[182090] in North American Network Operators' Group
Re: Dual stack IPv6 for IPv4 depletion
daemon@ATHENA.MIT.EDU (Tim Franklin)
Fri Jul 10 04:49:05 2015
X-Original-To: nanog@nanog.org
Date: Fri, 10 Jul 2015 09:48:56 +0100 (BST)
From: Tim Franklin <tim@pelican.org>
To: nanog@nanog.org
In-Reply-To: <57B47422-2B92-4C74-A3C8-FF2D980703F7@delong.com>
Errors-To: nanog-bounces@nanog.org
> And I=E2=80=99m saying you=E2=80=99re ignoring an important part of reali=
ty.
>=20
> Whatever ISPs default to deploying now will become the standard to which
> application developers develop.
>=20
> Changing the ISP later is easy.
I'm not even convinced of that. Once "/56" (or *any* value) is baked into =
the processes, hard-coded in systems all over the shop, assumptions made le=
ft, right and centre throughout the business, changing it will be hard. On=
ly the tech part of changing the ISP is easy.
It's the same reason it's so difficult to get IPv6 moving in some ISPs. Ma=
king the kit dance the appropriate jig in (modulo a few Luddite vendors and=
legacy kit that Just Won't Die) is quite straight-forward. Getting IT to =
update a field to not force [0-9]{1,3}.[0-9]{1-3}.[0-9]{1,3}.[0-9]{1,3}, or=
to add a new field, without a revenue stream attached is *hard*.
(Yes, it's part of our job as technologists to explain why we should do it =
anyway. That doesn't stop it being hard.)
Regards,
Tim.
