Francisco,

My understanding is that the expectation of RDAP is to implement the RDAP RFC’s 
as a Registration Data Directory Service (RDDS) and not to replace all features 
implemented in WHOIS by various providers.  You chose to provide elements of an 
availability service in the form of an unavailable feature in WHOIS.  In RDAP, 
the registry is not obligated to provide an availability check or reason why 
the object is not available in RDAP but simply return the data if it exists and 
the user is authorized.

I believe that an availability service is best implemented by checking the 
availability of one or more objects in a lightweight and highly performant 
manner, as opposed to implementing an unavailable feature on an individual 
lookup.  EPP has a separate check command from an info command for the reason 
of providing a lightweight mechanism to get the availability information of 
many objects in a single packet.  An unavailability feature of a lookup is 
technically possible, but I don’t believe it meets the needs for a scalable 
availability service.

Just as the availability check is being extended in EPP with items like claims 
information and fee information, an availability service runs a similar risk of 
getting heavyweight with features.  Some of these features are best handled in 
the OLTP system based on the complexity and required accuracy, so I believe 
it’s best to discuss who will use the availability information, what would they 
need, and what are the non-functional requirements for it.  This all needs to 
be done prior to proposing a protocol to handle it.


—

JG

[cid:image001.png@01D259D2.EB9A7C20]

James Gould
Distinguished Engineer
jgo...@verisign.com

703-948-3271
12061 Bluemont Way
Reston, VA 20190

VerisignInc.com<http://verisigninc.com/>

From: Francisco Obispo <fobi...@uniregistry.com>
Date: Friday, December 16, 2016 at 5:37 PM
To: James Gould <jgo...@verisign.com>
Cc: Andrew Newton <a...@hxr.us>, Registration Protocols Extensions 
<regext@ietf.org>
Subject: [EXTERNAL] Re: [regext] Using RDAP as a Domain Availability Service


Hi James,

From what I’ve seen, the expectation is that RDAP will substitute port 43 WHOIS.

We (uniregistry) currently do availability checks when the domain is not 
registered AND we also provide a very specific reason on why a domain is not 
available.

Customers also want to know if a specific domain is under a specific policy 
that might change in the near future, for instance when we launched out TLDs we 
had a large set that was held back for a period of time (A.K.A. names 
collision), and just showing as “object does not exist” did not provide 
information on why and whether the name would be available in the future. We 
had to become very explicit when our support queue was growing very large from 
customers asking for a specific reason and/or dates when they could register 
the names.

For us, RDAP has to cover both use cases. Whether a registry is obligated to 
provide an availability check or reason why the object is not available, should 
be mandated by policy and not in anyway by limitations of the protocol.

Best regards,

Francisco Obispo
CTO - Registry Operations
____________________________

[niregistry]<http://www.uniregistry.com/>

2161 San Joaquin Hills Road
Newport Beach, CA 92660

Office +1 949 706 2300 x4202
fobi...@uniregistry.link

On 16 Dec 2016, at 9:22, Gould, James wrote:

That is one of my concerns, RDAP is a lookup protocol and not the SRS and 
should only return data that exists with no reasons why it does not exist other 
than scenarios like authorization issues or when data can be found elsewhere by 
reference.  We have not defined the use cases and it should not be assumed that 
because RDAP provides registration data that it should also support 
availability logic like the SRS.  Can we clearly define the use case of this 
undefined service?  I view this as a separate service from the SRS and RDAP to 
meet the undefined needs of a yet to be defined set of users.
_______________________________________________
regext mailing list
regext@ietf.org
https://www.ietf.org/mailman/listinfo/regext

Reply via email to