[arin-ppml] Controlling the IPv6 address consumption rate

[William Herrin]

On Fri, Oct 15, 2010 at 1:51 PM, Heather Schiller
<heather.skanks at gmail.com> wrote:
> On Wed, Oct 13, 2010 at 11:16 AM, William Herrin <bill at herrin.us> wrote:
>> At the same time (maybe even in the same RFC), the IETF decided that
>> end users would not multihome using BGP with multiple ISPs under IPv6.
>> That too is the official standard.
>
>   I don't believe IETF ever "decided that end users would not
> multihome using bgp with multiple ISP's"   It was more like they
> repeatedly tried to create an alternate method of multihoming that did
> not rely on *deaggregation*

Hi Heather,

And we're still trying. I spent the last three years in the IRTF RRG
and we still don't have either a consensus approach or a technology
which can be proven to do useful work in even the corner cases. There
are some really clever ideas there with some real promise, but not
enough of them are being built and tried to answer some of the
identified unknowns and lock in on what might work.

> But then we
> passed the end user IPv6 assignment policy and instantly lost the much
> of the motivation to complete that work.

I've harped on it before, but if we had a pool of cheap, non-need
based registry-assigned addresses that were explicitly and
intentionally barred from BGP, the combination of all the kiddies
wanting PI and the availability of PI addresses might restore some of
the motivation to develop a tech capable of using them.


>> We always listen to what the folks in the IETF have to say, but
>> sometimes they don't know what the eff they're talking about.
>
>  Maybe rather than just listening to what the folks in IETF have to
> say, operators should participate more, in order to develop solutions
> that are operationally feasible.  Be a part of the solution.

In spirit I'm with you 100%. Pragmatically though, the number of folks
who enjoy both development and operations is a lot smaller than the
number who enjoy one or the other.


On Fri, Oct 15, 2010 at 2:44 PM, Joel Jaeggli <joelja at bogus.com> wrote:
>> On Wed, Oct 13, 2010 at 11:16 AM, William Herrin <bill at herrin.us> wrote:
>>> We always listen to what the folks in the IETF have to say, but
>>> sometimes they don't know what the eff they're talking about.
>
> I'm not convinced that you do listen, or that if you do listen that you
> agree.

Joel,

I can set your mind at ease for the second part. For the former,
you'll have to decide for yourself whether I made an adequate effort
to understand your viewpoint before disagreeing with it.


> It's pretty
> easy to characterize the activities of 15-18 years ago as deluded or
> misguided, it seems to be harder for us to accept that what was done was
> as good as it could be under the circumstances and that we need to move
> on changing what we have in minor ways rather than boiling the baby (who
> is now a teenager in any event) in the bathwater.

We should change things exactly as much as experience and a healthy
caution reveals is needed. If you did as a good a job on IPv6 as was
done on IPv4, you can expect as little change as IPv4 underwent. Did
you? And when we think back, just how minor were IPv4's needful
changes like CIDR and NAT?

Yesteryear's best decisions were rarely deluded or misguided.
Following them uncritically today can be.

Regards,
Bill Herrin



-- 
William D. Herrin ................ herrin at dirtside.com  bill at herrin.us
3005 Crane Dr. ...................... Web: <http://bill.herrin.us/>
Falls Church, VA 22042-3004