Tony Li allegedly wrote on 03 29 2009 2:04 PM: > Ok, but in a mobile network, it's usually the L2 point of attachment > that's changing, not the L3 point of attachment. When you do change > that (i.e., roaming), then your locator is indeed changing and all sorts > of registration mechanisms are in place to update your locator.
First, assuming your implied claim that "some locator must change when your point of attachment changes" is true, that still doesn't mean that a particular locator must change when your point of attachment changes. We still don't want "must" in the definition of a locator. See my previous mail for proposed words (at the top). Second, in reality there is no L2 locator involved. MAC addresses need not change even if forwarding is using them. And in the real world it's all done with tunnels anyway. >>> identifier An identifier is the name of an object; identifiers have no >>> topological sensitivity, and do not change, even if the >> >> "do not change" -> "do not have to change"? > > > Sure. Someone might want to change their identifier at the time that > they roam for privacy reasons. That's certainly reasonable. Right. See my previous message for proposed text. _______________________________________________ rrg mailing list [email protected] http://www.irtf.org/mailman/listinfo/rrg
