Agreeing on Do Not Track

Tom Lowenthal

1

Last week saw the first meeting of the W3C’s Tracking Protection Working Group, a cross-section of advertisers, browser vendors, publishers and public interest groups come together to agree on a standard for Do Not Track (DNT).

We believe the group’s mission is vitally important. Without a well-defined DNT mechanism, it has been common for Web users to have their reading habits recorded by companies they have never heard of, and have no relationship with. Sophisticated tracking techniques have been appearing faster than users’ ability to defend against them. The mission of this working group is to give readers a simple and meaningful way to regain control.

Last week’s Working Group turnout was comprehensive, with about 40 folks present from a variety of organizations — comfortably more than the ten required by the charter. The chairs, Aleecia McDonald (Mozilla) and Matthias Schunter (IBM), emphasized the need to work quickly in producing this consensus standard and outlined an aggressive schedule, with the next meeting at the end of October, and a Last Call Working Draft expected before the end of the year.

In two days, we made healthy progress, agreeing broadly on the DNT header in its current form, and brainstorming ways to extend it into a more refined standard. There was no shortage of ideas around the table and we put together quite a to-do list of topics to work on in the coming weeks and months. Although there was healthy discussion on most of the issues raised, there was a surprising amount of agreement on many of the substantive questions, which bodes well for the production of a robust and widely-used standard.

The entire workings of the group are public. If you’re interested, then you can sign up for the group’s mailing list, which is also publicly archived. The minutes from both days of last week’s meeting are available, and further updates, including minutes of the weekly conference calls will continue to be archived on the group’s page.

This post was co-written by Peter Eckersley of the EFF, and Tom Lowenthal from Mozilla.

One response

  1. Peter Cranstone wrote on ::

    Here’s the problem that the DNT header does not address – “Context”. DNT is a binary solution to a problem that is far more complex. It’s a just say No approach to something that needs a “maybe yes, if i can trust you and return value”.

    The web provides the perfect mechanism to add more context around privacy. DNT is just a single “header” what is required to solve the problem is additional headers.

    Cheers,

    Peter
    3PMobile