On Jul 21, 2015, at 7:48 AM, Sam Hartman <hartmans-ietf(_at_)mit(_dot_)edu>
wrote:
However, efficiency of implementation is a concern it's reasonable for
the IETF to value and we typically value it highly.
Being able to implement things like TOR on top of sox shims, and being
able to implement hidden services on top of SOX shims is very valuable.
It allows us to add support for hidden services without modifying the
end applications.
Absolutely. However, it is perhaps worth noting that there is a long-standing
solution to the problem that doesn’t require socks: nsswitch.conf. It’s not
the right architectural solution either, for a couple of reasons: not an
appropriate UI for non-hackers, and a bit too dependent on the list of things
switched being small. But the point is that this is actually a pretty
well-understood problem, and if, as a policy, we continue to add special-use
names as required, the solution to the problem of how to handle these
special-use names in the host stack is already well understood.