OpenAndTransparentW3C/AC Reviews

W3C doesn't currently report how its members influence the formation of working groups or allocation of staff resources. Should that change?

Background

W3C organizes much of its work into groups, such as a Working Group (WG) or Interest Group (IG), and allocates some its staff resources into those groups, using a measure called Full-Time Equivalents (FTEs). In other words, if the time of a single staff member represents 1 FTE, each group will be apportioned some percentage of that; for example, if the Foo WG has 0.2 FTE, that means that 20% of a staffer's time (or 10% each of 2 staffers) is dedicated to helping that WG in a variety of roles; these staffers are called Staff Contacts, or Team Contacts.

One of the benefits of W3C membership is that the members have a say in which groups are formed, and how these FTEs are allocated within those groups. This is important, because most W3C Team are funded by member fees, and we have limited resources, so members have some control over the priority and allocation of these limited resources.

W3C's oversight is provided by the Advisory Committee (AC); each W3C member is allowed a single AC representative (AC rep), who serves as the primary official interface between their organization and W3C.

When W3C is considering whether to form a particular group, the Team puts together a charter that describes the background and scope of the group, lists the deliverables (e.g. specifications), and provides details such as the proposed chairs and staff contacts (and their FTEs), the group confidentiality (public or member-only), meeting schedule and timeline, and other relevant information.

The AC is then asked, via a survey on our Web Balloting System (WBS, also called a survey, ballot, poll, or questionnaire), if they support the formation of this group, and the allocation of those resources; they are also asked whether their organization is likely to participate in the group. Each member evaluates their interest in the activity, with considerations of IP commitment, priority, relevance, and other issues, and responds to the WBS with one of the following options (with optional details in a feedback input):

• supports this Charter as is
• suggests changes to this Charter, but supports the proposal whether or not the changes are adopted
• suggests changes to this Charter, and only supports the proposal if the changes are adopted [Formal Objection]
• opposes this Charter and requests that this group be closed [Formal Objection]
• abstains from this review

These polls are open for a minimum of 4 weeks. At the end of the poll, the staff contact and domain lead answer questions, address open issues, and modify the charter according to the consensus of the AC; W3C management (W3M) evaluates the results of these surveys and discussions, and decides whether to form the group.

AC review results

The results of these AC reviews are usually open to the entire members (there is an option for an AC rep to provide Team-only replies). But they are not open to the general public.

This information seems relatively harmless, especially in the aggregate (e.g. not indicating which member responded or how they responded), and could be helpful to show the public a view inside our process.

AC Review Transparency Preferences poll

In 2008, W3C took a poll to ask our members if these poll results should be more transparent; the AC Review Transparency Preferences results (member-only) are available, but member-confidential.

Thus, we can't show who said what, or even give concrete numbers, but we can describe general conclusions.

Composition of the poll

The poll asked questions along several axes:

• Member Visibility vs Public Visibility
• During Review vs After Review
• Aggregated Results vs Details Without Attribution vs Individual Responses
• Compulsory Transparency vs Default Transparency vs Optional Transparency

This led to a relatively complicated survey that may have been difficult for members to assess and answer.

Low response rate

Only 16 of 416 AC reps responded to this poll, less than 3.9% of our members. Its not clear how conclusive this result can be considered.

General results

At the time, members were more comfortable with other members seeing the results than letting the public see the results. They were also more comfortable with results being available in aggregate than in detail. They were also more comfortable with transparency being opt-in, rather than compulsory or even default.

Member visibility of aggregated results

There was near-unanimous (all positive or abstain) consensus that aggregate results should be member-visible after the review was over: Member Visibility of Aggregated Results After Review (member-only).

Revisiting transparency?

In 2014, the world, the Web, and W3C have changed. Openness and transparency are more expected. We have seen the result of experiments in openness at W3C, and it has been mostly positive. A new, simpler survey of W3C's members now would likely have different (and more conclusive) results.

If you support a new survey of W3C openness and transparency of resource allocation and group formation, please email the public-openw3c mailing list, or tweet your opinion to @openw3c with hashtags #openW3C and #results, with a link to this page so people know the context. Tell us why you think this is important, don't just give a +1.