I totally agree. There is far too much focus on nailing down %100 of
requirements in the recent IETF past, hindering progress on solutions. Both
implementations and requirements are refined by the iterative process of
implementation and deployment feedback.
Tom
On Apr 3, 2015, at 1:03 PM, Benson Schliesser
<bensons(_at_)queuefull(_dot_)net> wrote:
Phillip Hallam-Baker wrote:
Another part of the problem is that often pre-WG lists are formed on
the basis of a solution rather than a set of requirements.
On the other hand, a solution implemented in running code helps the process.
If our goal is interoperable specifications (e.g. standards) then it's good
to use such solutions as a basis. There still is value in understanding the
context (requirements, use-cases, etc) but IMHO we should remember that those
are ancillary.
Cheers,
-Benson