Decius wrote:
] ] Host Identity Protocol [3] (HIP) defines a mechanism that
] ] decouples the transport layer from the internetworking
] ] layer, and introduces a new Host Identity namespace. When
] ] a host uses HIP, the transport layer sockets and IPsec
] ] Security Associations are not bound to IP addresses but
] ] to Host Identifiers. This document specifies how the
] ] mapping from Host Identifiers to IP addresses can be
] ] extended from a static one-to-one mapping into a dynamic
] ] one-to-many mapping. This enables end-host mobility and
] ] multi-homing.
]
]
] Abaddon and I spent some time last summer working on a protocl
] called Adaptive Addressing Protocol (AAP). It was an attempt
] to make Mobile IP less silly by allowing hosts to change their
] IP addresses without dropping connections. It did so by
] associating connections with a unique identifier, secured by a
] Diffie Hellman key exchange. Last night, Jeremy points me at
] this. This is APP, basically. Its amazing how close our
] designs are. Its also *really* frustrating. We were going to
] have a working demo this weekend for Phreaknic.
]
] I'm going to read through this stuff and see if there are any
] design decisions that we made that might be of value to this
] working group. Having said that, its worth memeing this
] documentation. This is the right answer for mobile IP
] and it will probably be the answer we end up with for
] multihoming of small networks because deploying something like
] this is less expensive then renumbering the internet
] geographically.

I've had this idea before as well. Yay!

RE: Host Identity Protocol