Re: Opportunistic Tunneling

[Pekka Savola <pekkas@netcore.fi>]

Tailed down Cc: a bit..

On Tue, 24 Feb 2004 itojun@iijlab.net wrote:
> >Both 6to4 and Teredo can work, to an extent, when relay routers don't 
> >exist.  TSP can't.
> 
> 	how useful is the IPv6 internet if you can talk with 6to4 users only?
> 	i think the argument is bogus.

Based on the deployments of Microsoft I guess some do view this as a
sufficient first step.. because you can still use IPv4 to those native
IPv6 users.

I'm not sure whether I agree this kind of fragmentation is a good
idea, but if the intent IPv6 deployment is host-vendor -driven (as
noted in the first message of this thread), mainly based on the desire
to run p2p applications directly, this may be a real consideration.  

> >There is path optimization between 6to4 nodes and between Teredo 
> >nodes.  There is no such thing with TSP (for the good and the bad).  
> >I.e., all the traffic must pass through the server/broker.
> 
> 	i think it causes serious misunderstanding if we characterize this
> 	property as indication of "opportunistic".  optimization property
> 	has to be discussed under separate thread.

I believe the word, "opportunistic" was in this context coined up by 
Rob Austein, and should be interpreted in the similar way as 
"opportunistic IPsec".  That is, you can talk directly with others, 
not just a [tunnel] server.

By that definition, the "path optimization" seems to be an integral
property of "opportunistic" methods.. because there are other methods,
which require more infrastructure (e.g., tunnel service such as STEP,
TSP, ISATAP etc.) which can be basically close to zero-configured.

I agree that "opportunistic" is probably a confusing term, but I
haven't been able to think of a better one.

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