Keith Moore wrote:
> ...
> bottom line: the Atlanta poll was essentially meaningless 
> except to indicate that there's a significant fraction of the 
> group that wants to discourage SLs to some extent.
>

I agree. The question that wasn't asked is why they want the
restriction. Of the group that indicated a desire to restrict use of SL,
how many of those were based on the simple fear of change, the unknown,
or that SL automatically leads to nat? 

To be fair, I suspect that the number of participants that have actually
worked through the applications and infrastructure that work or not is
about the same for either side, and collectively <10% of those
expressing an opinion in Atlanta. 

SL was not intended to be a general purpose addressing mechanism, and
trying to use it as such will have failure modes. We know that large
network managers will insist on being independent of their provider(s),
and we know that routing that independence is a hard problem. That is
the reason that multi6 was tasked with focusing on that specific issue,
without distraction. Unfortunately, that group has gone into hibernation
for extended periods of time, and with the latest emergence seems more
interested in architectural changes that affect all applications, rather
than engineering a manageable multi-homing approach.

Personally I want to see us make progress on this front, and I don't
care which WG gets it done. The IESG probably has a different opinion,
so it may be challenging to get documents out of draft state.
Fortunately we can do the work and worry about getting it to RFC later.

Tony




--------------------------------------------------------------------
IETF IPng Working Group Mailing List
IPng Home Page:                      http://playground.sun.com/ipng
FTP archive:                      ftp://playground.sun.com/pub/ipng
Direct all administrative requests to [EMAIL PROTECTED]
--------------------------------------------------------------------

Reply via email to