Press Release

OWL 2: Knowledge and Data at Web Scale

OWL 2 Connects the Web of Knowledge with the Web of Data

Contact -- Ian Jacobs, <ij@w3.org>, +1.718.260.9447
(also available in French; see also translations in other languages )

Press Release | Testimonials | W3C Press Release Archive

http://www.w3.org/ -- ?? October 2009 -- Today W3C announces an important revision and extension of the Web Ontology Language (OWL). OWL 2 is a Semantic Web standard that provides a framework for asset management, enterprise integration, and the sharing and reuse of data on the Web. This standard format for information sharing spans application, enterprise, and community boundaries - all of these different types of "user" can share knowledge and data, even if they don't share the same software. As well as extended representational capabilities, OWL 2 provides several profiles that allow for simpler implementation and greatly improved scalability. OWL 2 therefore brings the Semantic Web closer to its full potential by connecting the web of knowledge with the web of data.

OWL 2 provides increased expressive power and improved scalability

OWL 2 provides a powerful language for defining ontologies -- vocabularies where the meaning of each term is precisely defined. They are used to classify the terms used in a particular application, characterize possible relationships, and define possible constraints on using those relationships.

An example may help. A bookseller may want to integrate data coming from different publishers. An ontology can be used to provide a unified and user-centric view over the different publishers' data, and to augment it with background knowledge. This could include, e.g., specifying that the terms "creator" and "writer" used in different datasets are equivalent to the ontology term "author", that each book has at least one author, that each author has exactly one date of birth, and that books having the same author(s) and title are in fact the same book. OWL 2 allows this kind of information to be precisely specified, and this precision enables generic processors to infer extra information from the data, e.g., that two ISBN numbers refer to the same book (because the books they reference have the same author and title), that a book is always an answer to a query for "document with an author" (even if the author isn't explicitly specified), or that there is an error in the data (if different data sources give different DOBs for the same author). OWL ontologies may define only a few simple terms, as in the above example, or may include 10s or even 100s of thousands of complex definitions.

W3C published the first version of OWL in 2004; this technology has already been successfully deployed in such diverse application areas as Oil & Gas exploration, eBusiness, health record management, semantic desktops, and management of musical archives (see, for example, some of the Semantic Web Use Cases and Case Studies published at W3C).

OWL 2 is an extension of OWL, adding some new features that have been requested by users. These include, for example, greatly extended support for datatypes (users can now characterise literal values as boolean, as binary data, as time instants, as different types of numerical values, and even as restrictions such as numeric intervals, or strings with a particular length), extended annotation capabilities (including annotations on terms, definitions and even annotations themselves), better ways of identifying resources (e.g., the above use of author and title to identify particular books).

An important feature of OWL 2 is the introduction of Profiles. The full implementation of OWL aware software may be costly, both in complexity and usage. The OWL 2 specification therefore includes a number of profiles, essentially sub-languages of OWL that are easier to implement and deploy, or that can be more efficiently implemented. Applications, vocabularies and ontologies may restrict themselves to these sub-languages in return for improved efficiency and ease of use.

Finally, in addition to the above mentioned feature enhancements, the specification itself has also been extensively revised and improved with respect to both clarity and precision.

To get started with OWL, see the OWL Overview and OWL Primer.