This interview was given by Dr. Dave Raggett for the Loquendo Newsletter and for the VoiceXML Italia User Group. The interview took place at the beginning of March 2006, at Cannes during the W3C Technical Plenary Meeting.

[Paolo Baggia] Dr. Dave Raggett, would you like to say a few words about yourself?

[Dr. Dave Raggett] I am an employee of the UK company "Volantis" who specialize in content adaptation for Web applications on mobile devices. Most of my time is spent on assignment to the World Wide Web Consortium as a member of the W3C Staff. My current role is Activity Lead for W3C's work on multimodal interaction.

[P. B.] How did you come to find yourself taking part in the early days of the web? What did you do at that time?

[Dr. Dave Raggett] I have been involved in the Web since 1992 when I first met Tim Berners-Lee and started working with him on HTTP and HTML.
At that time, I was a researcher at HP's corporate research laboratories in Bristol, England. I had been working on applying hypermedia and expert systems to new ways to order HP's computer workstation solutions.

Trying to squeeze the sales manuals onto VGA sized displays proved to be a considerable challenge, and provided me with a great experience in designing graphical form-based applications. After meeting with Berners-Lee, I was keen to bring that experience to enriching HTML, and set about collecting ideas for what became known as HTML+.

Standards work on HTML moved from the IETF to the W3C in 1995 when I was on a two year assignment to MIT. Work on HTTP remained at the IETF. I had launched the HTTP working group after a Bird of a Feather meeting in late 1994. Larry Masinter (Xerox PARC) took over as chair, and I was free to focus on W3C work on HTML 3.2 and subsequently HTML 4.0.

Since then I have been involved in launching standards work in a number of areas at W3C, including XHTML, XForms, MathML, Voice Browsers and Multimodal Interaction.

[Dr. Dave Raggett] My interest in voice interfaces started in the mid-nineties as a way to provide greater accessibility for people with visual impairments. I organized a W3C workshop on Voice Browsers in 1998 and then set up the W3C Voice Browser working group early in 1999.

This proved to be a popular working group and much larger than most at W3C. We took on the task of refining VoiceXML which itself was inspired by HTML with links and forms, rather than conventional approaches to dialogue modeling. The charter also allowed for work on multimodal interaction and we eventually published a draft set of requirements, but this was put to one side as we pushed hard on completing VoiceXML and its related specifications on speech grammars and speech synthesis.

To satisfy interest in multimodal interaction, I organized a workshop that was co-hosted by W3C and the WAP Forum at hotel in Hong Kong in 2000. Despite the interest, it wasn't clear whether it was timely to start standards work. The Multimodal Interaction working group was eventually formed two years later in 2002.

[Dr. Dave Raggett] VoiceXML has been extremely successful in displacing other approaches for self-service call center solutions. The use of XML and web-server application technologies has helped to reduce the costs in developing and maintaining voice applications. Work on multimodal interaction has moved much more slowly, and I think it will be several years yet before we see multimodal really take off. Giving people a choice in how they interact with applications will be really valuable.
Voice is much easier than typing on mobile phone keypads, although sometimes the context makes typing the right choice.
Similarly, the persistent nature of displays offers benefits compared with voice which is by its very nature transitory.

[Dr. Dave Raggett] I am currently focusing on a synthesis of the Web and ubiquitous computing. The aim is to make it much easier to build distributed applications across a diversity of devices and in a variety of environments including home networks, offices, factories and many more. It includes extending the Web into the physical world of sensors and effectors, for example RFID tags.

[Dr. Dave Raggett] Essentially it means applying a web model to developing distributed applications, i.e. mark up and the DOM for scripting event handlers. The difference is that now events may come from remote sources. Device coordination involves the discovery of capabilities and services, and the binding of these into the application session.
Discovery can involve Web services in a distributed search, and this is where the W3C's work on the Semantic Web can pay dividends as a scalable way to combine distributed information sources.

I recently ran a W3C workshop on the Ubiquitous Web at Keio University in Tokyo. Participants shared use cases, discussed the relation to existing W3C work, and areas where further standards work would be needed. One of the major topics at the workshop was trust and security. There is plenty to think about, but I hope to come up with detailed proposals for new standards work later this year. I am also starting to plan for a follow on workshop on declarative web application modelling as a basis for content adaptation and for distributed applications.

[Dr. Dave Raggett] Largely yes. For a long while, the Web was for many people a read-only medium, but over the last few years wiki's and blogs have reversed this, and I expect the trend to continue as more and more applications migrate from standalone to web-based applications.

[Dr. Dave Raggett] I think the Web is still in its infancy and there is much left to do. There is a great deal of excitement and energy and I am very much looking forward to realizing the vision of the Ubiquitous Web.

[Dr. Dave Raggett] My main hobby is scuba diving, and I am working towards being a fully qualified open water instructor. The underwater world is very relaxing and it is wonderful to help introduce newcomers to the experience.

I also enjoy working on open source software projects, for instance, HTML Tidy, a tool for cleaning up HTML, and more recently Slidy (so named by my daughter) which provides a Web-based alternative to Microsoft PowerPoint. It has also been fun working with Loquendo on building demonstrators for adding speech interfaces to Web pages.

[P. B.] Many thanks, Dave Raggett.

Dave Raggett is a W3C Fellow from Volantis and has been closely involved in the development of standards for the Web since 1992, especially HTML, and has launched standards work on HTTP, MathML, XForms, voice browsers and multimodal interaction.

He is currently focusing on realizing the potential of the Web for distributed applications across a diverse range of devices (the Ubiquitous Web).
He is the author of several books as well as many conference papers and talks (one of his most recent talks was at Google, Montain View CA, February 1st, 2006).

You can find out more about his work on his home page, see: http://www.w3.org/People/Raggett