Henry S. Thompson
W3C Technical Architecture Group
1. What is the TAG?
The W3C (the World Wide Web Consortium, founded by Tim Berners-Lee, is responsible for most of the foundational standards which ensure the inter-operability of the technologies which make up the Web, such as HTML, XML, CSS, SVG and MathML. It also strives to protect the interests of all Web users, in areas such as accessibility and internationalisation.
The TAG (Technical Architecture Group has perhaps the widest ranging remit of any of the groups which do the work of the W3C. Its remit is to “to document and build consensus around principles of Web architecture and to interpret and clarify these principles when necessary, to resolve issues involving general Web architecture brought to the TAG, and to help coordinate cross-technology architecture developments inside and outside W3C.”
2. What is my role?
I’ve been an elected member of the TAG since 2005, with support from JISC for travel costs. Although TAG members don’t explicitly represent particular constituencies (they are elected by the W3C membership as a whole), I’ve tried to pay attention to issues of particular relevance outside the United States in general, and to the UK in particular.
3. What has the TAG been up to lately?
Since the publication of Architecture of the World Wide Web, Volume One, the TAG has mostly operated in a more focussed, issue-driven mode. There have been three specific topics under consideration over the last year or so: the future of HTML, the revision of HTTP and a cluster of issues around URIs, including persistence, semantics and conversion between different formats. (There is a full public listing of TAG issues and open action items available online.) On a broader canvas, the TAG has begun working towards a possible new publication on the Architecture of Web Applications. The following subsections look at each of these areas of work in turn.
3.1. The future of HTML
The TAG is not directly involved in the W3C’s work to produce HTML5, the next version of HTML. But it has been actively engaged in monitoring the progress of that work, particularly in areas of relevance to Web Architecture. This has involved the TAG in discussion with the HTML WG and its chairs on such matters as XML compatibility, modularity, (distributed) extensibility, accessibility and approach to language definition. In some of these areas, for example XML compatibility and language definition, TAG intervention seems to have had a significant positive impact, leading to new work and/or substantive revision to the HTML5 spec. In others, the discussions have been less fruitful, at least so far.
3.2. The revision of HTTP
The HTTP working group of the IETF has begun work on the first revision in over 10 years of the specification of the Web’s key transport protocol, HTTP. There has been excellent liaison between the TAG and the working group, with a number of the changes in the draft revision arising directly from TAG input. Modifying a specification of such importance requires great care, and the TAG is helping to provide independent review as the work goes forward.
URIs keep being used in new ways and in new circumstances. A number of issues have arisen or come to the fore recently at the intersection of their use on what one might call the ‘old-fashioned’ Web with their use on the ‘Semantic’ Web, or the Web of Linked Data, as it is now often referred to. These include deep questions about the precise meaning of response codes such as 200, 303 and even our old friend 404, more specific issues including the appropriate level of commonality for the interpretation of fragment identifiers (that is, the part after the hash (#) in a URI) across all the so-called ‘+xml’ media types and issues which are almost as much organisational as technical, notably the question of just how many places we need to define the mapping from the kinds of strings identifying web resources that we find in XML or HTML documents to the rather more constrained form the IRIs and/or URIs are mandated to take in HTTP requests.
3.4. The Architecture of Web Applications
The growth of ‘Web 2.0’ and the mobile Web has given rise to many new questions about how rich and powerful client-side actors, much more diverse than simple browsers, can and should be governed by the principles of Web Architecture as already understood, and to what extent we need new architectural principles in this area. A lot of the TAG’s current work is focussed on sub-parts of these questions, for example the use of URIs to ‘store’ client-side application state, privacy considerations which arise when device-resident sensors such as GPS expose APIs to web applications and security models and vulnerabilities. The TAG expects to publish drafts in these and related areas in the new year.
4. Find out more
The TAG conducts almost all its work in public, either in or linked from email sent to www-tag. Browsing the public archives is the best way to catch up on the current state-of-play. Minutes from the TAG’s weekly telcons and quarterly face-to-face meetings are always announced on that list. For a slightly longer-term perspective, the public archives of the public-tag-announce list give regular summary updates.
Finally, input on any topic that is, or that you think ought to be, under consideration by the TAG is always in order on the public mailing list.
Henry’s web page is here: