W3C home > Mailing lists > Public > public-tracking@w3.org > June 2012

Alignment of interests and meeting expectations

From: Peter Cranstone <peter.cranstone@gmail.com>
Date: Fri, 08 Jun 2012 10:24:04 -0600
To: <public-tracking@w3.org>
Message-ID: <CBF78344.2D71%peter.cranstone@gmail.com>
As I read through all of the messages on this forum Iım struck with a
feeling that the big hurdle with this recommended specification is one of
alignment and expectations.
 
As a user I expect that when I enable the ³Tell Web Sites to Not Track Me²
setting in my browser that they will do just that. In essence I expect
privacy. So the question to ask ourselves, is that a realistic and
achievable expectation? I think itıs reasonable to believe (certainly from
the forum posts) that itıs not.
 
Why not?
 
For a host of reasons, but they can be boiled down into a few simple items
 
1.    The user has no visual indication that the server has acknowledged
receipt of the header

2.    The user has no visual indication that the server will respect the
header

 
The internet is a client server environment. The client sends a request, the
server responds to the request. For this tracking mechanism to work there
must be a simple UI that alerts the consumer to the above items. The absence
of this can lead to deceptive results.
 
Think about what that means from a practical perspective. Content providers
are going to have to add new code to their web pages to meet that
expectation. Thatıs expensive work. Whatıs their incentive to pay for that
work? It could well be lost revenue. So why do it? We can sit and
pontificate all day long on a forum ­ but thatıs NOT where the rubber meets
the road. At some point someone needs to write the actual code that turns
this spec into something that has real meaning.
 
The easy part of this spec is sending a 1 or a 0 (well maybe not. No current
browser supports the ability to send DNT:0). The hard part of the spec is
the correct interpretation and expectation setting which will be done by the
server on receipt of that header. This is where alignment comes in. I know
there are competing interests and compromise is difficult, especially when
there are potentially billions of dollars at stake. So here we get to ask
ourselves another simple question ­ is it more important to make things
right vs. doing the right thing.
 
The former is management, the later is leadership.
 
What Iım seeing is management, trying to please all the stakeholders vs.
doing the right thing. Itıs important as we get to the stage where this
recommendation is released into the wild that there is clear leadership as
to itıs goals. The user expects not to be tracked ­ if the mechanism
provided by that spec does not meet the users expectation then the law of
unintended consequences steps in. And all the ³server side² arguments in the
world will not resolve it.
 
If this spec is truly going to work then someone needs to build a web site
which can be used as a test use case. It should showcase both first and
third parties. I should be able to turn on ³Tell Web sites to Not Track Me²
in my browser and then access that web site and see the results of the spec..
And donıt forget to think about Mobile as well. Coding a design requires you
to now think about more than one screen size.
 
Why expect others to follow the spec, if those who designed the spec are not
willing to as the saying goes, ³eat their own dog food².



Peter
Received on Sunday, 10 June 2012 23:32:33 UTC

This archive was generated by hypermail 2.3.1 : Friday, 21 June 2013 10:11:30 UTC