On Mon, Mar 03, 2003 at 04:00:29PM -0500, Bob Wyman wrote:
Single-user addresses since they require both sender and receiver to
have mail systems that understand the additional header fields. On the
other hand, handling of a single-user address of the form
"username+signature(_at_)example(_dot_)com" is something that can be deployed
immediately by any mail recipient who has a mail server capable of
ignoring the signature part when determing who the mail is destined for.
Thus, if you use sendmail for SMTP, or you use Earthlink or a number of
other mail providers, you can create and use single-user addresses NOW.
All you have to do is put the single-user email address into the
"reply-to" and "From" fields of your messages.
...
result in minimal disruption to the existing protocols. The single user
address (especially if based on the "+" convention) allows the method to
be used with NO modifications to existing protocols.
I disagree.
There are many MUAs and mail handling programs which can't cope with
+ mail addresses and fail to sort mails, recognize senders etc.
Maybe your MTA can transport these mails, but lots of mail programs
will fail. E.g. the complete mail routing software running on
the rackland server will fail.
Just as an experiment:
I just sent this message to the asrg(_at_)ietf(_dot_)org mailing list with
a sender address
From: hadmut+45852e998cdb18c752c9d9b6cf77aef7(_at_)danisch(_dot_)de
The mailing list processor didn't accept it and queued it
for approval. So these + mail addresses don't even work on this
mailing list. So your assumption of "minimal disruption" is wrong.
In contrast, a new header line is fully transparent and really can
be handled by any MTA and MUA.
Hadmut
_______________________________________________
Asrg mailing list
Asrg(_at_)ietf(_dot_)org
https://www1.ietf.org/mailman/listinfo/asrg