On Wed, 8 Dec 2004, JFC (Jefsey) Morfin wrote:
At 04:46 08/12/2004, shogunx wrote:
both count. if they do not understand it to the level of acceptance at
least, then how its built does not matter. if its not built correctly,
large percentages of migrators will drop anchor and turn around to v4 NAT
again.
True. Obviously the techology is of the essence. What I mean is that IPv6
will only take off the day the reason why IPv6 was designed is permitted to
be used (to be an IPv4 with larger addresses).
what is standing in the way at the isp level are the following:
a) support applications. we all know good network operators are usually
anal about security. anyone have a packet sniffer for v6?
b) revenue streams for isp's. currently, v4 addresses are a commodity
item. they cost the end users, no matter who you go to. are the isp's
willing to give up these revenue streams for better technology? perhaps
the independents, but asking a telco to give up a way to make money once
they have already found out how to extract it from the public is like
trying to get a sperm sample from your grandmother. good luck.
This means that users will
be permitted to freely innovate in the way they use the Internet in _not_
carring about the type of address they use. And that we do not block this
innovative usage in not permitting what this innovation may need, and in
not stabilizing the standards. Today I think these needs include legal
protection,
why and from whom?
regalian services,
please define?
permanent addressing,
solved. my tunnels provide with them my allocations, for all practical
purposes. i have had the same v6 addresses on my hosts since i
implemented v6 quite some time ago, with AAAA's pointing to the hosts i
wished to make public. of course, i have added many more hosts since i
implemented v6. people in the know are asking to colo in my home office
simply because i have v6.
independence from ISP,
solved. a tunnel is portable.
plug-and-play, ...
if you mean stateless autoconfig, then that is solved too. if it is not
wit you, then i suggest that you contact your OS vendor, or better yet,
move to a better OS;)
Obviously as you say. The "internat is the future", with NATs adding
functions over functions.
i'm just saying that since we have NATs, we already have layer 1 of the v6
network in place at the end user premises.
But we will then talk more of "corebox" than
NATs.
having built one of those and implemented it on the atlanta backbone some
months ago (remotely no less) there is a need for real large scale routing
hardware to handle v6 expansion at the backbone and isp level.
They started as NATs, but once they are under IPv6 - and not a NAT
anymore - they will continue to be here, and to provide an increasing pile
of services (starting with OPES, and their network overlay and all the
possible new architectural non-end-to-end systems
IMHO it is the end to end possibilities that are the most exciting.
.. and all the debates
this will rise). So, let talk of "interbox".
Exciting future.
Indeed.
scott
jfc
sleekfreak pirate broadcast
http://sleekfreak.ath.cx:81/
_______________________________________________
Ietf mailing list
Ietf(_at_)ietf(_dot_)org
https://www1.ietf.org/mailman/listinfo/ietf