Re: [registrars] Proposed Ballots

[Eric Brunner-Williams in Portland Maine <brunner@nic-naa.net>]

> > >- that delegants and operators be encouraged to leverage existing
> > >  registry protocols and not create new ones
> > 
> > This seems to imply that the best protocols already exist. I 
> > agree that life would be much easier if we had a single 
> > protocol to deal with. On the other hand I would not want to 
> > stifle innovation and potential future benefits to save a 
> > little time today. Besides, so far, even with EPP, each 
> > implementation has been different. I don't think we could 
> > support this ballot with this comment as part of it.
> > 
> 
> No, it implies that new operators should be encouraged to use existing
> protocols unless they have demonstrably better ideas.

There is an implementation of "EPP" using HTTP as the transport protocol.
There are implementations of "EPP" using SMTP as the transport protocol.

I put the tla "EPP" in quotes because there are IESG dicta that make it
unlikely that these mechanisms will be labeled "Draft Standard".

That doesn't mean these are "bad" ideas -- no one in the IESG has real
experience as either a registrar or a registry.

My point is, lets not take the word "protocol" as being without nuance,
we don't want to force all future registries into the same ultra-expensive
boat that NeuStar and VGRS chose to sail. The IESG is a political as well
as a technical body, and on registry protocols, their function is more of
the former than the latter.

>                                                   ...  These are the same
> rules that we played by in the last round and instead of creating a
> bunch of new RRP registries, we ended up with a bunch of players that
> worked together and settled on a better idea. We should continue to
> encourage this spirit of cooperation.

For the record, having been working for one of the four-and-a-fraction
parties that created EPP (NS, AF, GNR, RCOM, + VGRS), sitting across from
the Liberty/Afilias/Tucows player(s), I agree.

> If you have a better way to get these points across, please put forward
> an amendment - positive criticism is always a useful tool for change.

I'm sure we want to encourage schema compatibility, without welding our
futures to version 1 of EPP.

I'm sure we want to discourage schema incompatibility, where an operator
chooses to stick in some GOTO-esque kludge (e.g., .US nexus requirement).

An operator who manages to make some wildly ideosyncratic mechanism the
registration protocol for that registry, will have managed to restrict
that market very differently from any other recently created. Some will
find that an attractive proposition.

Eric