On 15 Apr 2019, at 17:54, Asmus Freytag (c) <asmusf@ix.netcom.com> wrote:

On 4/15/2019 9:46 AM, John Levine wrote:

In article <09cbecda-324d-eb4d-bd45-fb4e64c71a72@ix.netcom.com> you write:

That's the real danger of understanding UA as "blind acceptance" vs. 
universal support for well-behaved (if non-native) identifiers. 
"Well-behaved" almost has to become more narrowly defined than the 
"anything goes" or "any PVALID goes" from E-mail or domain name standards.

Quite right.  I wish I know where the bad idea came from that you're
supposed to accept every technically valid but illegible and
misleading IDN or e-mail address.  It's never been true with ASCII
addresses and it's even less true with IDNs and UTF-8 mailboxes.

How do we make sure that UA doesn't become synonymous with "uncritical acceptance"?

A./

In a previous email you wrote: "Those issues may require a discussion whether "Universal Acceptance" and "Uncritical Acceptance" are the same thing". (Note - I Title cased for better effect)

Seeing the two together in a sentence is extremely effective and to me effectively summarises the issues.

I consider defining good mailbox naming practices for the whole of Unicode would be a huge undertaking and there would be many overlaps and possibly contradictions with other standards.

So, instead, what of defining good mailbox naming practice for just one of the orthography categories? This could serve to illustrate both good practice (Universal Acceptance) and bad practice (Uncritical Acceptance).

André Schappo