It does sound like there may be multiple IEEE solutions that might be 
considered as potentially helpful to homenet scenarios.

Just as information to those unfamiliar with IEEE:
1905 is an "entity" based IEEE group (i.e., companies are members, not 
individuals) that is very distinct from 802 (which, like IETF, is 
individual-based), and had little interaction with 802. 1905 set its own 
meeting schedule that had nothing to do with 802. We were fortunate to have 
Philippe participate in 1905, as someone who was also participating in 802. 
IEEE-SA (which IETF has a liaison officer for) oversees all the groups.

I think Philippe brings up some interesting points that we should be able to 
discuss more once people can see the specs.

A use case I wanted to explore is where homenet router 1 has an Ethernet port 
connected to an Ethernet to powerline adaptor; elsewhere in the home is a 
powerline to Wi-Fi brick (pure AP, no routing), and then homenet router 2 has a 
Wi-Fi to Ethernet adaptor plugged in to  its Ethernet port. The 2 routers each 
think they are using Ethernet to communicate with each other. If we consider 
algorithms like in Appendix A of RFC 6126 (babel), where it's suggested that 
different cost computations be used for wireless and wired, the routers 1 and 2 
 would use a "wired" cost computation, because they think they are using 
Ethernet. But since there's a wireless link between the 2 routers, maybe the 
wireless cost computation would be more appropriate. If only they knew.

In a homenet environment, if the xNCP and routing protocol messages are being 
received with regularity between routers 1 and 2, there is no reason to expect 
topology changes have occurred between routers 1 and 2. So a "dynamic" topology 
discovery mechanism may be overkill. A less chatty (more static) protocol may 
be sufficient. 

In any case, my main thinking is "if information from 1905 happens to exist in 
a homenet, it may be useful to describe how to make use of it". I think its 
main usefulness would be as info into various internal algorithms. Not as 
competition to xNCP or routing protocols. The main algorithms I think it could 
inform would be the routing protocol cost algorithm and potentially host 
application source address selection algorithms. But it's not necessary to any 
of this, and I definitely don't want to suggest a dependency on it. Just better 
algorithmic behavior with info than without.
Barbara




_______________________________________________
homenet mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/homenet

Reply via email to