Re: Fwd: New Version Notification for draft-gieben-epp-keyrelay-00.txt

On 24/01/13 13:22, Antoin Verschuren wrote:

> Op 24-01-13 11:27, Klaus Malorny schreef:
>
>> interesting draft. On the first glance, I'd like to note the
>> following:
>
> Hi Klaus, thank you for your feedback. Comments below.
>

Hi Antoin,

>> 1. why is the use of the authinfo only optional?
>
> Because it is also optional in the EPP transfer command.
> A registry may declare a transfer token mandatory in local policy, but
> a registry may also determine that transfers that are not
> authenticated by a token will proceed if the losing registrar does not
> object within a certain time frame. A registrar may object if the
> transfer is not authenticated by a token, but he may also choose not
> to object, again by his local policy.

Ok, although I consider the support for transfers without authinfo as a backward 
compatible feature no longer required and desired today.

> This is a very good suggestion. We will update the message queue
> format to include the keyrelaying registrar in the next version of the
> draft.
>
>> 3. How long shall the current registrar/name server operator add
>> the DNSKEYs to the respective zone?
>
> This question is process, not syntax.

I cannot agree. There are certain expectations with the protocol, namely that 
the current registrar performs the necessary steps to have the DNSKEYs published 
on the domain's name servers, and this is already "process". But defining only 
the addition and not the removal is only a half thing. Without clear semantics, 
i.e. with uncertainties about the consequences of its use, the value of a 
protocol decreases a lot. Of course, the semantics need not to fully cover 
Peter's/Marcos' paper.

Regards,

Klaus

_______________________________________________
provreg mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/provreg