On 5/14/2013 3:46 PM, Andrew Sullivan wrote:
To be fair, for what it's worth as a WG chair I've had the latter
experience at least as often as the former in the use of DISCUSS, and
I've observed some DISCUSSes cleared without any change at all to the
document in question.
We suffer a continuing logic error in the IETF. We use "sometimes it
happens the other way" as if that negates the existence and problem
cause by what is being criticized.
So, yeah, of course a Discuss /sometimes/ causes a small delay with no
changes. /Sometimes/ ADs use the sledgehammer of the Discuss to ask for
a bit of conversation. That's all irrelevant.
What's relevant is the nature of the mechanisms, its capability, and the
cost it can and does impose on authors and the working group.
When a serious defect is identified, it's entirely worth the cost.
When it isn't, it isn't.
In all cases, the person imposing the cost has an obligation to
facilitate closing it, including making clear the criteria for closing
it. It is unreasonable to have those who must do the work to clear it
play a guessing game.
On 5/14/2013 4:03 PM, Ted Lemon wrote:> On May 14, 2013, at 6:30 PM,
Dave Crocker <dhc(_at_)dcrocker(_dot_)net> wrote:
And of course, that's still everyone's preference. But the
reality is that the imposition of the Discuss is an assertion
that changes are being required.
No, it absolutely is not. That may have been the theory when you
were AD, but I can tell you from personal experience dealing with
DISCUSSes on drafts for which I am the responsible AD that that is
not the theory now.
It's the practice now. It's also often/typically the intent now.
"'DISCUSS' is a blocking position; the document cannot proceed
until any issues are resolved to the satisfaction of the Area Director
who issued the DISCUSS."
This, of course, sounds very conversational (except for the blocking
part.) And sometimes that's how it is used. Sledgehammer to demand
polite conversation. Talk with me, or else.
But look over the Discuss Criteria (Section 3.1) and at least half of
the bullets pertain to core technical or procedural failings. These are
not matters of misunderstanding, resolved with some polite
conversations. They require change.
d/
--
Dave Crocker
Brandenburg InternetWorking
bbiw.net