Re: EPP statuses and other questions

[Michael Graff <Michael_Graff@isc.org>]

"Hollenbeck, Scott" <shollenbeck@verisign.com> writes:

> People said that they wanted to include additional information describing
> why a status has been applied, to add clarifying info, etc.  The server
> might not care about the reason, but it's not necessarily there for server
> interpretation -- it's there because it describes something about the
> object.

I still don't think the registry is the place for this additional
information.  I'm certain people want to store things like credit
card numbers to bill people with, but the registry isn't the place for
that, either.  I know my example here is absurd, but since the data
isn't there for the operation of the registry, it should not be in
the registry, IMHO.

Why flags were set on domains, hosts, etc. seems like it should go into
the registrar database, instead.  But, I understand why it's there,
even if I think it's a mistake.  :)

> The unique identifier for domains and hosts is the domain name and host
> name, respectively.  The valid syntax for these names is defined in multiple
> normative references listed in the specs.

Ahh, I missed that part, sorry.

> As I've said in a private response to a message you sent me off-list, the
> contact mapping is a general purpose mapping that is not necessarily for use
> only with domain names and the like.  The idea is that the identifier can be
> something that is easy for a human to remember, much like those you provide
> when you register for services like email accounts, web site accounts, etc.
> Implementers and server operators can define an implementation profile that
> requires certain strings in the ID, or restricts use of certain strings, but
> those limits are not something that should be built into the protocol
> itself.

I can see the need for this sort of thing, even if I don't like it.  I
assume this is so I can move my contact/domains to a new registrar without
having to change contact handles, etc.

--Michael