To:
"Andrew Sullivan" <andrew@ca.afilias.info>, <ietf-provreg@cafax.se>
From:
"Hollenbeck, Scott" <shollenbeck@verisign.com>
Date:
Tue, 19 Jul 2005 17:30:47 -0400
Content-class:
urn:content-classes:message
Sender:
owner-ietf-provreg@cafax.se
Thread-Index:
AcWMoqJ/yhoe+ma/TdWYH3IFn+snnAAA3FkQ
Thread-Topic:
[ietf-provreg] EPP Document Updates
Subject:
RE: [ietf-provreg] EPP Document Updates
> -----Original Message----- > From: owner-ietf-provreg@cafax.se > [mailto:owner-ietf-provreg@cafax.se] On Behalf Of Andrew Sullivan > Sent: Tuesday, July 19, 2005 4:29 PM > To: ietf-provreg@cafax.se > Subject: Re: [ietf-provreg] EPP Document Updates > > On Tue, Jul 19, 2005 at 02:47:50PM -0400, Hollenbeck, Scott wrote: > > Thanks for the thoughts, Ed. > > > > > I talked this over with the developers in-house - I'd (we'd) > > > recommend altering the pendingDelete rule (first option). > > > > That's what I originally intended with 3915, but it may mean a text > > change to 3731 to ensure that the intention is clear. Not > necessarily a > > big deal. > > Something that isn't clear to me -- and I know that Janusz disagrees > with me about this, and I'm not even sure what I think about it > myself -- but why does the pendingDelete status prevent other changes > on a domain? If you want to prevent updates, why not add > serverUpdateProhibited too? (The same goes for other pending status > values, and other objects.) More memory taxation... The pendingDelete status (originally PENDING-DELETE) has been in the specs since the very first individual submission. The restriction went in between -04 and -05 of the working group versions. I just spent a few minutes looking through the archives from 2002 to see what (if anything) was discussed, but a quick glance didn't turn up anything. I *think* the idea was that someone (maybe me after talking to registrars) wanted pendingDelete to be a final, irrevocable "warning" period before a domain would become available for re-registration. Once the removal process started, it should be hard or impossible to prevent it from completing. I agree that the transform restriction seems to be kludgy in and of itself. The protocol already includes features to prevent transforms, if desired by the registry, by adding status values as Andrew suggests above. In short, I believe that the right thing is to remove the "pendingDelete prevents transforms" restriction from 3731. -Scott-