applicability statement vs technical specification

On Sun, Oct 05, 2003 at 12:12:41PM -0400, Alan DeKok wrote:
| 
|   My opinion is that the document should discuss the detailed design
| of the system *last*.  e.g. A contrary example is that SPF has a short
| introduction, and then gets right into design.  The introductory
| paragraphs should be expanded to address detailed benefits and
| limitations of addressing the problem statement in the first paragraph
| of Section 1.

Perhaps we should put forth an applicability statement separate from the
technical specification.  I focused on design because an RFC is not the
place for rhetoric; it is a specification, not an exhortation.

|   We should start off with a statement of the problem, along with the
| benefits and limitations of any method of to address that problem.
| This will help limit the opposition to the proposal, as it's problems
| will be stated up front.
| 
|   I hope that beginning the document that way will help avoid some of
| the arguments seen on the main ASRG list, for some of the proposals.

Whose opposition are you concerned about?

I don't see the adoption process as a story of convincing people that
we're right; I see it as a story of giving what some of the people want
to those people that want it.  After all, this is a voluntary standard.
The people we need to convince are the MTA developers and the major
ISPs, not the nitpickers and edge cases.  Debate with them goes past the
point of diminishing returns.