Re: Out-of-area ADs [Re: IETF areas re-organisation steps]

[Nico Williams <[e-mail]>]

On Mon, Dec 29, 2014 at 08:09:28AM +1300, Brian E Carpenter wrote:
> On 29/12/2014 07:33, Dave Crocker wrote:
> ...
> >> However, we have not found that ADs are so specialized that there
> >> is a "correct" AD for every WG,
> > 
> > raises the possibility that AD job descriptions ought to make
> > explicit reference to cross-area skills?  This, of course, leads to
> > the challenge of figuring out what that means, in pragmatic terms.
> 
> This ties back into Nico's point about maybe flattening the hierarchy (*)
> and essentially abolishing areas as such. I have much sympathy with that,
> although it's a bit scary. [...]

I get the fear.  But I think the risks are manageable with a simple
prescription:

>                     [...]. But it means that we would indeed change the
> criteria for picking ADs. [...]

Here's my high-level proposal:

   We need an IESG mostly populated by generalists who can specialize,
   and a few specialists who can generalize.

It's a compromise, of course.  You've all heard the about the jack of
all trades, master of none...  And having a specialist for every
possible, relevant specialization is expensive (which I think is why
we're discussing this _at all_).

But here's the thing: it is actually expected and required of most
people, over their lengthy (nowadays) lives, that they will adapt and
change all sorts of things, changing careers, and becoming experts in
new fields.  At least smart people anyways, which the IESG had better be
chock-full of.

It's fair to say that some IESG members will be more expert in
"security", and others in "transport", but damn it, it ought to be the
case that the neither is completely ignorant of the principles and even
details that the other deals with.

How the IESG members spread the workload amongst each other is not that
interesting to me, really -- I have better things to concern myself with
(since I'm not on the IESG).  That they do it effectively is, and that
they don't consequently drop important things by the wayside is too.  We
already have directorate reviews to assist IESG reviews, and IESG review
already provides an opportunity for the full power of all the IESG's
members to be applied to every document crossing its path.  What more
can we ask for?

>                    [...]. We wouldn't be looking for, say, a Transport AD
> who is a widely recognised expert on congestion control, or a Security AD
> who is competent to verify a crypto algorithm. In fact, over-specialisation
> would be a *disqualification* for serving on the IESG.

Yes-ish.  I don't mind having a few specialists whose job is to prevent
the sorts of gross mistakes that... the IESG is there to prevent.  I
think the specializations needed would be few, maybe just three,
roughly:

 - security (the sorts of people who can tell that your protocol has
   security problems; not necessarily cryptographers),
 - networking (the sorts of people who can tell you that your protocol
   will cause harm to the Internet, or will fail to scale),
 - and operations (the sorts of people who can tell you that you're
   using DNS incorrectly).

All other specializations should be selected for (or against) on the
basis of what is needed more of at the time.

> Serving as a Gen-ART reviewer has been a great experience for me, but has
> often forced me out of my technical comfort zone. It would be the same
> for the IESG in such a new order. They would have to depend more than
> today on expert reviewers.

Yes, and they've already been doing that anyways.

Nico
--