- From: Roy T. Fielding <fielding@gbiv.com>
- Date: Wed, 12 Sep 2012 12:44:02 -0700
- To: <rob@blaeu.com>
- Cc: <public-tracking@w3.org>
On Sep 12, 2012, at 12:06 PM, Rob van Eijk wrote: >> As I've said multiple times now, if the WG disagrees with the text >> in the spec, then the right way to do so is to object to the text >> with a specific change proposal, in writing, on what must be changed >> to resolve that objection. Nobody has done that. > > I submitted text already. Text which has nothing to do with the default issue, but appreciated nonetheless. > http://lists.w3.org/Archives/Public/public-tracking/2012Sep/0136.html > > I propose text in the TPE in chapter 3 that is clear enough, for > example: > > "Implementations of HTTP that are not under control of the user, > including Web Servers, MUST not drop or modify a tracking preference". The current text says "Implementations of HTTP that are not under control of the user MUST NOT generate or modify a tracking preference." I am pretty sure that was intended just for user agents and things that might alter the user agent configuration, but I'll have no objection to making it similar to the intermediary requirement if we make it clear that an HTTP endpoint doesn't have to hold onto its messages forever. In other words, all messages get dropped at the conclusion of processing, so please be specific. Or, are you trying to accomplish something else? The signal sent by IE 10.0 is not a tracking preference because it fails to match the semantics for DNT. If you want to forbid the way that Apache addressed that issue, this change won't accomplish it. You would have to request a change to DNT's semantics that would allow its tracking expression to not be the user's personal preference. ....Roy
Received on Wednesday, 12 September 2012 19:44:20 UTC