RE: [gnso-pednr-dt] Whois Output from "Thin Registry"

["Tim Ruiz" <tim@xxxxxxxxxxx>]

My personal opinion:

In reality the actual state of the registration is exactly what is
displayed in the Registrar AND Registry Whois. At the Registry the Whois
may correctly reflect that the domain has been auto-renewed for its
customer (the Registrar). At the Registrar or Reseller the Whois
correctly reflects that the domain has expired for its customer (the
Registrant). This is an issue of education more than anything else. The
Registrant is a customer of the Registrar or Reseller and must rely on
what the Registrant or Reseller are telling them to understand the true
state of their domain.

Third party Whois services are impossible to control. Registrants should
be working with their Registrar or Reseller and they need to be educated
to that effect. Trying to standardize what Registries do with
auto-renewal (or not as the case may be) conflicts with free market
development by restricting what Registries may decide to do to
differentiate themselves from their competitors. The confusion about
exipiration dates is due to a lack of understanding about how the system
works and who the customer actually is for each component part. Again,
eductation is the key.

Tim

-------- Original Message --------
Subject: RE: [gnso-pednr-dt] Whois Output from "Thin Registry"
From: Alan Greenberg <alan.greenberg@xxxxxxxxx>
Date: Tue, September 22, 2009 7:53 pm
To: "gnso-pednr-dt@xxxxxxxxx" <gnso-pednr-dt@xxxxxxxxx>


To put this in perspective for those who were not on the call, the 
question that was raised was "does the registry whois data have an 
expiration date". The reply of several of us is that it must, but 
Mike Palage offered to verify and send the result.

Anecdotal evidence (by a number of respondents to the ICANN call for 
comments among others) indicates that the only date some people see 
post expiration but before renewal by the registrant is the registry 
date that is now a year out. Based on Rob's e-mail, at least some 
registrars will still show the now-past expiration date. Perhaps just 
another of the variations among registrars.

Alan

At 22/09/2009 06:29 PM, Rob Hall wrote:

>Ron,
>
>This is not a byproduct of whois, but rather how auto-renew works.
>
>In order to help safeguard against deletions by mistake, the Registry
>renews everything automatically. This saves the Registrar from having
>to send an explicit renew command to the Registry and ensures everything
>gets renewed. The downside for the Registrar is that the Registry bills
>the Registrar immediately for the next year on auto-renew, so Registrars
>need to have large accounts of money at the Registry just to handle
>renewals that a Registrant may not have paid for.
>
>Instead of an explicit renew command, the Registrar must send an
>explicit delete command to delete a domain.
>
>Frankly, I think this is a great safeguard, and I can tell you that in
>Registries that don't auto-renew, mistakes get made and clients are
>affected. I would think consumers would prefer this auto-renew
>safeguard.
>
>An outcome of this, is that the day of expiry, the expiry date gets
>pushed out to the next year. This is because the Registry has no
>information to the contrary. It is impossible for the Registry to show
>any other date.
>
>But in reality, this thin Registry whois is simply a method if finding
>out who serves the actual whois. Most sites I know of serve the
>Registrar whois, and never just the Registry whois. It is this
>Registrar whois that typically has the proper expiry date, as this is
>one of the ways we inform our customers of this date. I know of no
>Registrar that puts an expiry date in the Registrar whois that is not
>accurate to the current state of the domain in the Registrars system.
>
>And keep in mind that our customers are coming to our Registrar to
>renew, not typically doing a whois search elsewhere. Our customers also
>typically log in to their account, and we go alert them at that point of
>any domains needing attention.
>
>Some Registrars choose to delete the day after expiry. I know we used to
>do exactly that. This threw the domain on Registry hold and started the
>RGP clock ticking. Registrars save money doing this, as the Registrar
>immediately gets a refund from the Registry of the fees it was charged
>for the next year.
>
>Additionally, most Registrars also put up a landing page, so the
>customers website and email stops resolving, as another indication that
>they better action the renewal if they want it.
>
>I don't think we should be trying to re-write the whois setup or
>changing auto-renew with this working group.
>
>I prefer to focus on our mission, as I am not sure we are actually
>accomplishing our goal by diverging down tangents.
>
>Rob.
>
>
>
>-----Original Message-----
>From: owner-gnso-pednr-dt@xxxxxxxxx
>[mailto:owner-gnso-pednr-dt@xxxxxxxxx] On Behalf Of ron wickersham
>Sent: Tuesday, September 22, 2009 6:10 PM
>To: gnso-pednr-dt@xxxxxxxxx
>Subject: Re: [gnso-pednr-dt] Whois Output from "Thin Registry"
>
>
>
>On Tue, 22 Sep 2009, Michael D. Palage wrote:
>
> > Expiration date is included in thin registry output.
> >
>---snip---
>
> > Updated Date: 07-apr-2005
> >
> > Creation Date: 07-jul-1998
> >
> > Expiration Date: 06-jul-2014
>
>yes, an "expiration date" is included in whois responses...
>
>but if that Expiration date were to say 20-sep-2010 (and Network
>Solutions
>asked me to help them know their registration status ;-) ) then on this
>date, 22-sep-2009, i could not tell them that their domain was ok,
>because
>it might in reality have just expired and that although the domain's DNS
>
>servers are working today, i can't assure the registrant that they will
>be
>working tomorrow.
>
>isn't it reasonable that whois (thick and/or thin) unambiguously tell
>everyone (both the registrant or a registrant's advisor and the general
>public as well) the actual state of the registration?
>
>and that the registrar's and/or reseller's web page also display the
>actual state of the registration?
>
>this is not to say that the installed whois mechanisms have this
>capability, nor that it would be trivial to implement changes. but i do
>
>want to point out that the current system is not transparent and does
>not
>give the registrant "opportunity" to properly monitor and inform
>himself/herself to take action to recover a domain which has just
>expired.
>
>-ron
>
>Ron Wickersham