Yesterday I posted a message that said that I agreed with the IPv6 working
group chairs that rough consensus was reached to deprecate IPv6 site local
addresses. That said, I do have an issue on the discussion that led up to
that consensus decision. I do not think there was much of an actual
discussion on the topic.
The working group chair's presentation on the site local options listed
five options for the working group moving forward in regards to the site
local question. These options ranged from eliminating site local addresses
to fully embracing the concept and working out all the details of how to
use them. But they only discussed the middle three options. They reported
that the consensus in the Atlanta meeting was to not support outright
elimination or full embrace so those options were not included in the
chair's presentation of the advantages and disadvantages of the various
options.
The discussion during the chair's presentation basically did not touch on
the pros and cons of having site local addresses per se - a few 'they
should just go away' statements were made but no exploration of the issues.
The open discussion after the presentation also did not explore the issues
but there were a greater number of people who felt that SL addresses should
be eliminated from IPv6.
As I mentioned in yesterday's note - Thomas and others noticed the
sentiment against SL and the chairs wound up asking the question they did
(about deprecating SL) as a result.
But, for what it's worth, I do not think that there was sufficient
discussion of the option of deprecating SL addresses before the consensus
check was made. So, in a way, I think the consensus was wrongly reached,
even if I agree that consensus was reached.
At this point I do not know what to suggest be done. As I will explain in
another note, I am in the minority on this topic but even so I'm not sure
that much good would come from readdressing the issue. It may be best to
stick with the consensus for now and only look at the issue again if it
becomes clear that the future usefulness of IPv6 depends on the IETF
understanding the use of private addressing in IPv6, and that may never
happen.