Folks,
The only thing I'd add is to ask folks to also bear in mind
our charter [1] deadlines when suggesting new document
structures. ...
A split is guaranteed to cause that date to slip, fullstop.
Although we certainly need to pay attention to impact on the schedule, the
amount of impact is significant.
Given how many months it took to get this working group chartered, and how much
time we have had to spend on the threats document, the idea that we would be
concerned about a very minor slippage would be rather strange, if that slippage
buys some strategic advantage. Yes, that is a big "if".
Partitioning of documents is considerably more than a bookkeeping exercise in
abstract aesthetics. It has a fundamental impact on clarity and schedule, and
often produces components of specification that are far more crisp.
It also permits being much more clear about critical dependencies and often can
permit reducing them, for any initial deliverables of basic utility. It permits
deferring complex or secondary topics, taking them out of the critical path for
initial utility.
Too few documents is at least as bad as too many. I actually suspect that it is
worse, given how bogged-down larger, more complex documents tend to get. As we
have already seen, a particular challenge for a diverse group trying to specify
an "interesting" system, is focus. Larger, more complex documents tend to
encourage distraction and lack of discussion focus.
d/
--
Dave Crocker
Brandenburg InternetWorking
<http://bbiw.net>
_______________________________________________
NOTE WELL: This list operates according to
http://mipassoc.org/dkim/ietf-list-rules.html