Re: draft-bourbaki-6man-classless-ipv6-00

[Mark Smith <[e-mail]>]

On 4 June 2017 at 11:33, Lorenzo Colitti <[e-mail]> wrote:
> On Sun, Jun 4, 2017 at 2:34 AM, Roger Jørgensen <[e-mail]> wrote:
>>
>> > Please stop trying to make IPv6 be the same as IPv4. That will take away
>> > our
>> > ability to make the Internet better once IPv4 is gone.
>>
>> oh totaly agree, but it's already lost, the "operators" don't want to
>> move/change.
>
>
> But they *are* changing. For example: almost 20% of Google users these days
> come in over IPv6. Pretty much all of them are connected to networks that
> support SLAAC and use /64 prefixes. That's a huge number now matter how you
> look at it, and it's showing no signs of slowing down.
>

I think the mistake people are making is that they're viewing lack of
enterprise IPv6 adoption to indicate a lack of any IPv6 adoption. It's
an easy mistake to make, I often do it myself.

IPv6 isn't solving business problems that most enterprises have, which
that is why it isn't being adopted by them. That's ok, technologies
aren't adopted by businesses for technologies sake.

It is solving the business problem that residential service providers
have, which is why they're adopting it so quickly. Specifically, the
problem it is solving for them is it is minimising the amount of
Carrier Grade NAT capacity they'd have to buy.

Some other enterprises are using it, I know it is being used in
Advanced Metering Infrastructure (i.e. electricity smart meter
networks).

Even when an enterprise encounters an IPv6 only website that they need
to access, their solution will be to put in an IPv4/IPv6 proxy on the
edge of their network, as that is consistent with their private
internal addressing / middlebox/security box on the edge of the
network model.

Regards,
Mark.