[ppml] Policy Proposal: Definition of known ISP and changes to

[Brian Johnson]

cliffb at cjbsys.bdb.com wrote:
> 
> > 
> > 
> > cliffb at cjbsys.bdb.com wrote:
> > >
> > > "If there is really a desire to get v6 started, ARIN 
> should give every
> > > entity that has existing IPv4 addresses equivalent IPv6 
> > > addresses under
> > > whatever provisions they are under now, including no fees for
> > > grandfathered PI addresses like mine."
> > >
> > 
> > Isn't there already a prefix set aside for mapping ipv4 
> into ipv6? If
> > there isn't, wouldn't doing this imply that anyone who has 
> ipv4 space
> > can simply map this into ipv6 and away you go?
> > 
> > Example: I have an IPv4 allocation of 12.34/16.	My IPv6 
> allocation would
> > be XXXX:XXXX:XXXX:XXXX:XXXX:XXXX:12.34.00.00/96. (since I am not sue
> > off-hand if the prefix exists, or if it exists, what it is, 
> I am using
> > Xs)
> > 
> > If such a prefix exists, it would be very easy to use a 
> dual stack and
> > assign the IPv6 address that matches the IPv4 address.
> 
> How would that work with routing.  v6 only goes to /64 and 
> thus I think there
> would be a lot of v4 tunneling to a gateway to get the v6 in and out.
> 
> 
> Cliff
> 

I have no idea why there would need to be a limit to network routing
prefix sizes other than the obvious bloviating over routing table size.
Not that I don't care, but whenever new technology emerges there is
always a delay to market with equipment that can handle it (at least in
a free market there is).

Is there a technical reason for the /64 barrier or is it just a policy
decision?

- Brian