[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]


To: Eric Brunner-Williams in Portland Maine <brunner@nic-naa.net>
CC: Daniel Manley <dmanley@tucows.com>, "'ietf-provreg@cafax.se'" <ietf-provreg@cafax.se>
From: Klaus Malorny <Klaus.Malorny@knipp.de>
Date: Thu, 06 Sep 2001 11:16:25 +0200
Sender: owner-ietf-provreg@cafax.se
Subject: Re: Questions about containers

Eric Brunner-Williams in Portland Maine wrote:
> 
> containers "live" in the registry. end-users don't set them up and instantiate
> them in the registry, as they've no access except as mediated by a registrar.

Hi Eric,

Of course you are right that containers live in the registry. Therefore I
could try to hide the existance of containers. Unfortunately, there exists a
whois service. I assume that it would also give detailed information about the
containers the domains are linked to (otherwise, this service wouldn't be very
practical for professionals). But this service is also available to end-users
and resellers, so I as a registrar/reseller (and not the registry) have the
pain to educate them. And even much more trivial concepts are sometimes quite
hard to explain to the masses.


BTW, some more comments, independent from the discussion above:

* the response to <domain:info> request does not disclose whether an attribute
(e.g. host or contact) is derived from its container or whether it has been
inserted directly into the domain. Even a comparison with its container would
not reveal this information. That's bad.

* There's a big chance that an update of a container may render objects that a
directly or indirectly linked to that container void, e.g. a domain defines by
itself 12 name servers, and the the user attempts to add two name servers to a
container of it, thereby exceeding the limit of 13 name servers. Reparenting
objects and containers also has a big probability to fail as suddenly some
required contacts may be missing. Of course, such attempts should result into
an error. But the document does not describe how this is done. I think the way
this is reported must be defined clearly, and all objects that would become
void must be reported.

* although I see some benefits in the container model, I am realizing slowly
that it does not help to overcome my "single registrant - multiple registrars"
and "registrar - reseller" problems in any way.



Klaus Malorny


___________________________________________________________________________
     |       |
     | knipp |                   Knipp  Medien und Kommunikation GmbH
      -------                           Technologiepark
                                        Martin-Schmeißer-Weg 9
     Dipl. Inf. Klaus Malorny           44227 Dortmund
     Klaus.Malorny@knipp.de             Tel. +49 231 9703 0

Home | Date list | Subject list