On Jun 30, 2011, at 5:47 AM, Fernando Gont wrote:
Jari,
On 06/30/2011 06:38 AM, Jari Arkko wrote:
But their architecture is largely done and cannot be easily affected.
Vendors are now looking into adding IPv6 into their home routers and
other devices. I want to be able to show them how to do it right. They
can, of course, replicate everything exactly as in IPv4. Much of it is
right, of course, but on some areas I think we can do better. This is
why the working group should focus on IPv6.
My point is: Will implementation of the produced RFCs lead to home
networks in which stuff works for IPv6 differently from how it works for
IPv4?
hopefully!
e.g., your home network would have multiple subnets (thanks to
PD), but a single IPv4 subnet?
for that specific example, not clear. my impression is that the architecture
of home networks should work regardless of whether there's a single subnet or
multiple subnets.
If our work focuses only on IPv6, I get the impression that we're
heading in that direction.
nothing says that some results of the work can't also apply to IPv4. but
people are far too mired in outdated assumptions today, such as the idea that
every network needs a NAT or a firewall that filters based on IP addresses and
ports.
If HOMENET is going to improve stuff that we already do with IPv4 (by
leveraging IPv6), then that's fine...
no, that's not fine. that's painting ourselves (and the Internet) into a
corner.
Keith
_______________________________________________
Ietf mailing list
Ietf(_at_)ietf(_dot_)org
https://www.ietf.org/mailman/listinfo/ietf