Re: LIN6 i-d for multihoming and mobility support

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

marcelo bagnulo;

> > Are you seriously considering that users are willing to memorize
> > and type in from key boards lengthy public keys or its signatures?
>
>
> Well, transport and network protocols use 128 bit long strings to identify
> the other end of the communication, so it would be useful to provide a
> mechanism to prove the ownership of those strings

Correctly qualifying, 128 bit psuedo random ID for secure
identification is completely insecure and useless unless there
is a *SECURE* mapping between the ID and a lot more handy ID to
identify the owner.

Without such mapping, such proposals are incomplete.

Note also that, if you assume something like PKI or secure DNS,
you should also assume that session keys are shared and sessions
are secured without bothering multi6 group. That is, there is no
point to have the 128 bit long strings.

						Masataka Ohta