Hey mark!

 ml> systems that display the message's written time in local time or
 ml> UTC if configured to do so

I've yet to see any BBS get that right if indeed the BBS software can be
configured to do so.  Do you have or know of a working example that indeed gets
 it right?

 ml> you cannot convert to/from local/UTC if the TZ isn't known

Understood.  Speaking for myself, I have never bothered and don't use the FTN
datetime for anything, nevermind converting it even if only display purposes. 
What I was taught AGES ago is the data is the data even when it is obviously
wrong, and that to tamper with raw data is EXTREMELY vorbotten.

If I cared, for local display I would do something like this (using the
datetime and TZUTC from "MSGID: 1:3634/12.73 5c3e3137";

  TZ=UTC date --date="15 Jan 19  14:13:10 -0500"

which outputs "Tue Jan 15 19:13:10 UTC 2019" on the machine I am currently
using to reply.  Assuming a FTN compliant "TZUTC: 0000" that translates to;

  TZ=UTC date --date="15 Jan 19  14:13:10 0000"

which outputs "date: invalid date '15 Jan 19  14:13:10 0000'".  However if I
add the '+' character to the start of the offset string then it outputs, "Tue
Jan 15 14:13:10 UTC 2019".  In other words proper datetime applications based
on the strftime function will only work with utc offset data with the correct
prefix for east (+) or west (-) of prime meridian.  Your mileage may vary.

Note that this reply doesn't contain a CHRS control line given that it is pure
ascii.  :-)

Life is good,
Maurice

... Don't cry for me I have vi.
--- GNU bash, version 5.0.0(1)-release (x86_64-pc-linux-gnu)
 * Origin: Little Mikey's EuroPoint - Ladysmith BC, Canada (2:280/464.113)