From W3C Wiki
Jump to: navigation, search

The focus of this document is the structure of the W3C's Advisory Board (AB). The AB was not part of the Consortium when it was created in 1994. The W3C Director approved the creation of the AB in 1998.

Some Consortium Members consider the current structure as sub-optimal. This document includes some historical information about the AB's current structure, problems with the current structure, alternative structure proposals and recommendations.

This document is a WorkInProgress (WIP). Feedback is welcome, preferably by directly editing this document or by sending email to public-w3process@w3.org (archive) with a Subject: prefix of [ABStructure]. We are also interested in input from non-Members i.e. The Public.


The AB was created in 1998. It's role is defined in the W3C Process Document (PD). The PD states the AB will consist of 9 elected Members and that individuals nominated for the AB do not necessarily need to represent a W3C Member.


  • What are the requirements and use cases that led to the creation of the current AB structure?
A: I believe this came out of the Peabody meeting, in which it was determined that W3C needed a process and formal structures for the members to keep tabs on what W3C was doing, and enable them to provide guidance on what the W3C should be doing. This question should be answered by people who were involved. -chaals
  • What problem(s) is the current AB structure designed to solve?
A: group of members small enough that team feel comfortable consulting them in confidence, representative enough of the members that their input will generally be what the full membership would say, and committed to being available more often for more time than most AC reps. -chaals
  • Do the requirements and use cases from last century still hold today, and thus, still warrant an elected Advisory Board?
  • Why does the Staff need 9 proxies to represent the Consortium's ~400 Members?
A: Rather than consult with the AC representatives on topics they feel are or may be sensitive, staff will have more private discussions with people they select. This effectively reduces representation of members, and openness of process. - chaals
  • What mechanism(s) are non-Members supposed to use to provide advice and recommendations to Staff?
A: Join... ;) More seriously, they can also use the w3process community group, or pass recommendations to someone who is a member or staff to bring them up through the existing processes. - chaals


  • Nine AB members is too narrow to represent the almost unbounded interests of Consortium Members plus the World Wide Web.
  • Election are run in a non transparent way and basically no data are available, so that it's impossible to determine if people are playing the voting system - JC


Here are some alternative ways to structure the AB, not necessarily mutually exclusive options:

  • Make the AB self-selected
    • The earliest proposal appears to be by Arthur Barstow in June 2009
    • Michael Champion's May 2013 proposal
    • Dissent from JC
  • Move the AB's Process related tasks to an AC Task Force or Community Group; Wayne Carr June 2014
    • Support from JC

At their February 2012 Meeting, the AB discussed term limits and other mechanisms to increase the diversity of the AB (more discussion)

Some institutional memory is extremely helpful in AB discussions. Some continuity as well as some turnover makes the AB more effective and reduces the likelihood of them repeating themselves without knowing. - chaals

Efficiency of these small committees are strongly related to their memory. Replacing everyone is surely a good way to make sure the group becomes irrelevant - JC