W3CInteraction Domain

With financial support from:

W3C Workshop on
Metadata for
Content Adaptation

Chairs' Report: Workshop Executive Report

28 October 2004

The W3C/DIWG Workshop on Metadata for Content Adaptation took place on 12-13 October 2004, in Dublin, Ireland and was hosted by MobileAware Ltd. The 30 participants represented a diverse community of: mobile operators, content adaptation specialists, accessibility specialists, content rating providers, device manufacturers, telecommunications researchers and academics. The two-day workshop included two keynotes, 12 presentations, several debates and two breakout sessions. All of the presentations, minutes and summaries are available on the W3C web site [1].

The key findings of the workshop are as follows:


It is necessary for (fragments of) content to be labelled according to the role it plays, as intended by the author. Knowing "why" a particular piece of content is present will enable an adaptation solution to deal appropriately with that content. Roles can include concepts such as "navigation", "sidebar", "footnote", "explanation", "illustration", "main message" and many more. Some work must be done to develop a key/initial set of roles, and some guidelines on how adaptation should react to these roles.


The relationship between (fragments of) content should be captured in metadata. Relationships can include "explains", "illustrates", "must accompany", "alternative to" and many more. The "importance" of a fragment is a relationship between the fragment and its containing document. Relationships between metadata may also be required.

Inferred metadata

Where possible, metadata should be derived from the existing markup, in order to reduce the burden on the author. It should be possible for the author to add metadata but should only be necessary where there is a clear need and a tangible return on investment (of effort).

Cross-domain use-cases

The adaptation requirements of the mobile web, of Accessibility and of search/rating have technical similarities. Use-cases from automotive, mobile and accessibility domains clearly demonstrate the potential for sharing solution concepts across these communities.

Importance of RDF and OWL

The potential of RDF and OWL is clearly recognised. The participants applaud activities that make these technologies available to the diverse communities and in ways that fit with current and evolving practices. In almost all cases, the participants referred to RDF when considering technical solutions with respect to the representation of metadata.


The DIWG sees Role and Relationships as key concepts to be explored under its renewed charter. Technology to represent and process these concepts within an adaptation framework will be required. Liaison with other groups (particularly Semantic Web and others involved in OWL/RDF) will be essential. The shared ideas and objectives of other groups, such as WAI, shall be the basis of future joint discussions, which will be encouraged. We expect to learn from each other. Finally, the plea to avoid introducing unnecessary burden for authors shall be heeded.

[1] http://www.w3.org/2004/06/DI-MCA-WS/

This report has been approved by DIWG, to be issued and read in conjunction with other workshop results.

Metadata for Content Adaptation

Towards an accessible device independent Web

Workshop home page: http://www.w3.org/2004/06/DI-MCA-WS/

Rotan Hanrahan
Last updated $Date: 2004/10/28 21:13:49 $Author: rhanraha $

Valid XHTML 1.0!