Re: ACTION-212: Draft text on how user agents must obtain consent to turn on a DNT signal

On 11/19/12 6:32 PM, David Singer wrote:
> I take it, following this thread, that you OK with even stronger language for sites, when they are getting consent for an exception?  Sites have every incentive to get users to agree, and it's easy to call the API to log the exception with the UA.  In contrast, the browsers cannot 'hide' whatever their options are, and have little direct incentive to mislead anyone.

As I've said before, I am not necessarily opposed to guidelines for how 
sites/servers get consent. In fact, this could be very helpful, because 
companies will get some advice on how to stay within the lines and be 
good actors.

However, you also have to recognize the imbalance between servers and 
UAs. Servers have incentives to do a good job of getting consent, and 
they face real (legal) consequences if they don't. UA's on the other 
hand have few incentives to fully inform their users about the effects 
of DNT, and have virtually no consequences for doing an inadequate job 
of it.

Received on Sunday, 25 November 2012 17:52:40 UTC