Hi Mario,

> On 17 Dec 2025, at 4:42 pm, Mario Loffredo <[email protected]> wrote:
> 
> Hi Gavin,
> I reviewed the draft and here is my feedback:
> 1) I believe "related" is too general a link relation type to trigger a 
> referral request from a domain registry to a domain registrar. To avoid 
> ambiguity, it would be appropriate to register a specific relation type at 
> IANA, such as "rdap-sponsor." It might also be helpful for servers to return 
> a consistent return code; that is, an object could contain multiple "related" 
> links but only one "rdap-sponsor" link. Therefore, the request for that link 
> could succeed or fail. The language in Section 3 does not help to clarify the 
> problem, since servers may apply different policies in returning any of the 
> links with the same relation type. 

This draft does not currently *require* the use of "related" links for 
registrar RDAP records. It could define a new relationship type, but links with 
the "related" type (along with the "application/rdap+json" media type) are 
already widely supported on both client and server.

> 2) Because the RDAP response to a referral request should not include a body, 
> using the GET method to receive a 200 return code as described in this draft 
> would not conform to its use in RDAP under Section 4.1 of RFC7480.
> The HEAD method also could not be used, as it would need to determine the 
> existence of the data—in this case, the desired link—but not to return the 
> link itself. If, however, the server returns a 301/302 return code, using the 
> Location header for redirection would not comply with Section 5.2 of RFC7480. 
> This section refers to redirecting the original query, whereas in this case, 
> the Location header would contain a different one.
> Perhaps a solution could be to use the 303 return code provided by section 
> 15.4 of RFC9110:
> Redirection to a different resource, identified by the Location header field, 
> that can represent an indirect response to the request, as in the 303 (See 
> Other) status code.

The 200 status codes in the examples are typos that will be fixed. The text of 
the draft does not contemplate using 200.

Thanks,

G.

--
Gavin Brown
Principal Engineer, Global Domains & Strategy
Internet Corporation for Assigned Names and Numbers (ICANN)

https://www.icann.org

_______________________________________________
regext mailing list -- [email protected]
To unsubscribe send an email to [email protected]

Reply via email to