(Removed Cc: iesg)
On Mon, 20 Sep 2004, Harald Tveit Alvestrand wrote:
--On mandag, september 20, 2004 14:38:51 -0400 Michael Richardson
<mcr(_at_)sandelman(_dot_)ottawa(_dot_)on(_dot_)ca> wrote:
Harald> And - here I am making a real leap of faith - if the IETF
Harald> recommendations for NAT devices make manufacturers who
Harald> listen to them create NAT devices that make their customers
Harald> more happy, then many of these new NAT devices may be
Harald> conformant to IETF recommendations.
Do we really want customers of NAT devices to be happy?
Given that I'm one of them, and will continue to be one until the IPv4
Internet fades to where I can ignore it.... yes.
The point is not whether the users behind an IPv4 NAT are happy or
not.
The point is which kind of applications you can reasonably expect to
deploy behind an IPv4 NAT, and be happy.
I agree with Harald that v4 NATs are going to be here a decade from
now. But that's irrelevant, if those people using the NAT only use
simple client-server applications.
What matters is the peer-to-peer, etc. applications which typically
require solutions like STUN, TURN, Teredo, etc.
My argument is that we can simplify the architecture significantly if
we can assume that v4 NAT can be treversed using one particular
mechanism (e.g., Teredo), and all the applications which have a more
complex model than just client-server are just recommended to use IPv6
instead.
If we can provide a reasonably working IPv6 connectivity solution(s),
we wouldn't have to try to figure out how to make NATs behave better,
how to build robust applications to work with these nicely or badly
behaving NATs, etc.
--
Pekka Savola "You each name yourselves king, yet the
Netcore Oy kingdom bleeds."
Systems. Networks. Security. -- George R.R. Martin: A Clash of Kings
_______________________________________________
Ietf mailing list
Ietf(_at_)ietf(_dot_)org
https://www1.ietf.org/mailman/listinfo/ietf