On Jun 21, 2013, at 2:56 PM, John C Klensin <john-ietf@xxxxxxx> wrote: > While I agree with the above (and am still trying to avoid > carrying this conversation very far on the IETF list), I think > another part of the puzzle is that there are also situations in > which technical considerations imply real constraints on policy > alternatives. Some obvious examples include physical constants > like the speed of light, others, only slightly less obvious, > include things like the design of the DNS as a simply hierarchy > that cannot support symmetric aliases (i.e., anything that would > support an actual "came from" function or a list of all of the > names that point to a given note). The policy folks ignore > those constraints, or treat them as subject to policy-making > decisions at the risk of being ridiculous and/or causing > considerable harm to the Internet. While they are less obvious > in this community, I suggest it works the other way too -- there > are policy and economic decisions and realities that are as much > constraints on the technical solution space as those technical > constrains are on the policy ones, with just about the same > risks of ridiculousness or damage if they are ignored. Agreed. I believe that there is a better understanding of this situation now than in the earlier days (including among governments who are beginning to seriously engage with ICANN's GAC.) > That is, again, why it is so unfortunate that the original model > of the IAB/PSO as one of ICANN's three "everyone has to work > together" pillars was abandoned... and more unfortunate that it > was replaced on the ICANN side by approximately nothing other > than some committees and other bodies that could easily be > ignored and on the IETF side by depending on individuals with > feet in both camps to speak up. It's difficult to lay blame anyone from walking away from the PSO approach; in ICANN's early years it always seemed to be a vestigial structure serving little purpose. The lack of apparent value was amplified when ICANN changed its proposed structure (from being oversight and coordination between true independent supporting organizations) into a heavily DNS-focused direction by opting to absorb the DNSO internally in the initial Singapore meeting. If ICANN were operating solely in a coordination and oversight role, with policy, process, and protocol development done in supporting organizations, then it would have been a lot easier to make the liaison and coordination function successful, both between pillars (DNSO, ASO, PSO) and to/from governmental types. For some reason, doing that in the margin of a 97% DNS-focused omnibus policy/oversight/coordination/operation organization doesn't provide the necessary level of attention. FYI, /John Disclaimers: My views alone. Apologies for length; I lacked the time to write a shorter reply.