Re: [v6ops] The need for local-ipv4 socket transition solutions -- NAT64/DNS64 remains insufficient

On Tue, Mar 24, 2015 at 8:06 AM, James Woodyatt <jhw@nestlabs.com> wrote:

> On the other hand, I would be disappointed if such a consensus were to
> emerge. It would mean that we would never be able to sunset IPv4, and as an
> endpoint software developer, if every host I want to deploy on has a CLAT
> provided by the system, then I'd observe that my incentives to stop writing
> IPv4-only software logic would evaporate for all practical purposes.
>
> I expect my opposition would be trammeled by bulldozers, but I would
> oppose.
>

Rev'ing up the bulldozer (puffs of smoke fill the air)

I agree with your principles, and i believed in them too at one point, but
i became painfully aware that this view is grossly misaligned with
reality.  Once abandoning this view and accepting that i cannot boil the
IPv4 ocean, i experienced a great deal of internal peace as i took control
of my internet numbering destiny.

As we know, in mobile, there are 3 dominate operating systems.  Each and
everyone of them ships with fundamental software -- core functionality --
that cannot operate without IPv4 sockets.  Beyond the IPv4 dependent
software that ships on a clean build, the apps stores are littered with
IPv4 dependencies -- in top 100 apps.  The typical number in the app store
is 20% of the top 100 fail on IPv6 + NAT64/DNS64, recently re-verified by
myself.

Even if these amazingly-wealthy talent-stuffed software factories were able
to clean up their own ipv4-dependent software, and painfully thrust
standards on 3rd party app makers, they cannot clean up the world wide web
(as noted in the first message of the thread).  They cannot clean up
enterprise VPNs.

And no, no ISP / Mobile provider is willing to accept 1% of the web
breaking  (including Alexa top 100 web pages) while their competitors
deliver 100% of the web.

Shorter Cameron:  Failure to deploy a proper IPv6-only solution to the host
is creating real and meaningful harm to IPv6 deployment at large.  The
network that i work at has hit a ceiling on IPv6 deployment due to a single
major host operating system lacking this basic functionality (Windows Phone
and Android support 464XLAT, deployed to 10s of millions of users who have
never heard of IPv4).

IPv4 dependence is hurting the internet and allows for middle-box
proliferation.

So, perhaps a BCP from the IETF would make it clear what is best and what
is not-best.

CB

>
> On Tue, Mar 24, 2015 at 9:13 AM, Lorenzo Colitti <lorenzo@google.com>
> wrote:
>
>> On Tue, Mar 24, 2015 at 9:11 AM, Ca By <cb.list6@gmail.com> wrote:
>>>
>>> Oh, I see why you started this thread now. Do you expect that the IETF
>>>> would be able to reach consensus on such a statement?
>>>>
>>>
>>> Yes. Are you in opposition?
>>>
>>
>> Oh, of course not. I'm the maintainer of a popular 464xlat
>> implementation, remember?
>>
>> _______________________________________________
>> v6ops mailing list
>> v6ops@ietf.org
>> https://www.ietf.org/mailman/listinfo/v6ops
>>
>>
>
>
> --
> james woodyatt <jhw@nestlabs.com>
> Nest Labs, Communications Engineering
>
> _______________________________________________
> v6ops mailing list
> v6ops@ietf.org
> https://www.ietf.org/mailman/listinfo/v6ops
>
>