- From: Rigo Wenning <rigo@w3.org>
- Date: Wed, 24 Oct 2012 17:53:04 +0200
- To: public-tracking@w3.org, ifette@google.com
- Cc: Mike O'Neill <michael.oneill@baycloud.com>, "Roy T. Fielding" <fielding@gbiv.com>, Nicholas Doty <npdoty@w3.org>
Ian, you know I'm personally against the distinction. As far as I understand you suggest to just return the value (can do DNT) on request. But this doesn't return what the service actually does. It can decide to still track you because it thinks it is a first party. But the user wants to know whether she is tracked. This is part of the response. Self determination. So if you suggest a simple response, you also suggest removing the distinction between first and third parties. For the EU stuff, this would be perfect. For usability, clean protocol and semantics it would be nice. But we have all those first parties that still want to track and benefit from the exception. And we have seen some of the ad representatives starting to consider to remove the 1st/3rd party distinction. Yet the arithmetic that comes with this model doesn't sound sweet to other participants that want to save some income over a transition phase via the 1st/3rd party definition. This is why the matrix (for the US solution) is so complex. For EU, the answer is simply. They respond 3 until they get DNT:1. We shouldn't overestimate the readability of tokens. The user agent UI (if any) will give meaning to the token. So the text only has to reflect the functional properties (this directed to Mike O'Neill) and can be left as is even for the EU solution. Best, Rigo Rigo On Tuesday 23 October 2012 16:02:37 Ian Fette wrote: > I still don't understand the need for this. The server should > simply state "Yes, I support DNT" or "No, I don't support DNT" > (or alternately "Yes, I'm honoring your request" or "No, I'm not > honoring your request.")
Received on Wednesday, 24 October 2012 15:51:16 UTC