In response to the ICANN enquiry, the NRO and IAB submitted comments to suggest the constituency-based approach to transition planning, with an assumption of 3 constituencies or communities of interest in IANA services, namely names, numbers and protocol parameters. Although the CG has not agreed as such, it seems that we are proceeding in this direction. I am wondering now whether this needs more work and refinement, as we discuss the (un)likelihood of consensus within one or more communities. I think the "3-legged stool" view of IANA is valid and useful for this exercise - because the 3 communities, while they share an interest in IANA, are customers of a set of services which are almost entirely independent. So, in theory at least, we can do our work independently without too much risk of conflict. But we discussed this is breaking down in connection with the “names community”, because, it seems to me, such a community really does not exist, and may not be viable due to fundamental differences. Is there a solution for us in subdividing the names community further into GNSO, CCs, and Registries, as in fact the CG constituencies reflect. My question is whether the transition plans of these groups actually do need to be shared, or if they can be asked to lay out their plans/requirements independently, for us to then reconcile. We may find that their differences can actually coexist in the final plan, without breaking anything. Thoughts? Paul. ________________________________________________________________________ Paul Wilson, Director-General, APNIC <dg@apnic.net> http://www.apnic.net +61 7 3858 3100 See you at APNIC 38! http://conference.apnic.net/38