Re: introduction to the TPE

On Apr 10, 2014, at 10:49 AM, Dobbs, Brooks wrote:

> Okay so "forbidden" is a bit of a straw man here.  I don't think anyone said that it was forbidden.  I do maintain that  high level discussions that occur in an intro should stay with the things that compliant participants MUST do.

I don't, nor am I aware of any specification, at any SDO, that would
support such a view.  In fact, they almost always state what can be
done with the protocol, not the minimal subset of what has been required,
and often include wishful thinking about what might be possible
in the future.

As much as I appreciate all of these meta-discussions about how to
properly edit a specification, please keep in mind that I do have a
bit of experience in the matter and don't really need any more input
on things that fall in the realm of editorial preference.

What I need is input on the accuracy of what has been written, in
accordance with the WG decisions, and insight into what might be
missing to enable implementation, preferably in the form of patches
(before/after text) that can be easily adopted without several more
rounds of WG review.

If sending DNT:0 on every request was impossible or even discouraged
by the protocol, then I would have made your suggested edit.  However,
that is not the case, and the spec has not been interpreted that way
in practice.  We should therefore not discourage browsers from
implementing such a configuration on their own.

....Roy