Re: IETF59 minutes and presentations

[Pekka Savola <pekkas@netcore.fi>]

On Fri, 12 Mar 2004, Alain Durand wrote:
> ==> I do not see how this scenario is fundamentally different
> from the road warrior using IPv4 on a dial-up modem.
> In many cases, he can call a well known centralized 1-800 number
> that will create a suboptimal long path or he will have a phone number
> for a local PoP, that will hopefully provide shorter RTT.
> There is no 'find me the nearest PoP' protocol here.
> 
> If we look at a tunnel broker as a virtual IPv6 ISP,
> can someone point me how things are different?

Road warrior has made a contract with an ISP, which is his back-up
long-distance number.  If we assume that all the users would have to
make a contract/signup/whatever with a long-distance virtual IPv6 ISP,
then this would probably be the case.

What I'd like to get is when the local ISP is offering the tunnel
service, the user would get directed there automatically, and would
become aware of it so that he could start using that service instead
(instead of using long-distance service).

That is, I think we want to avoid these "long-distance calls", and we
want to make it easier for the user to notice whether there is support
in the local ISP.  Additionally, the configuration of tunneling IMHO
should be close to zero-config.  (Looking up tunnel server addresses,
signup forms, or whatever in the web is unacceptable -- and that's one
reason why the tunnel broker model hasn't seemed to fly so well so
far..)

If we want to make this easy for the users, that seems like something 
that would have to be done.  Especially if we insisted that we 
wouldn't need Teredo.

Hopefully I answered your question..

-- 
Pekka Savola                 "You each name yourselves king, yet the
Netcore Oy                    kingdom bleeds."
Systems. Networks. Security. -- George R.R. Martin: A Clash of Kings