Re: Fwd: New Version Notification for draft-crocker-email-deliveredto-05.txt

On 8/12/2021 7:49 PM, John Levine wrote:
> It appears that Ned Freed  <[email protected]> said:
>> It's also different in that the use of the field in those MTAs is strictly for
>> loop detection, which makes the format a local matter. The draft aims for a
>> different - and in some cases disjoint - logging usage, which really calls for
>> a common syntax.
> 
> As far as we know, all of the existing usage is for loop detection, so it would
> be reasonable to say the contents of the header is a token or sequence of tokens

The strings we have documentation for do not look nearly as abstract as 
the term token implies.  Rather, they look like email addresses.

When you produce fielded examples of strings that look more opaque 
and/or random, a change in terminology might be worth considering.


> with semantics local to the MTA that adds the header. 

The strings we have documentation for sometimes demonstrate local 
idiosyncrasy, but only sometimes.

For the most part, they seem to reflect the transit destination email 
address.


> I suppose we could say something
> like what we say with Message-ID, to use a local domain to avoid collisions.

The current syntax already specifies use of a domain name.


> What is not reasonable is to overload this existing header with new, incompatible
> semantics and no way to tell the difference. 

Please provide details that substantiate a claim of incompatibility.

The current spec supports documented existing use, except for the 
example with preceding commentary free-form text.


  The obvious solution, of course,
> is to use a different header name.  If the goal is to record the SMTP envelope
> address, Envelope-To seems to match what Exim does.  If the goal is something else,
> make up a new header name.  It's not like they're hard to invent.

Constantly repeating the same statement of your preference is 
constructive how, exactly?


d/


-- 
Dave Crocker
Brandenburg InternetWorking
bbiw.net