"Adult supervision"

[Mark Allman]

> > Should IESG members really have to debate with each document
> > author or working group chair (for instance) whether it's okay
> > to assume that a device or server will only be accessible from a
> > local, trusted network and that therefore no authentication is
> > needed?
> 
> Absolutely not.  It would take forever.  ADs are too busy already.
> But the contrary position is also wrong.  You can't just say
> "look, man, everybody knows you can't just have no authentication,
> so shut up and go away."  There is a middle path, where you write
> up documents documenting issues like this and publish them as
> RFCs, and then you can say "look, man, I don't have time to
> explain this to you, but your protocol needs to conform with
> RFCmumble, please go read it."

A few thoughts...

  + I agree that "this will not fly" types of comments are not
    useful.

  + I do not want the ADs spending their time making arguments for
    widely held beliefs (that really will not fly).

So,

  + A little of what you are suggesting has been done (e.g., RFC
    2357, "IETF Criteria for Evaluating Reliable Multicast Transport
    and Application Protocols").  Maybe more is necessary.

  + There is another middle ground.  It would be nice if rather than
    saying "this will kill the Internet" or some such the AD would
    send a note saying something like: "I think the IETF consensus
    is against protocols that do not have {authentication,
    congestion control, ...}." **and then** (as a first step) plumb
    this person together with a few subject matter experts as folk
    they can go talk to in more depth about why consensus is that
    way and whether or not the person has a special case that may,
    indeed, fly.

    In other words, the ADs can instigate good discussion on the
    merits without having to spend a huge amount of time being
    directly involved (at least as a first step).  And, the person
    bringing some proposal gets good technical thoughts on the
    merits of the question.

allman


--
Mark Allman -- BBN/NASA GRC -- http://roland.grc.nasa.gov/~mallman/