On Jul 4, 2017, at 1:09 PM, John Levine <johnl(_at_)taugh(_dot_)com> wrote:
Special use TLDs are the North Korea of the IETF. Everyone agrees
they're a problem, but there's no agreement about what to do, with
attitudes ranging from resigned acceptance to aggressive
countermeasures.
The countermeasures all have problems. Some make implausible
assumptions like knowing in advance where all the missile launchers,
er, stunt resolvers are. Others would take strong action that would
produce a toxic rain of collateral damage on their allies.
Actually, one of the reasons we did a problem statement document and not a "why
are these useful" document as Randy has suggested is that there is no consensus
even that special-use names are a problem. A lot of people think that it's
gTLDs that are the problem, not special-use names: gTLDs, by allocating names
that ought to have been special-use names, have created a serious problem in
that we can no longer use special-use names freely, and that not all uses of
special-use names remain possible.
I don't mean to open a discussion into that topic here—I realize that there is
no consensus on it, and that this position is not shared by various people,
yourself included. I just want to point out that you've expressed it as
universally true that people think special-use names are bad, and that is not
in fact the case: if it were, we wouldn't need to publish this document.