Re: [P2PSIP] New draft: HIP BONE

Hi Salman,

thanks for your comments.

> My general concern with running a peer protocol (RELOAD or P2PP) on top 
> of HIP is that one will need to understand somewhat two completely 
> different documents (HIP and peer protocol document) to implement a 
> working solution. That increases the bar for someone who will like to 
> quickly design and develop a working p2p solution.

that is because you get things like mobility and multihoming. If you 
want to implement these features without using HIP you will need to 
anyway read more specifications (e.g., mobile IP and SHIM-6).

> I think there is a need to carefully perform message overhead analysis 
> of HIP under different levels of churn.

The only difference is that, every time you a node establishes a 
connection with another node, there will be a HIP four-way handshake 
instead of a TLS or DTLS handshake.

> I am a bit confused about 
> splitting the functionality between the overlay layer and the HIP layer. 
> Is peer protocol no more than an encoding mechanism when implemented on 
> top of HIP?

No, the peer protocol provides all the functions it provides today 
except for the CONNECT and TUNNEL primitives (and NAT traversal).

If you mean that the peer protocol runs inside secure connections 
managed by HIP, that is also the case today: the peer protocol runs 
inside secure connections managed by TLS.

Cheers,

Gonzalo

_______________________________________________
P2PSIP mailing list
P2PSIP@ietf.org
https://www1.ietf.org/mailman/listinfo/p2psip