Filter-based routing table management (was: Re: minimum IPv6 announcement size)

[jcurran at istaff.org (John Curran)]

On Sep 26, 2013, at 1:43 PM, Randy Bush <randy at psg.com> wrote:

> y'know, it's funny.  there is a market in ipv4 space.  there is no
> market in routing table slots.  perhaps this is not conspiracy but
> rather the market is speaking.

That easily could be the case.  So how well is has the current model
been working out for IPv4?  It appears that only feedback mechanism
is the "The Cidr Report" <http://www.cidr-report.org/as2.0/#Gains>, 
and last time this topic came up was almost exactly two years ago 
back at 375000 routes... is this success?  Perhaps more conference 
banners are needed?

> of course, we can use the idea of a market in routing table slots,
> rack space, or coffee to distract from the artificial problems in
> the only actual market, ipv4 address space.

No desire at all to distract from the discussions on that topic, and 
in fact, I'd encourage folks (including yourself) to pop on over to 
PPML and make suggestions for policy changes as desired.  That's not
an option available to me, and I was simply commenting on the fact 
that we're recreating the same IPv6 routing table feedback system 
which gave us our present "success" with the IPv4 routing table.

FYI,
/John