Re: [ietf-provreg] FYI: EPP implementation by the Polish registry

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

I'm not twisting your words, I'm not doing anything but try to get to the
bottom of something.
    
>     Correct.
>     
>     So there no semantic consequence to the registrar being, or not being,
>     "based within the European Union."
> 
> Not with respect of the privacy issues. That is under Dutch regulation,
> as I have said before.

OK.
http://www.nic.nl/sidn/flat/Deelnemers/Deelnemer_worden/index.html
is garbage, in this context.

>     The registrar may (hypothetically) provision some datum to the .nl registry,
>     with some (equally hypothetical) binary toggle set to ZERO, or to ONE, with
>     out difference, independent of the registrar's location. Is this correct?
> 
> Yup, see above.

Good.

>     What contract?
> 
> Between registry and registrar.
>     
>     Better still, if a registrar doesn't need to signal in-band it accepts
>     anything, then it doesn't need to even know whatever that useless thing
>     is.
> 
> You are twiting words. See my earlier response how this might work.

Really? It seems that Scott (ages ago, London-context) said "contracts" as
the mechanism. I'm sorry but I don't know which earlier response you are
referring to. Could you provide a date? I'll read it.

>     8.4 has a MUST in it, which appears to have no meaning at all if I
>     finally understand you. Everything is in the contract -- Scott's opening
>     position, that bilateral out-of-band mechanism (contract) covered the
>     requirement.
> 
> A MUST in a requirement document might work out differently in practice.

In theory ... Yeah. Patrick's quote.

>     Well, you're a co-chair, so I guess we're finnished with
> 
>        The protocol MUST provide services to identify data collection policies.
>     
>     and are on to
>     
>        "the proper element in the protocol which help to implement a policy"
>     
>     Gosh this is such fun. Not.
> 
> I was not talking as co-chair.
> 
> 	jaap
> 

These two sentences mean very different kinds of things.

One attempts to describe a non-trivial policy space, without knowing a priori
the constraints. A mechanism of description. A "what".

One attempts to select from a set of elements (generally) the one(s) which can
effect a particular policy (possibly pluralized). A mechanism of effectation.
A "how".

Usually "how" comes after "what" (except in cowboy and indian movies), not
as an alternative to "what".

Eric