TAG work in recent months, February 2009

[Note that the TAG's previous report was published in October 2008, which was shortly before the 2008 W3C Technical Plenary / Advisory Committee Meetings.]

This is a report from the W3C Technical Architecture Group to the W3C membership on TAG activities from October, 2008 through early February, 2009.

TAG Membership and Leadership

Since the previous status report, there have been several changes in the membership and leadership of the TAG:


Immediately after the publication of the previous summary, the TAG met in joint session with the HTML Working Group at the Technical Plenary. The groups explored a number of areas including:

Discussion continues in the HTML working group, the TAG and elsewhere regarding all of these issues, and we believe that our joint work is bearing fruit in a number of ways.

At the joint session, several members of the TAG advocated modularization of the HTML 5 Draft Recommendation. Since then, HTML 5 editor Ian Hickson has sent a list of sections and corresponding work estimates for the HTML 5 draft:

Ian's list will be input to further analysis of the possibilities for modularizing the HTML 5 specification. Another related development was HTML Working Group (former) co-chair Mike Smith's 14 November invitation to comment on a draft, HTML: The Markup Language, which is an experimental effort to explore modularization of the HTML 5 specification.

Also at least indirectly following from the TAG's discussions with the HTML working group at TPAC, Adam Barth worked with Ian Hickson on Content-Type Processing Model, a 9 January Internet Draft that demonstrates progress in factoring out a part of HTML 5 that overlaps with the HTTP specification. (See also The HTTP Origin Header, a 21 January Internet Draft by Hickson and Barth.)

TAG members Dan Connolly and Larry Masinter are also working with Ian Hickson and others on a draft, planned for early March, that will attempt to address concerns relating to treatment of URIs in the HTML5 draft specification.


The TAG has spent many years discussing a wide range of issues relating to the ways that languages evolve, interoperation of systems on the Web in which sender and receiver are built to understand somewhat different versions of a language specification, and in particular issues relating to the evolution of XML-based languges. This work has been tracked as our ISSUE-41, What are good practices for designing extensible XML languages and for handling versioning? Draft findings in three parts were developed and repeatedly revised, with the latest versions available at Extending and Versioning Languages: Terminology, Extending and Versioning Languages: Compatibility Strategies, and Extending and Versioning Languages: XML Languages. The TAG has also in recent months considered a formalism for language versioning, developed by TAG member Jonathan Rees.

The TAG's work on versioning has already proven useful to other working groups such as XML Schema, which cites the TAG's earlier draft findings as an influence on the design of versioning features in XML Schema Definition Language (XSD) 1.1 Part 1: Structures.

On its teleconference of 29 January 2009 the TAG resolved not to pursue publication of this work as a TAG finding in the immediate future. Instead, the TAG has agreed that outgoing TAG member David Orchard, who edited these drafts, will publish the versioning strategies draft in the form of W3C Note(s) under his own name, citing the TAG's contributions to the content, and incorporating the formalism discussed above.

Error handling

Typical Web browsers accept and attempt to render and provide scriptable control over a wide range of nonconforming documents that are served with Content-types such as text/html. For example, browsers tolerate and process documents that have improperly nested tags, documents that are missing required markup, and so on. In order to promote interoperability when dealing with such "erroneous" content, while minimizing the need for implementors to reverse engineer one another's browsers, the HTML 5 draft provides detailed specifications not only for dealing with "correct" input, but also for the processing of such erroneous documents.

The TAG on 9 December 2008 reopened its ISSUE-20: What should specifications say about error handling?, an issue that the TAG originally raised in 1992. The TAG has begun exploring the tradeoffs between convenience for users vs. promoting the publication of correct content on the Web. We are also considering the appropriate layering of specifications for such error handling.

TAG Finding: The Self-Describing Web

On 7 February 2009, the TAG published a new finding, The Self-Describing Web. The finding describes features of the Web that support reliable, ad hoc discovery of information, with examples using formats such as Atom, RDFa, GRDDL, XML, and RDF. It also explains how the specifications needed to interpret documents retrieved from the Web can be determined unambiguously, starting with the specification for URIs (RFC 3986). The TAG thanks the many participants in the TAG's www-tag@w3.org (archives) mailing list who contributed to the development and review of this finding.

Use of HTTP Redirection — access to resource descriptions and metadata

During this period the TAG has discussed extensively questions relating to accessing descriptions of resources, as opposed to the representations returned with HTTP status code 200, and the role if any of HTTP redirection in accessing such descriptions. This work is being tracked, at least for now, as part of the TAG's ISSUE-57: The use of HTTP Redirection. There have been some suggestions to broaden this analysis to include other questions relating to metadata on the Web, including for example, site-related metadata. In conjunction with this study, the TAG is evaluating and working with the authors of these documents, among others:

About the Technical Architecture Group

The Technical Architecture Group (TAG) was created in February 2001. Three TAG participants are appointed by the Director and five TAG participants are elected by the Advisory Committee. The mission of the TAG is stewardship of the Web architecture. Included in this mission is building consensus around principles of Web architecture, resolving issues involving Web architecture, and helping to coordinate cross-technology architecture developments inside and outside W3C.

Details on TAG activities can be found from the TAG home page. The TAG meets weekly and sends summaries of its activity to the AC, Chairs, and public TAG mailing list (www-tag archive). The TAG welcomes public discussion of open issues, as well as proposals for new issues, on that same list.

TAG Participants

Following the TAG election results announced in January 2009, the TAG participants are:


Noah Mendelsohn, TAG co-chair
Dan Connolly, W3C TAG team contact

$Revision: 1.21 $ of $Date: 2009/02/28 15:57:06 $

    Valid XHTML 1.0 Transitional     Valid CSS!