> -----Original Message----- > From: regext <[email protected]> On Behalf Of Andrew Newton > Sent: Thursday, August 3, 2023 11:48 AM > To: Arnt Gulbrandsen <[email protected]> > Cc: [email protected] > Subject: [EXTERNAL] Re: [regext] Proposed update to draft-ietf-regext-epp- > eai > > 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. > > On Thu, Aug 3, 2023 at 10:49 AM Arnt Gulbrandsen > <[email protected]> wrote: > > > > Hi, > > > > > If we’re going to stand firm on the current working group consensus, > > > then I believe the question to be answered is why EPP is special and > > > shouldn’t be required to align with accepted practice? > > > > What's accepted practice? > > > > We have phone numbers, postal addresses and email addresses in whois. > > All are known to be unreliable, only mostly work AIUI, and that's accepted. > > IMHO, it's more nuanced than that. Text-capable phones and all-ASCII email > addresses can be verified via automated processes, and there are some > regular processes in both the INR and DNR spaces to periodically re-verify > them. > > EAI sorta breaks that model in that an SMTPUTF8 address can work for Alice > but not for Bob. > > I think requiring the collection of 2 email addresses impinges on policy (and > is > unworkable in many of today's registries). But I think Jim has a point. Are we > painting ourselves into a corner? > What's to stop us from being here 3 years from now because a policy body is > pointing at RFC 6530 and saying "it says to have an ASCII alternative"* ? > > Instead of "MUST have ASCII with SMTPUTF8" or "MUST have either ASCII or > SMTPUTF8", maybe this should be more flexible and allow for both an ASCII > and > SMTPUTF8 address > so long as at least one is required and they are clearly discernible as to > what > they are.
[SAH] I've come around to this position, too. I like the idea of adding support for a second email address, which could be either an all-ASCII address or an SMTPUTF8 address. This would allow us to support EAI, but it also adds flexibility to support an alternate email address that can be used for things like account recovery if the original email address becomes unusable. I'm comfortable with the idea of requiring a second all-ASCII address if one of the addresses is an SMTPUTF8 address. Scott _______________________________________________ regext mailing list [email protected] https://www.ietf.org/mailman/listinfo/regext
