To:
ietf-provreg@cafax.se
From:
Eugenio Pinto <eugenio.pinto@fccn.pt>
Date:
Wed, 07 Jun 2006 18:12:54 +0100
In-Reply-To:
<20060607161610.GC29821@afilias.info>
Sender:
owner-ietf-provreg@cafax.se
User-Agent:
Thunderbird 1.5.0.4 (Windows/20060516)
Subject:
[ietf-provreg] <info> and <roid>
Andrew Sullivan escreveu: > In any case, since all the commands are per-object in EPP, I'm > having a hard time understanding the motivation for a bulk query. So if you have a use case in mind, I'd be interested to > hear it. > > A Andrew, I could tell you some different use cases where there are needed bulk queries. But, unfortunately for my arguments, they are applied only in the transition from non-EPP environment to EPP... So, I think you're wright. We don't need bulk queries unless our Registrars don't have a coherent information system. But, that's their problem. --------------- Another thing that confuses me is why there isn't a <domain:roid> element at the <domain:info> command? A <domain:roid> element contains the Repository Unique Object IDentifier assigned to the domain object when the object was created. It makes all sense that we can retrieve the information we need from the <roid>. I've noticed that DNS.PT is not the only Registry that has different domain objects with the same name. Even if in EPP everyone has assumed only ACTIVE domains were to be provisioned and that the name is a unique identifier on the set of ACTIVE domains, it's a very limiting approach using it as the identifier for the object. Just imagine two different objects with the same name on pendingCreate status. We almost don't know how to query such a domain yet. I know that .br defined an extension for just the same problem. We were almost to do it too. But that's the kind of thing that must be changed in the core specification. Am I wrong? Best Regards, Eugenio