Re: Confused on "IETF Consensus"

["Theodore Y. Ts'o" <[e-mail]>]

   Date: Wed, 12 Jun 1996 18:22:58 -0700
   From: Fred Baker <[e-mail]>

   Flip side of that is that, quite frankly, those who have been involved in
   writing the document are in most cases the best technical staff we have on
   that particular job, and the ones that are going to live with the
   consequences of their consensus (assuming it indeed exists). Those
   individuals have "paid their dues" so to speak; they have gone through the
   effort to do the Right Thing, and Get It Right. They have documented,
   implemented, argued, interoperability tested, and in some cases spent
   nights on their customer's premises making sure it worked for real people
   who are really depending on it. They have frequently put not only blood,
   sweat, and tears into it, they have put significant money in as well.

While this works well for technical issues such as defining a protocol,
it's not clear how well it works when the same "best technical staff"
are inventing policy.  The problem is that the wg may not include
everyone who must "live with the consequences of the decision" in the
wg, since the IETF in general has a disproportionate number of engineers
over normal network users --- not surprising, really, given our charter.

To give an example other than the one which sparked this whole
discussion: Suppose you have a wg where one or two developers decide
they wanted to be lazy and allow the use host tables in IPV6
implementations.  This would have made a complete hash of the
requirement it be easy to renumber the addresses IPV6 hosts.  

The problem was that people who would have to live with the consequences
of that decision -- the users --- weren't represented in the working
group!  As I understand the history, this particular controversy had to
bubble all the way up the IESG before it could be resolved sanely.

The bottom line is that working groups *do* need checks and balances,
becuase often times the interests of the implementers --- which are the
majority of the people in most well-functioning working groups --- do
not always reflect the best interests of the IETF or the Internet as a
whole.

It would be wise for us not to forget this.

						- Ted