Maybe I'm unduly pessimistic, but I would have trouble working up a lot of
enthusiasm for doing work on a project if I knew that some gatekeeper
working in parallel might later tell me that I was wasting my time.
I have trouble working up a lot of enthusiasm for contributing to work that is
going off in the weeds. What I have in mind is not a "gatekeeper" but external
review. The external review would not be expected to tell a WG that it were
wasting its time (though that is a possibility), but rather, how to make better
use of its time by producing a specification that was more relevant.
Apparently you think that an artist is the best judge of the relevance of his
Although I completely disagree with your prediction that DKIM will be
useless, there are worse problems than standards that turn out not to be
used, so long as they are designed so they don't interfere with other
more useful work. We seem to be managing that last issue reasonably well.
DKIM is a good counterexample of that also. The existence of DKIM distracts
attention from more useful work that could be done - partly because it's
consuming energy from those who would work on more useful goals if they were
chartered, partly because of the need for damage control, and partly because of
the widespread assumption that since IETF has chartered DKIM that DKIM is "the"
solution that will be promoted by IETF.
Ietf mailing list