Re: prerequisite for change (was Re: draft-housley-two-maturity-levels)

[Dave CROCKER <[e-mail]>]

On 1/30/2011 8:06 AM, Andrew Sullivan wrote:
> On Sun, Jan 30, 2011 at 07:49:44AM -0800, Dave CROCKER wrote:
>> The current proposal specifies a second maturity level that does not
>> permit changing the technical specification.
>
> Yes, I know.  I fail completely to see why anyone would ever do the
> work for such movement of maturity level.  The proposal seems to me to
> be something along the lines of giving gold stars to protocols with
> people who are willing to do the busywork.


One of the challenges in the discussion of this topic on the IETF list is some 
very different models of the role of a standards label.  Given the frequent view 
that only technical details matter, one would think that we are composed only of 
engineers who give little thought to organizational, social and psychological 
factors that might also affect adoption decisions...

One curiosity about this is that our documentation style for specifications 
permits far more tutorial and explanatory content than most other standards 
group, which one might take to indicate that we actually worry a bit about those 
who will adopt our work and that we actually want to help them.  Yet our 
organization mode tends to consider 'publication' the end of our concern.

As folks in the communications game, the IETF tends to have a curious lack of 
interest in the 'feedback' portion of a Shannon-Weaver model[1], when it comes 
to protocol adoption.  It's as if we stopped reading after the first Shannon 
paper.[2]

In any event, I thought the value proposition of the proposal's second label had 
been discussed at some length:

    For significant portions of industry, the decision whether to adopt new 
technology relies heavily on an assessment of its stability and likely success. 
  Some specs are nascent and fluid.  Some are failed.  Others are successful. As 
of now, it can be difficult for such folk to distinguish the real maturity of 
IETF proposals, since most are at Proposed.

    So the proposal defines a second stage that serves the purpose of making an 
official statement of stability and success, by virtue of noting significant 
deployment.  That can be helpful during the "market expansion" phase of adoption.

d/


[1] <http://en.wikipedia.org/wiki/Shannon%E2%80%93Weaver_model>

[2] <http://plan9.bell-labs.com/cm/ms/what/shannonday/shannon1948.pdf>
-- 

   Dave Crocker
   Brandenburg InternetWorking
   bbiw.net