inline... On Fri, Jan 3, 2025 at 4:43 PM Jasdip Singh <jasd...@arin.net> wrote: > > https://github.com/anewton1998/draft-regext-rdap-extensions/issues/42 > > > > >> Section 2.5, paragraph 2 > > >> RDAP extensions not defined by the IETF MUST use the extension identifier > >> as a prefix in accordance with this document, [RFC7480], > > > > > I'm not very much convinced if this requirement is practical. > > > > [JS] Not sure if a non-IETF extension would get the same rigorous scrutiny as > an IETF extension. This prepending is an insurance for preventing namespace > conflicts down the line. For emulation, there is a precedence for it in, for > example, the media types area with the vendor and personal trees.
Extension identifiers are needed to avoid namespace collisions. That's basic RDAP from the beginning. This text is simply noting that the IETF can sometimes go without them because the IETF presumably can avoid colliding namespaces within its own specs. But even then the suggestion is a SHOULD in that the IETF needs to know what it is doing. > > > > > 1. what about the specifications, which are work in progress and not yet > > under scrutiny of IETF review and change control? > > > 2. what about a general purpose extension which won't get defined by IETF > > and would like to approach it later? What about them? Is there something so onerous in this requirement that they must break the rules? > > > > [JS] If so, it would at that time be considered from an IETF extension angle > and would most likely get a new extension identifier which may or may not be > used for prefixing. > > > > > 3. would it not be just enough to have a proper review on IANA level when > > identifiers are being registered to assure no general terms are being > > overloaded by a specific extension or no conflicts between the extensions? > It would not be enough, because that supposes the IANA or the DEs need to deconflict every JSON name etc.. and that would lead to mistakes. -andy _______________________________________________ regext mailing list -- regext@ietf.org To unsubscribe send an email to regext-le...@ietf.org
