Reasonable to use crypto in all communications? (Re: Fwd: Minutes/ Notes)

[Pekka Nikander <pekka.nikander@nomadiclab.com>]

> So i guess that my question would be: is it a reasonable approach to use
> crypto in all comunications? (such as HIP)
I don't think so.  There certainly is some communication that
does not need any protection at all in addition to what is
provided by the application layer.

On the other hand, it has been stated that there certainly is
some communication that does not need the added flexibility and
robustness provided by the id/loc separation.

Now, there is a good question:  Do we get four distinct subsets,
each of a reasonably big size, or is any of the four subsets
so small that we can almost ignore it:

  - non-robust, non-flexible, non-protected traffic (current IP)
  - non-robust, non-flexible, protected traffic (current IPsec?)
  - maybe-multihomed, maybe-mobile, non-protected traffic
  - maybe-multihomed, maybe-mobile, protected traffic

The id/loc mapping itself needs some kind of protection, or otherwise
we get new nasty DDoS attacks using all hosts, even uncompromized
ones, as zombies.  The actual traffic may or may not need protection.
It is a know deficiency in the current HIP spec that it does not
directly support non-protected traffic.  However, if unencrypted,
non-integrity protected ESP was allowed, one could use the SPI
in the ESP header as a kind of condensed identifier, without
any cryptographic protection.

--Pekka Nikander