Monthly Archives: July 2013

What should the W3C TAG do next?

I am running for election to the W3C Technical Architecture Group (TAG).

What is the TAG known for? Well, primarily the Architecture of the World Wide Web, Volume 1 which was created in 2004. This document summarized principles, constraints, and good practice notes as of that time.

Since then the TAG has reviewed architectural issues on what appears to be an ad hoc basis, producing what are known as “TAG Findings“, expert answers to specific questions.  I surmise that the visibility and usefulness of these has been less than it could be, probably since W3C working groups are not always either aware of them at all, or not sure which might be applicable to the problems they are working on, or if they’ve seen them, not sure how to apply them to the specific problem they are facing.

The TAG has also devoted much effort on work that has not reached conclusion, given the wide constituency and difficulty of the problems they have worked on. In some cases the work has stopped with “draft findings, in other cases only discussion on the mail list. A lot of good ideas and information may be lost but it is hard to tell.

The W3C is in a sea of change, including movement toward “living documents” as opposed to versioned dated documents. The idea of versioned documents was simple – a group agrees to approve a definitive version as of a date in time, making it a standard. The benefits of this approach are clear – there is a static document that can be referenced, will not change, and for which consensus can be clearly recorded. The issues are also clear: with faster development cycles static documents can become out of date more quickly making it misleading to have a static approved document as the target of links when newer corrected material is available yet not readily found.  The solution to this issue is to have continuously updated drafts, so that the latest version is the definitive version, so called “living documents”. The concern here is that the process can spin out of control, with editors adding what they will without checks and balances – “if you don’t spot it, you must approve it”could become the new mantra. We need both – a rapid cycle time as well as clarity of approval and agreement.  (The solution in progress appears to be pull requests in git accepted by those we trust).

What does this have to do with the TAG? Well the TAG is part of this sea of change as well, as reflected in the previous TAG election where a common theme was that the TAG need work less on abstractions and closer to the needs of developers and working groups. There is a desire that the TAG produce material relevant to current work on web applications (and other topics) and that this material can be easily found and used.

It seems the time has come for an new volume of the Architecture of the World Wide Web, addressing topics related to Open Web Platform applications, APIs and programming, open data, and security and privacy.  I’d argue for a new volume, as I’d rather not see history rewritten (e.g. to expunge XML, despite it’s continued use in various communities). It is usually easier to correct and improve a draft than to create a new one, so the TAG should seed the process of “documenting and creating consensus around principles of Web architecture” by creating a new Architecture of the World Wide Web volume and working with the W3C community to get it right. This continues the original TAG mission yet represents a change from the recent past in how it is done.

I suggest that focusing on addressing high priority issues of the web architectural evolution and interoperability with the focus of producing a specific document can give the TAG focus and enable  feedback, organizing “findings” and issue resolution into a specific result that can be readily shared. I can help with this, both with the writing and creation of this new work as well as collaborating with the TAG, in the Working Groups and communities (such as PING and others). What I can do, as a member of the TAG, is to get this started, organized, written down, and communicated so that the work of the TAG is visible and useful to the community and so the community can get involved to make it better.