Re: Today's call: summary on user agent compliance

On Jun 6, 2012, at 11:48 , Aleecia M. McDonald wrote:

> We did NOT hear a view that the specification should require publishers to honor DNT:1 signals from non-compliant User Agents.

I think that I have consistently argued that when the two ends adhere to the protocol (i.e. their expression and responses are correct), then it's non-compliant to do other than the protocol requires, both in email and on the call.

You might have good reason.  But it's still not compliant.  I sent you "Please do X", and you replied "No, I won't, I don't believe you."  I don't think you can describe that as *compliant*.  You might think it *justified*.

Note well that this goes as much for servers claiming compliance and expecting to be treated as compliant, when in fact they implement something else (e.g. "do not target"). What is sauce for the goose is sauce for the gander, as they say, and if we write anything about one, we should write about both.






David Singer
Multimedia and Software Standards, Apple Inc.

Received on Wednesday, 6 June 2012 22:00:32 UTC