Re: A proposal for publishing nominee names

[Theodore Ts'o <tytso@mit.edu>]

On Sun, Jul 14, 2002 at 06:18:58AM +0900, Harald Tveit Alvestrand wrote:
> 
> When faced with gaming attempts in other contexts, my advice has almost 
> invariably been to expose the actions to public view.
> With an informed and fair-minded populace, that will lead to the attempted 
> manipulators getting egg on their face.
> (Without that populace, we're up the creek anyway.)
> 
> But in the case of nomcom, gaming can go undetected or be impossible to 
> expose in this way - BECAUSE of the confidentiality rules.
> 
> If the publication of names makes some kinds of gaming visibile to the IETF 
> community, I think we've gained something, not lost it.

Something else to consider is that as long as nomcom as a certain
amount of discretion (which we can give them, since if we can't trust
the nomcom, the process is sunk anyway), the gaming problem can also
be addressed by allowing the nomcom decide that someone who has
flooded the nomcom with thousands of names can be safely ignored, thus
(a) saving the nomcom from having to ask everyone if they want to be
considered, and (b) those names would therefore not have to appear on
the list.

Preventing denial of service attacks is hard in protocols because
computers don't have the judgement that humans about deciding when
they are under a DOS attack.  But humans can easily decide that if
they are getting bombed with suggestions, even if they are coming from
different sources, that a DOS attack is taking place, and as long as
we give the the appropriate amounts of discretion, we can trust them
to take appropriate action --- possibly up to and including making the
public that someone has attempted to game the system by flooding the
nomcom with all possible IETF members as suggestions, for example.

So Dave's argument that the system can be gamed in this particular
argument is at best just a reminder that we need to give the nomcom
appropriate amounts discretion, and at worst, completely specious,
since I don't believe the problem is would be something that we would
really need to worry about in real life.

						- Ted