Hi, I think I must be misunderstanding what you're saying, because my current interpretation of this message makes it sound like the Internet is one big LAN, and I can't imagine you believe that. More below. On Tue, Apr 21, 2015 at 10:17:10PM -0400, Avri Doria wrote:
that the protocol operational community has is that they define all of the directories held by IANA and can change those with protocol actions. The transition does not change this.
Yes.
Nor am I recommending we tackle this particular problem at the moment. But I want to make sure it is flagged.
What is the "problem", exactly?
One of the long term issue we have is that the IETF is also the one that controls the DNS protocols, DNSSEC, and any new protocl elements that may be created in the future. The degree to which IANA adopts future protocols and protocols elements is a decsion that is somehwat orphaed at the moment. Who makes those decisions.
I don't see what issue there is here. The IETF makes protocols. Suppose it invents a new protocol. Suppose that the name policy community thinks it's a stupid or useless protocol. In that case, that community won't implement the new protocol and won't use it. This is not a theoretical exercise: we have more than one worked example. For instance, in perhaps the most obvious case, IETF developed in a WG (a previous one named CRISP, having nothing to do with IANA transition) the IRIS protocol. This was supposed to be a WHOIS replacement and was carefully designed to meet the requirements that came "over the wall" from ICANN. It was a complete failure: virtually nobody implemented it and the one registry that did either has since shut it down (or is in the process of doing so). IETF has no cops or army. It makes protocols and they are either voluntarily deployed, or they're not. I don't see why the case of the names community is somehow extra special.
And this is where the financial comes back in, because as the sole financial source, ICANN does control the funds available to any protocol innovation.
Really? I had not observed ICANN's close involvement in the development of DANE, SIP, XMPP, and so on, let alone its significant contributions to ALTO or OpenFlow. It sounds like you have a centralized, managers-of-the-Net picture of how all this works, and I don't think that's how the Internet actually functions. (I also don't believe that's what you think really happens, so I'm pretty sure I must be misunderstanding you.) A -- Andrew Sullivan ajs@anvilwalrusden.com