, esp. RDF Semantics
- two complete OWL Lite consistency checkers (i.e. 2 which pass almost all OWL Lite consistency and inconsistency tests and moreover claim logical completeness)
- Each test (except the extra credit tests) is demonstrated to be passed by some implementation
- two reasoners implementing (different) substantial subsets of OWL DL
- two reasoners implementing useful subsets of OWL Full and passing at least 80% of the postive entailment tests
- two owl syntax checkers passing all tests
Details of the feature at risk
The following tests would change from being OWL Full (in)consistency tests of OWL Full files, to being OWL Lite and OWL Full (in)consistency tests of OWL Lite files:
- http://www.w3.org/2002/03owlt/editors-draft/draft/proposedByFunction#Restriction-002
- http://www.w3.org/2002/03owlt/editors-draft/draft/proposedByFunction#Restriction-003
The following tests would change from being OWL Full (in)consistency tests of OWL Full files, to being OWL DL and OWL Full (in)consistency tests of OWL DL files:
- http://www.w3.org/2002/03owlt/editors-draft/draft/proposedByFunction#disjointWith-004
- http://www.w3.org/2002/03owlt/editors-draft/draft/proposedByFunction#disjointWith-006
- http://www.w3.org/2002/03owlt/editors-draft/draft/proposedByFunction#disjointWith-008
- http://www.w3.org/2002/03owlt/editors-draft/draft/proposedByIssue#I5.26-001
- http://www.w3.org/2002/03owlt/editors-draft/draft/proposedByIssue#I5.26-002
- http://www.w3.org/2002/03owlt/editors-draft/draft/proposedByIssue#I5.26-003
- http://www.w3.org/2002/03owlt/editors-draft/draft/proposedByIssue#I5.26-004
- http://www.w3.org/2002/03owlt/editors-draft/draft/proposedByIssue#I5.26-005
The following similar tests would be unchanged (in OWL Lite or OWL DL):
- http://www.w3.org/2002/03owlt/editors-draft/draft/proposedByFunction#Restriction-001
- http://www.w3.org/2002/03owlt/editors-draft/draft/proposedByFunction#Restriction-004
- http://www.w3.org/2002/03owlt/editors-draft/draft/proposedByFunction#equivalentClass-009
- http://www.w3.org/2002/03owlt/editors-draft/draft/proposedByFunction#disjointWith-003
- http://www.w3.org/2002/03owlt/editors-draft/draft/proposedByFunction#disjointWith-005
- http://www.w3.org/2002/03owlt/editors-draft/draft/proposedByFunction#disjointWith-007
- http://www.w3.org/2002/03owlt/editors-draft/draft/proposedByFunction#disjointWith-009
The following similar tests would be unchanged (in OWL Full):