Behavior of user agents after granting exceptions

Hi Folks,


another request for clarification. The underlying question is to what
extend the actual DNT headers sent depend on the prior granted exceptions.

Please respond if you believe that I misrepresented our current consensus:

1. A user agent receives a JavaScript request for a user-granted
exception from a site. This javascript call conveys the _desire_
    of the site to obtain the indicated type of user-granted exception.

2. A response of 'true' indicates that the user granted this exception
(at this point in time). 'false' indicates that the 
    user or user agent was unable to obtain such an exception.

3. At a later point in time, the user agent sends a DNT;0 or a DNT;1 header

4. For most user agents, [3] will depend on [1+2], i.e., sites with an
exception will receive DNT;0 while sites without
    receive DNT;1

5. We nevertheless permit any other behavior of user agents, e.g.,
    a) User agents ignoring the requests for exceptions (while returning
true or false when the API is called)
    b) User agents returning TRUE for the Javascript call and then later
still sending DNT;1 (somewhere or everywhere)
    c) User agents using other algorithms to determine whether to send
DNT;0 or DNT;1 (and for the return value of the API call).


One consequence of this approach is that if a first party needs to
_ensure_ that certain third parties are indeed exempted, then
the only definitive and final reference are the actual DNT headers
received. While in practice, the return value for the Javascript call is
a strong indication, it is not definitive.

I believe that we want to foster innovation in user agents and allow
user agents with limited (or no) javascript capabilities. Therefore I
believe this to be a viable approach. Please respond if you disagree or
if I misrepresented our current view.

Regards,
matthias

Received on Friday, 13 April 2012 02:02:35 UTC