Re: Architecture [Re: Agenda for Vienna]

[Thierry Ernst <ernst@sfc.wide.ad.jp>]

> > All of the stuff we're discussing right now needs support from both 
> > ends in order for multihoming on one end to work. This means change is 
> > NOT optional, we need to get at least 95% of the internet to adopt this 
> > or we have failed.

Definitely, if a solution does not require changes at the end nodes,
this solution should be favored. New protocols shouldn't require change
at all implementations to work. This is a bit too late. If we need to
update exisiting implementations everytime there is a new feature on the
table, IPv6 cannot be deployed.

My point is that modifications at the other end should be done ONLY as
an OPTION to make a feature work BETTER. Same as MIPv6: if you don't
implement the CN operation at the other end node's side, it still works,
but not efficiently. 

My 2 euro cents.
Thierry.