[6bone] Re: RFC 2772 input from RIR space holder

[Tim Chown]

On Fri, Nov 22, 2002 at 12:55:12PM +0100, Jeroen Massar wrote:
> 
> Apparently you don't know that you can BUY IPv6 transit and connectivity
> in many different parts of the world (Actually I think almost
> everywhere).
> 
> You might contact:
>  - VERIO
>  - NTT / IIJ
>  - GlobalCrossing
> 
> and many others, as for endpoints, in Japan you can get a native IPv6
> only line if you want it. That they don't sell it to your house yet
> doesn't mean it doesn't exist. In the Netherlands one can currently
> get Xs4all's PowerDSL which gives you native IPv6 and IPv4 on ADSL.
> Chello has native IPv6 for most parts of europe as far as I understood.

All this is very nice, but it doesn't mean stable, production quality IPv6
networking on an international basis.  Buying it is one thing, being able
to use it is another.   Finding paths that perform as well for IPv6 as for
IPv4 is very rare.  I suggest you try some globetrotting to find the
reality :(
 
> So YES there is REAL commercial activity. And these people want it
> to work as they got clients paying for it. Just like IPv4.

But are those clients paying to trial IPv6, or paying to make money with
IPv6 now?  I severely doubt it's the latter (but would be very happy to
be proved wrong :)
 
> You might also note that the 6bone should have nothing to do with RIR
> space.
> And currently it has due to the 6bone mess.

My suggestion is that while 3ffe: prefix sites are perhaps the greatest
culprits, the 2001: space has its share too.  Thus the way to get some
kind of structure and predictability (and thus reliability) is for 
mutually interested networks to get together to talk policy - this is
happening now for the major academic/research networks, some of whom are
still in 6bone space (note that technically under the RIR rules, an NREN
cannot get 2001: allocations as most do not serve 200 universities).

Tim