Volantisphoto Dave Raggett's Bio

I am employed by Volantis who have generously placed me on assignment as a member of W3C staff, reporting to Philipp Hoschka.

I am currently the W3C Actvity Lead for Multimodal Interaction, and have been trying to enourage people to do enough standards work to catalyse the widespread adoption of multimodal interfaces. This is going slower than all of us anticipated, but the long term prospects for richer ways to interact with devices and services are very bright.

I have been developing a way to add speech and VoIP capabilities to modern Web browsers via a local HTTP-based speech proxy, and exploiting XmlHttpRequest (AJAX). This uses the Loquendo libraries for speech synthesis and recognition for the speech server, but end users only need to install a small portable proxy server. The decision to implement a proxy server came after experience with implementing a binary Firefox extension for speech synthesis which I demoed at the W3C Technical Plenary in early 2005. It turned out to be very challenging to add support for speech recognition, and would in any case only work with Firefox, while the proxy server would work with any modern web browser.

In the spirit of weaning people away from Microsoft Office, and towards Web-based formats, I have developed a cross browser slide show tool (HTML Slidy) and am working on adding wysiwyg editing capabilities, building on top of the "design mode" capabilities in most modern browsers.

There are huge opportunities for Web based services for small to medium sized enterprises, helping people to organize their information and building upon knowledge of relationships between people to provide smarter services. Today's standalone desktop applications will soon look very antiquated.

I have helped to drive W3C's work on context awareness for web applications, initially in the Multimodal Interaction working group and more recently in the Device Independence working group. The Delivery Context Interface provides a means for web applications to dynamically adapt to the device capabilities, user preferences and environmental conditions (e.g. geographical location, battery level and RF signal strength) and has been developed together with France Telecom, IBM and Nokia.

I have also worked with Nokia on launching work on remote user interfaces at the IETF, presenting at the BoF in Paris a few months ago, helping with the charter, and more recently co-editing the Internet Draft for the requirements. The aim is to define a protocol for use with XML-based user interfaces, where the user interface and application logic may be on different machines, and coupled via an exchange of XML DOM events and update/mutation operations. The Widex working group has recently been approved by the IESG and the first meeting is expected to occur at the 64th IETF in Vancouver in November 2005.

My current vision building exercise is the "Ubiquitous Web". I am organizing a W3C workshop on this for March 2006 (Of course my participation in the actual workshop will rely on finding new sources of funding). The Ubiquitous Web is about giving Web developers new power to create distributed applications that break free of the traditional client/server model. See my presentation at the W3C Seminar on embedded multimodal interfaces.

Standards are a powerful tool to build prosperity, but need to be combined with proactive work on building visions and demonstrating them through solid implementations.


My involvement with the Web started in 1992: developing experimental web browsers and servers, authoring the HTML+ and HTML 3.0 specs for the IETF, the Arena browser, coining and helping to launch VRML, setting up the IETF HTTP WG (which I chaired in the early days before handing over to Larry Masinter), and setting up the W3C HTML-ERB which has by now evolved into a number of working groups, including HTML, CSS, DOM, Math, and XForms.

In the Fall of '95 I brought the major players in the browser wars together around the same table to work on joint specifications for HTML. We initially worked on the OBJECT tag and later on, style sheets, improvements to forms, etc. The work on style sheets was spun off into the CSS working group. The work on HTML has led to HTML 3.2 and more recently HTML 4.0. I organized a W3C workshop on the Future of HTML in 1998 and then created a briefing package which was accepted by W3C members. The new HTML working group (members only) came into operation in early September 1998 and has subsequently driven the transformation of HTML into the XML markup language, now known as XHTML.

In 1996 I organized a workshop on High Quality Printing and the Web. This led to work on richer style sheets, and better ways to specify fonts including support for downloadable font subsets. The ideas for improving the faithfulness of color rendition over the Web have yet to be fully adopted. This is something that HP feels is very important to the future of electronic commerce -- for instance if you select a pullover from a catalog on the Web that matches the color of a skirt you own, you would be pretty unhappy if when it appeared in the post, it had a different color!

The work on richer style sheets has led to a range of new features in the 2nd release of CSS from W3C. Though my hopes for using CSS for flexible layout have yet to be fully realized. More recently I have worked on ideas for new ways to specify presentation and behaviour for XML and HTML based upon declarative style rules and object oriented scripting. The result called Spice blends CSS and an extended form of JavaScript. I have been closely involved in work in ECMA to develop revisions to the ECMAScript standard, inspired by my work on Spice.

A spin off from the work on Spice is a tool for cleaning up bad HTML. The HTML Tidy utility, which is available as open source from W3C. It provides detailed feedback on errors and can help you spot areas where you can improve the accessibility of your HTML documents for people with disabilities. With tidy you can fix up a whole directory of HTML files with one command! I have also worked hard in the HTML 4.0 spec to ensure that you can write accessible HTML. This has focussed on labels for form fields, a way to group form fields and ways to support effective browsing of tables using speech or Braille. I greatly enjoyed working on accessibility especially my visit to the White House, a few years ago.

In 1993 I proposed ideas for embedding mathematical expressions in Web pages as part of my html+ proposal. This evolved into the HTML 3.0 proposal (see [1]). In 1996 I set up a W3C Math working group that grew to include the major players in this area. The group has developed the Mathematical Markup Language which will very soon become a W3C Recommendation. MathML is a verbose low-level format for machine to machine communication. I have developed a higher level language inspired by how people speak when reading mathematical expressions e.g. to someone over the phone. EzMath is available as a free editor and plugin and was presented along with Spice at the WWW'7 conference in Brisbane.You can use EzMath as a simple means to create MathML or directly in its own right.

I am very interested in speech and the Web. Nicholas Negroponte head of MIT's Media Lab made the case for this very well a few years ago when he pointed out that as computers shrink they become too small for keyboards or screens. Speech recognition and synthesis become the only practical means to interact with them. I am now involved with work on ways to exploit robust speech recognition and high quality speech synthesis that is actually bearable to listen to. I expect W3C to play an important role in helping to make the Web accessible from a broad range of devices: cellular phones, car based systems, handheld and palmtop computers, Web television sets and regular unmodified phones.

HTML is often used as a core component in Web based applications. To improve its effectiveness, I helped to instigate work on separating the user interface from the application data and logic. This work started as a next generation forms subgroup within the HTML working group, but has now been spun off as a separate XForms working group, and now a Candidate Recommendation.

I ran a W3C workshop on voice browsers in October 1998, bringing together researchers and developers to advise W3C on next steps. Following the workshop, I drafted a charter and went on to sent up a W3C working group for voice browsers. This has developed a suite of standards including VoiceXML 2.0. In September 2000, I organized a joint W3C/WAP Forum workshop on multimodal interaction, and in 2002, set up the W3C Multimodal Interaction working group. Both of these working groups have attracted broad participation from industry and look set to transform the nature of the way we interact with the Web. In the future, I look forward to an increasing focus on natural language and the fuzzier aspects of human reasoning.

I am married with two teenage children and live in Wiltshire in the west of England, following my return from two years of living in Boston. I work for W3C on assignment from Canon, although I spend most days telecommuting to MIT from home, or on the road clocking up yet more air-miles.

Previous Work Experience

Here is my curriculum vitae/résumé.

From February 2003 to February 2006, I was employed as a consultant by Canon with my time divided between my work for W3C and consultancy for Canon on W3C related technologies. I am currently the W3C Activity Lead for Multimodal Interaction, and was also the Activity Lead for Voice Browsers up until January 2005.

From October 2000 through January 2003, I worked for Openwave as an architect for Internet standards. This gives me tremendous opportunities to influence the convergence of mobile, voice and traditional Web technologies.

From 1995, I worked for two years at MIT's Laboratory for Computer Science in Cambridge Mass. as part of the W3C team on assignment from HP Labs. I then returned home to England, continuing my involvement with the Web as a W3C Fellow.

1985-1994 working for HP Labs, at Bristol England on knowledge-based systems, hypertext and user interfaces. I started work on the Web in 1992 and gradually shifted my focus until I was working full time on this in 1994.

1984 working for Hewlett Packard's Office Productivity Division

1980-1984 working for Research Machines Limited in Oxford on architecting and implementing system software for local area networks on Z-80 computers for schools and colleges.


I studied Physics at the University of Oxford, and stayed on to get a doctorate in Astrophysics. During this time I developed a fascination for computers and went on to spend a year as a research associate at the Machine Intelligence Research Unit at The University of Edinburgh in Scotland. In 1995 I spent six months at the Logic Programming Department of Imperial College, London, as an Alvey Journeyman.

Historical Stuff

HTML+ or as Postscript (222417 bytes)
This was a proposal I made for extending HTML, first published as an Internet Draft in 1993, and in summary form at the 1st international World Wide Web conference held at CERN in 1994. HTML+ played a key role in the development of forms for Mosaic, and later for HTML tables.
The Arena browser
I developed Arena in 1993/94 to demonstrate the ideas in the HTML+ specification. This was the first browser to support background images, tables, text flow around images and inline mathematical expressions. The browser was further developed by CERN and later W3C as an experimental platform for CSS and libwww. Here is a copy of a presentation made using Arena at Web World, in Orlando in early 1995. Here are screen shots of figures, tables and math.
HTML 3.0 or as plain text (381229 bytes)
An extended version of HTML+ this was submitted as an Internet Draft in 1994. Like HTML+ it was never standardized, but helped to stimulate further work on features like tables. This is also available as a compressed plain text version with form feeds 125,701 bytes, and as a compressed tar file of the source html files 156,346 bytes. Note the plain text uses carriage return, line feed pairs and form feeds between pages.
Mediated Digest Authentication
This is a scheme for using trusted third parties to mediate authentication of web servers and clients. It is a superset of the Digest Access Authentication scheme, and is based on the MD5 algorithm. As such there are no export or import restrictions.
Extending the Web to support distributed VR (the seminal paper for VRML)
I organized a birds of a feather meeting on virtual reality at the 1st international World Wide Web conference held at CERN in 1994 and presented my vision for VRML (a term I coined). I also got to give the same paper 4 weeks later at the Internet Society conference in Prague, talking along side people working on VR projects for the US Department of Defense. The constrast in style and culture was really striking! VRML got off to a huge start with thousands of people joining the new email list. See the VRML home page for up to date details on VRML.

Volantis Email: dsr@w3.org, phone/fax: +44 1225 866 240 mobile: +44 7966 964 409 (GSM)