Re: [regext] OK, What Next? (was RDAP Extensions Approach Analysis v2)

["Hollenbeck, Scott" <[e-mail]>]

> -----Original Message-----
> From: Gould, James <jgould=[e-mail]>
> Sent: Monday, June 27, 2022 10:32 AM
> To: Hollenbeck, Scott <[e-mail]>; [e-mail];
> [e-mail]
> Subject: [EXTERNAL] Re: RE: Re: [regext] OK, What Next? (was RDAP
> Extensions Approach Analysis v2)
>
> Caution: This email originated from outside the organization. Do not click links
> or open attachments unless you recognize the sender and know the content
> is safe.
>
> Scott,
>
> The language of the RFCs will support any of the three approaches, where
> the key aspect that may or may not have been discussed originally in the
> working group is versioning.  The only reference to versioning is the use of
> RDAP conformance "rdap_level_0" value and there is no mention of the use
> of versioning in the RDAP elements.  The versioning gap is what is driving the
> discussion now.
>
> Do you have a technical issue with supporting the registration of extension
> prefix identifiers (e.g., " lunarNIC") that ensures uniqueness of the RDAP
> elements (path segments, response members, etc.) along with the use of
> that prefix identifier to provide versioning in the RDAP conformance values
> (e.g., "lunarNIC_level_0", "lunarNIC_level_1", "lunarNIC_level_N")?  A client
> can use the registered prefixes to easily identify the conformance value,
> which may include the version number.  This is what is defined in Approach B,
> which is not as flexible as Approach C, but it ensures that there is a linkage
> between the extension elements and the RDAP conformance value.
> Support for Approach B does not require any change to 9083.

[SAH] Yes, I have an issue with registering one value and using another as a prefix. Here's why:

Imagine that "lunarNIC" is registered with IANA and returned in the rdapConformance data structure. What prefix value should a client expect to see in an extended response? "lunarNIC", "lunarNIC_", "lunarNIC_level", "lunarNIC_level_", "lunarNIC_level_0", "lunarNIC_something_else_who_knows_what_v1", or some other production that is prefixed with the registered identifier? There's no clear answer in 9083, and no way to unambiguously tie the prefix to the rdapConformance value (and thus, the associated specification that defines the extension) unless they match exactly. That's the technical issue that makes it difficult to build interoperable implementations.

Scott