Hello mark,

On Tuesday June 09 2020 11:08, you wrote to me:

 ml> @PID: Synchronet 3.18a-Linux  May 23 2020 GCC 7.5.0
 ml> @TZUTC: -0400
 ml> @TID: SBBSecho 3.11-Linux r3.173 May 23 2020 GCC 7.5.0
 ml> @CHRS: CP437 2
            ^^^^
Note that your software marked your reply to my message as encoded in CP437.

Look here:

 MvdV>> QWK/BW was never popular ìere and AFAIK was never used by points.
                                 ^

Instead of an 'H' we see an 'i' with accent grave. The infamous hex 8D in
CP437. My mesage however was encoded in CP866. That is Cyrillic and in CP866,
hex 8D is the letter pronounced as 'i' in 'Putin" which looks very much like
the latin 'H'. So someone using a reader that propely displays CP866, would not
have noticed anything odd other than the capitalisation.

When answering my message your software sees a character >127 which is not part
of a well formed UTF-8 sequence and wrongly decides it is CP437.

This is exactly what I expected to happen to your software when an encoding
other than ASCII, CP437 or UTF-8 is used. It illustrates why I stated that
automatically determining the encoding after it is out of the hands of te
writer is not a good strategy.

QED.

Cheers, Michiel

--- GoldED+/W32-MSVC 1.1.5-b20170303