Charles Lindsey wrote:
I see nothing in RFC 2822 to indicate how that display-name is intended to
be used. Clearly it is useful to display it for the edification of users
(and that is actually all that is needed in this case to indicate to users
which of the various mailboxes in evidence is/are lists and which are
humans). But there is nothing to preclude MUAs from using that information
for any purpose, and there is no reason why the convention of using it
should not be enshrined in some future standard or BCP.
Try RFCs 1958 and 2277, paying attention to the difference between
protocol elements and text. Text is for human consumption and is
in some language; protocol elements are simply keywords. Display
names, comments, and the field bodies of unstructured fields are
text, which is why precisely those elements are the ones which may
contain encoded-words (RFC 2047/2231/errata). Every past attempt
to overload display names, comments, and/or unstructured field
bodies with some sort of protocol content has been an unmitigated
disaster.
Indicating a mailbox as a list isn't terribly informative; it may
be a list that expands to one mailbox (or none), the author may
have guessed incorrectly and indicated a non-list as a list or
vice versa, and
John List:john(_dot_)list(_at_)foo(_dot_)example(_dot_)net;
is at best ambiguous.