W3C home > Mailing lists > Public > public-tracking@w3.org > January 2014

Re: fyi: doNotTrack attribute in API

From: Marcos Caceres <mcaceres@mozilla.com>
Date: Fri, 17 Jan 2014 09:15:12 +0000
To: "Roy T. Fielding" <fielding@gbiv.com>
Cc: Matthias Schunter <mts-std@schunter.org>, "public-tracking@w3.org (public-tracking@w3.org)" <public-tracking@w3.org>
Message-ID: <51590C5877AF4B728348FEBE0564F7AD@mozilla.com>



On Wednesday, January 15, 2014 at 3:57 AM, Roy T. Fielding wrote:

> On Jan 14, 2014, at 11:14 AM, Matthias Schunter (Intel Corporation) wrote:
>  
> > Hi TPWG,
> >  
> >  
> > enclosed some technical feedback received via the public mailing list.  
> >  
> > I appreciate the feedback and deem the changes to be non-controversial,
> > and I have asked Roy/David introduce corresponding edits.
> >  
> > Please comment if you see a need for further discussion in the WG before fixing the spec.
>  
> The main point reflects the discussion of the WG. The second point about
> readonly can be fixed.
>  
> > -----Original Message-----
> > From: Marcos Caceres [mailto:mcaceres@mozilla.com]  
> > Sent: Wednesday, January 08, 2014 7:59 AM
> > To: public-tracking-comments@w3.org
> > Subject: doNotTrack attribute
> >  
> > There seems to be some inconsistencies between the spec and browsers with regards to the `doNotTrack` IDL attribute. In the spec, it says it’s on the `window` object in some places, but in other places it says it’s on the Navigator interface. In browsers, it’s exposed on the Navigator interface (at least in Chrome, and FF), as it should be.
>  
>  
>  
> The browser implementations were premature (and undefined). In this case,
> it is desirable to supplant those implementations as out-of-date.
> Concern about deployed content being broken is not yet relevant given
> the lack of implementation by sites.

Ok, but if implementations are going to remove this, then they should be putting this behind a flag until the spec is stable (i.e., this should not be exposed to content by default).  
  
> > Also, why is it not read only in the spec? It’s implemented as readonly in browsers - and it wouldn’t make any sense to make it writable … I’m sure it’s just an oversight :)  
>  
>  
> Yes, that was an oversight and has been corrected. Thanks.

If this value can be set while a user is viewing the site, how do scripts get notified that the value has changed on a per site basis?  
  
Received on Friday, 17 January 2014 09:13:28 UTC

This archive was generated by hypermail 2.3.1 : Wednesday, 7 January 2015 14:41:42 UTC