Hi, I've reviewed this draft. I think this draft provides useful ideas and closes up to the LC, but the following points should taken into account.
1. multicast router support I almost agree with Behcet at this point. This draft should focus only on the case MAG acts as MLD proxy. All descriptions related to the case MAG acts as multicast routers (i.e. PIM routers) should be completely removed. (I would say WG does not need to decide only one solution. It sounds reasonable for this WG to provide both scenarios, "MAG=MLD proxy" and "MAG=PIM router". Both are beneficial for some condition or situation. The decision which scenario should be selected is an operator's or provider's choice, accouring to their conditions or situations. Well, it does not relate to the direction of this draft. I'd like to stop discussing about it here.) 2. local subscription This draft uses both "remote subscription" and "local subscription". But I don't think it is a bit confusing to use these words without clarifying the terminologies. The point of this confusion comes from several misuse of "direct routing" and "localized routing". Localized routing is formally defined in RFC6705. In my sense, local subscription includes both direct routing and localized routing, while this draft only focuses on direct routing. So, my recommendation is to remove these words, "remote subscription" and "local subscription", from the draft, and just use "subscription via MTMA" and "subscription via direct routing". 3. direct routing Similar to above. It'd be better to clarify that direct routing this draft focuses on does not include localized routing. Lastly, I have one question. Does this draft consider the case in which MAG enables *both* a tunnel to MTMA and direct routing? I could not find such consideration... Regards, -- Hitoshi Asaeda _______________________________________________ multimob mailing list multimob@ietf.org https://www.ietf.org/mailman/listinfo/multimob
