W3C home > Mailing lists > Public > public-tracking@w3.org > November 2011

Re: Issue-4

From: David Singer <singer@apple.com>
Date: Wed, 09 Nov 2011 11:42:00 +0900
Cc: "Aleecia M. McDonald" <aleecia@aleecia.com>, Tracking Protection Working Group WG <public-tracking@w3.org>
Message-id: <F88C6CE8-C80D-4729-81B1-BEF5E264FD9D@apple.com>
To: Karl Dubost <karld@opera.com>

On Nov 9, 2011, at 11:23 , Karl Dubost wrote:

> Le 8 nov. 2011 à 20:11, David Singer a écrit :
>> In general, protocols are designed such that default values can also be explicitly stated, and currently we don't have a "not stated" explicit value, either.  Do we need one ("DNT:<blank>", perhaps)?
> what about a simpler no DNT header == not stated. 

well, that's where we are today.  I don't mind, I am just pointing it out that the more usual technique is to say, in a protocol, something like:

"The optional field 'meadow' can take the following values: grass, clover, corn, or mixed;  if the field is not stated, 'mixed' is the default value."  That is, the assumed default is also a permitted explicit statement.

> It would give the possibility of a service along
> "We noticed that your DNT header was not set, 
> if you put it on, it means blah
> if you put it off, it means foo
> you can leave it undefined, it means bloubiboulga."

"We noticed your header was missing, or set explicitly to "declined to state".  There are advantages to an explicit statement…"

David Singer
Multimedia and Software Standards, Apple Inc.
Received on Wednesday, 9 November 2011 02:42:35 UTC

This archive was generated by hypermail 2.3.1 : Friday, 3 November 2017 21:44:42 UTC