Re: [v6ops] draft-palet-v6ops-nat64-deployment-02 comments

On Mon, Jul 16, 2018 at 5:03 AM Brian E Carpenter <
brian.e.carpenter@gmail.com> wrote:

> Fred,
>
> On 16/07/2018 23:47, Fred Baker wrote:
> >
> >
> >> On Jul 15, 2018, at 9:52 PM, xiechf.bri@chinatelecom.cn wrote:
> >>
> >> One of the debate about NAT64 is that it will slow down the transition
> process of IPv6, for it allow the existance of IPv4, especially in the
> content side, the OTT, especially smaller ones,  will have less incentive
> to move to IPv6. I do not mean that this point is right,  but this draft
> should give some explanation or clarification to this issue.
> >
> > To my way of thinking, which may or may not be correct, that can be
> argued either way. The existence of stateless and stateful translation has
> enabled 464XLAT to be used to make some networks IPv6-only (the CLAT being
> stateless and the PLAT being stateful) although the entire world has not
> switched. I would expect that few would even consider turning IPv4 off
> without some means of accessing IPv4-hosted content. So the existence of
> translation hasn't slowed that rush; it has enabled it. On the other hand,
> if IPv4-hosted content is still accessible, there is also no pressure to
> make it IPv6-accessible.
>
> That's true, but it applies regardless of whether the client accesses the
> IPv4 content via a dual stack ISP or a NAT64 solution. The only way to
> force a content provider to use IPv6 is by removing its IPv4 clients.
> I don't think many consumer ISPs are going to do that.
>
> So what we need is for IPv4 to be slower and less functional than IPv6,
> which we can only reasonably arrange by making IPv6 better...
>
>     Brian
>

While setting out with the objective of making a customer facing service
less good is impractical in a real business, the situation is that
nat64/dns64 engenders indirection and session creation latency , while
native ipv6 goes directly.

To that end, nat64 deployment encourages end to end ipv6

> _______________________________________________
> v6ops mailing list
> v6ops@ietf.org
> https://www.ietf.org/mailman/listinfo/v6ops
>