[arin-ppml] The role of NAT in IPv6

[David Farmer]

Thank you Joe,

I believe this is a much more appropriate attitude to base address 
policy regarding NAT and IPv6 on.

Joe Maimon wrote:

> nat = napt66;
> 
> I am unconvinced that stub/end/residential/consumer user deployment via 
> DHCP-PD or whatever iteration of the solution to the problem of 
> extending routing to these classes of user's device(s) (which to be 
> precise is a form of inter-autonomous system routing) will prove more 
> satisfactory and successful when compared to the complete workaround to 
> the problem (while creating others) that nat allows for.
>
> The market, consisting of users, providers, manufactures and default 
> "default settings" will decide this one, as they did for IPv4.
> 
> Eventually.

Yes, the market should decide this, not a policy fiat from ARIN or even 
the IETF.

I think a NAT66 device, requesting via DHCP-PD or being statically 
assigned at least a /64 or maybe in some cases a larger block and doing 
NAT addresses to address translation, not NAPT, could be very 
interesting.  Such a solution could be useful in a number of situation, 
and in that case the use of NAT has nothing to do with address conservation.

Embracing NAT as a possible solution, rather than treating it as 
something to be exterminated, could lead to some interesting solutions.

> However, it is a fact that there are many people who continue to assert 
> that nat has positive security implications for them.
> 
> And others have posted still other motivations for nat that even readily 
> available globally unique addresses may not obviate to the complete 
> satisfaction of potential nat users.
> 
> So while I expect that nat utilization will almost certainly lessen from 
> its current state of near universal presence, perhaps to the point of 
> rarity (at least for the end user masses), I neither believe it is fated 
> for extinction, nor do I believe behaviors assuming or intending 
> extinction are prudent.

Yes, one can hope that since IPv6 no longer has a need to conserve 
addresses, that the near universal presence of NAT will fade.  And, only 
those that have some other motivations to use NAT will inflect its costs 
on themselves and the rest of us.

> I do believe it is worthwhile to invest in policies and technologies 
> that remove or lessen motivations for nat. I do not consider my views 
> contradictory.

I agree it is a worthy policy goal to actively encourage non-NAT 
solutions.  However, it is important for that be done through true 
incentives for non-NAT solutions and not by creating artificial 
disincentives in policy for NAT solutions, because a majority dislikes 
NAT.

Again, thank you.

-- 
===============================================
David Farmer               Email:farmer at umn.edu
Networking & Telecommunication Services
Office of Information Technology
University of Minnesota	
2218 University Ave SE	    Phone: 612-626-0815
Minneapolis, MN 55414-3029   Cell: 612-812-9952
===============================================