Thanks! > On 17 Mar 2025, at 2:08 pm, Tom Harrison <t...@apnic.net> wrote: > > Hi Mark, > > Thanks for your review. > > On Sun, Feb 16, 2025 at 07:28:17PM +1100, Mark Nottingham wrote: >> Section 10.2 has a title of "Link Relations Registry" but the URL >> given is "https://www.iana.org/assignments/rdap-extensions/ >> rdap-extensions.xhtml" -- it looks like the intent is to register >> link relations given the content of the entries, so the URL needs to >> be corrected. > > This has been updated. > >> Assuming that's the case, I see that 'up' is already registered, so >> that registration is not necessary. >> >> Both that and the remaining requests are extremely generic terms, >> and so aren't appropriate to register in a specific protocol's >> document; experience shows that doing so misleads readers to believe >> that these terms are specific to that protocol and not useable >> elsewhere. >> >> Instead, I'd suggest registering them in a separate, standalone >> document, using generic descriptions that are broadly applicable; >> your document can then refine their semantics in a specific context. >> I suspect the HTTPAPI WG would be amenable to such work. >> >> Alternatively, you could register link relation types specific to >> your application; for example, "rdap-up" and so forth. > > We've updated the registrations to be RDAP-specific, by prefixing them > with "rdap-". > >> I also notice that you define compound link relation types, e.g., >> "up-active" and "top-active". Consider defining a separate "active" >> relation type (again, either generically, or specific to your >> application, depending on your preferences), which can then be used >> in combination with other types -- e.g., "top active" and "up >> active". > > Similarly here, we've removed the "up-active" and "top-active" > relations in favour of a single "rdap-active" relation, to be used in > conjunction with "rdap-top" and "rdap-up". > > -Tom
-- Mark Nottingham https://www.mnot.net/ _______________________________________________ regext mailing list -- regext@ietf.org To unsubscribe send an email to regext-le...@ietf.org
