At 3:53 PM -0600 3/27/06, Arvel Hathcock wrote:
The following point is against splitting the document:
(a) Less convenient for implementors and those who need to reference
the documents
The following points are in favor of splitting the document:
(a) It can help speed up the "process" by deferring controversial
topics for later
(b) It has the benefit of allowing easier revision in the future
There are costs and benefits for both sides, of course.
The WG should *not* want to speed up getting the normative base spec
out by deferring controversial topics to the non-normative overview.
This is a recipe for disaster, one that has been seen in many other
IETF WGs. We will end up arguing about what we "meant", and what we
"should have meant", and thereby greatly delay the document that will
be of most use to users of the protocol (as compared to the
implementers).
As I said at the mic last week, I think we should have a single
document that contains the protocol and the discussion around it.
That document does *not* have to contain deployment ideas (such as
Appendix B) or suggestions for signer and verifier policies, but
should be good enough for a developer and an early deployer to create
an interoperable implementation by reading the document.
_______________________________________________
NOTE WELL: This list operates according to
http://mipassoc.org/dkim/ietf-list-rules.html