Re: [BEHAVE] RFC6147 and RFC7208 interoperability issues

[Keith Moore <[e-mail]>]

On 2/7/22 12:18, David Conrad wrote:

> You are aware of what the BEHAVE working group's charter was, right?
yes.   Perhaps a noble attempt that failed, perhaps an errand for 
fools.  At the time BEHAVE was chartered I thought there might be a way 
to make NATs saner by making applications explicitly aware of them and 
letting applications explicitly control NAT bindings.   In hindsight I 
think it was far too late to fix NATs by then, but maybe it never was 
possible anyway.
> As far as I can tell, Klaus has identified a specific issue in which existing IETF specifications are deficient in order to facilitate interoperability in a specific protocol when NAT64 is in use.

To the extent that the specifications are deficient, it's because the 
approach was inherently deficient.

> Reiterating yet again how you think NAT is a bad idea does not appear to be constructive or helpful in any way.
Not buying it.    The problem with that attitude is that it makes it 
always okay to promote and encourage bad ideas, and to perpetually keep 
patching up the harm that they cause... but it's never okay to 
discourage them.
> The need to deal with NAT (in its myriad forms) is simply reality and will remain so for the foreseeable future.

Saying "X is reality" is a lot like saying "it is clear that..." or any 
other assertion that lacks support.   It's transparently vacuous.

NATs are reality like COVID is reality.  That's not a justification for 
encouraging more spread of COVID.

Keith