RE: Towards Text for ISSUE-176... (intermediaries)

Matthias,

I don't believe the statement is too strong and there was a time when the working group mostly agreed on this approach.  This feels to be a subjective judgment late in the game to reverse course so we should review the motivations that arrived at the draft language prior to making a change without stronger evidence of a need to do so.  Unless something has radically changed, those of us on the publisher and 3rd party side of the discussion still feel  very strongly that for the standard to be viable (voluntarily adopted) we'll need to ensure user preferences are explicitly expressed and intermediaries make this very difficult to do (do to their very nature there is very little to no interaction with the actual user).  That said, the current draft language still allows that outcome with the appropriate controls in place ("specifically installed for...").

I would ask we leave the language "as is".

- Shane

From: Matthias Schunter (Intel Corporation) [mailto:mts-std@schunter.org]
Sent: Monday, January 21, 2013 6:29 AM
To: public-tracking@w3.org (public-tracking@w3.org)
Subject: Towards Text for ISSUE-176... (intermediaries)

Hi DNT Team,


The following issue is still open:
- ISSUE-176: Requirements on intermediaries/isps and header insertion that might affect tracking
  http://www.w3.org/2011/tracking-protection/track/issues/176

I believe that we discussed this issue earlier and our spec currently contains this text:

An HTTP intermediary must not add, delete, or modify the DNT header field in requests forwarded through that intermediary unless that intermediary has been specifically installed or configured to do so by the user making the requests. For example, an Internet Service Provider must not inject DNT: 1 on behalf of all of their users who have not expressed a preference.

I believe that this statement is too strong. I suggest to augment it along the line of "If an intermediary wants to add, delete, or otherwise modify any DNT information, it needs to satisfy the constraints for user agents put forward in this document."

This would permit tools to change the information as long as they ensure that the resulting values, e.g., still reflect user preference. It also allows to add DNT support in an intermediary or OS as long as the requirements that we described are met.

Opinions?


Regards,
matthias