--On Monday, July 27, 2015 2:34 AM +1200 Brian E Carpenter
<brian.e.carpenter@xxxxxxxxx> wrote:
>...
> But it was always clear that the IETF
> kept the power of decision for assignments of domain names for
> technical uses, which has nothing to do with string lengths.
At the risk of hair-splitting, it was always clear that the IETF
always had (and kept) the power of decision over name properties
such as characters to be used. Lengths were another matter and
probably fell under the same general policies that resulted in
decisions about particular parameter values being treated as
requests. In particular, IANA made the decisions about what
names or parameter values were actually allocated, not the IETF
or the requester.
However, until circa 1998 - 1998, the IETF _never_ had the
"power of decision" over what went in the root, technical or
otherwise. RFC 1591 was never an IETF document. It was an IANA
document for community information, one that described
IANA-developed procedures and plans. The IETF was never
involved in, e.g., ccTLD assignments. Had the IETF asked (or
"decided") that additional TLD names (whether shorter or longer)
be delegated for technical purposes, I don't know what would
have happened, but that statement from 1591 about anticipation
of new gTLDs may constitute a hint.
That model started changing in the last 1990s and changed
considerably as ICANN took over and the staffing and complexion
of IANA changed. In particular, the IETF started assigning
specific values and directed IANA to put them into registries
and started reserving DNS names, notably the "example" set. The
snit with ICANN staff that resulted in RFC 3172 could be seen
from the IANA perspective as an attempt to assert the authority
IANA had before ICANN came into being but we viewed it as
interference in authority the IETF and IAB acquired as part of
the transition to ICANN.
john