Re: GSE

[Masataka Ohta <mohta@necom830.hpcl.titech.ac.jp>]

Iljitsch;

> > > That being said, I think GSE could be a subset of a more general
> > > approach.
> 
> > Wrong. Any approach containing GSE as a subset is worse than GSE.
> 
> With a solution that can work with any kind of structure for the
> identifier it would be easy to implement GSE for those who still want
> it,

Wrong.

GSE is impossible to implement, not because of its structure for
the identifier but because of its ignorance of the end to end
principle.

> without betting all our cards on one horse or a cliche of similar
> sentiment.

GSE was a dead horse from the beginning.

It has been harmful for healthy horses.

> > > However, using identifiers with regular IPv6
> > > unicast semantics will make the transition a lot easier as it allows
> > > interoperability between multihoming-aware and non-multihoming aware
> > > systems and/or providing the multihoming support in separate boxes.
> 
> > Transition from what? IPv4? OSI? Or?
> 
> Transition from existing v6.

An interesting theory.

							Masataka Ohta