[wg-c] straw poll results

[Jonathan Weinberg <weinberg@mail.msen.com>]

Straw poll results

	29 people voted in the straw poll.

	To start with the raw scores:  In question one, five voted for option one
(usually with caveats); none for option two; thirteen for option three;
eight for option four; and three for "other" (this category includes people
who had voted for one of the options, but who wrote messages to me
explaining their votes making it clear that they hadn't intended those
options or making it impossible to tell what they intended).  In question
two, five voted for option one; two for option two; thirteen for option
three; six for option four; and two for other.  In question three, four
people voted for option one; seven people for option two; one person voted
for option three outright and two more explained that they favored this
option for the initial rollout only; three people voted for option four;
two people voted for option five outright and two more favored this option
for the long term only; and ten for some form of "other."

	Some more useful results:

	(1) There was a strong majority in favor of the proposition that ICANN
should play some part in choosing the new gTLDs.  Nineteen people
subscribed to some variant of this proposition, with eight urging the
contrary proposition (that ICANN should choose registries wholly without
regard to the gTLDs that the registries propose to run, and then should
allow the registries to pick their own names and associated charters), and
two splitting the difference.

	(2) There was a majority in favor of the proposition that the initiative,
in selecting the new gTLDs, should come from the registries themselves.
For the initial rollout, seventeen people subscribed to some variant of
this proposition, with eleven urging the contrary proposition (that ICANN
should first select the new gTLDs, and later solicit applications from
registries to operate those gTLDs), and one being unclear.  For the long
term, nineteen people agreed with some variant of the proposition that the
initiative, in selecting the new gTLDs, should come from the registries
themselves, and nine urged the contrary.

	These two results suggest that the only recommendation that has a chance
of winning rough consensus in the WG is the one recently urged by Sheppard
and Mueller, under which registries apply describing their proposed TLD,
and an ICANN body or process makes selections taking into account the
characteristics of both the registry and its proposed TLD.  This result did
not attract a majority of those who responded.  Only about ten people
supported it for the long- term, and eight for the initial roll-out.  But
this proposal is the compromise middle ground between those who would have
registries choose gTLD names and charters without ICANN involvement, and
those who would have ICANN choose gTLD names and charters before soliciting
applications from registries.  The straw vote suggests that if we are to
reach any form of compromise rough consensus on this issue, this proposal
is how we will do so.  I am going to issue a consensus call urging that the
group adopt this compromise position, which is not *too* far from that of
any of the respondents, and I strongly urge all of you to support it.  We
need to wrap this process up.

	(3) There was a strong majority in favor of the proposition that the
process should have room for both limited-purpose TLDs (which have a
charter that meaningfully limits who can register there) and
general-purpose TLDs (which have an "open" charter that does not
significantly restrict registration in that TLD, or, perhaps, have no
charter at all).  For the initial  rollout and the longer term, at least
twenty people endorsed this proposition (answering either that ICANN should
select such a mix or that the registries should choose their own names and
charters, which would in turn ensure such a mix).  I'm going to issue a
consensus call on this point as well.

Jon