True Multihoming solutions (Was: At the risk of being declared off topic)

On Wed, 2005-01-12 at 10:26 -0800, Owen DeLong wrote:

I realize that this is more of an IETF issue than a NANOG one, but, I'd like to find a couple of people with some protocol background and a strong operational background that would be interested in trying to see if we can come up with a way to develop a version of IP which did not require a flag day, but, would facilitate a transition towards an IP where the End System Identifier (Host Address) could be globally unique to a given system (think PI space) and essentially a layer 4 construct, but, the Routing Tag (Network Address) could, instead, be dynamically determined and separate information.

Check HIP: http://hip.piuha.net/ As that is exactly what you describe. Greets, Jeroen