What does "Convert ... upon sending" mean?

[Ken Moffat]

On Tue, Apr 23, 2019 at 03:48:39PM -0600, Paul Gilmartin wrote:
> 
> ... notice the 5 non-ASCII octets, a0, b0, c0, d0, and e0 in the last
> two lines.  After a trip through AOL:
> 
> 512 $ od -tx1 *SCII.txt | head -17
> 0000000    20  20  20  20  48  6f  73  74  3a  20  49  42  4d  2d  31  30
> 0000020    34  37  20  20  6f  75  74  70  75  74  3a  20  66  72  6f  6d
> 0000040    5f  49  53  4f  38  38  35  39  2d  35  0a  20  20  20  20  20
> 0000060    20  20  20  20  20  30  20  20  31  36  20  20  33  32  20  20
> 0000100    34  38  20  20  36  34  20  20  38  30  20  20  39  36  20  31
> 0000120    31  32  20  31  32  38  20  31  34  34  20  31  36  30  20  31
> 0000140    37  36  20  31  39  32  20  32  30  38  20  32  32  34  20  32
> 0000160    34  30  0a  20  20  20  20  20  20  20  20  20  20  30  20  20
> 0000200    31  30  20  20  32  30  20  20  33  30  20  20  34  30  20  20
> 0000220    35  30  20  20  36  30  20  20  37  30  20  20  38  30  20  20
> 0000240    39  30  20  20  61  30  20  20  62  30  20  20  63  30  20  20
> 0000260    64  30  20  20  65  30  20  20  66  30  0a  0a  20  20  20  30
> 0000300    20  20  30  20  20  20  20  20  20  20  20  20  20  20  20  20
> 0000320    20  20  30  20  20  20  40  20  20  20  50  20  20  20  60  20
> 0000340    20  20  70  20  20  20  20  20  20  20  20  20  20  20  ef  bf
> 0000360    bd  20  20  20  ef  bf  bd  20  20  20  ef  bf  bd  20  20  20
> 0000400    ef  bf  bd  20  20  20  ef  bf  bd  20  20  20  ef  bf  bd  0a
> 
> 
> ... each non-ASCII octet has been changed to Replacement, "ef  bf  bd".
> 
> > Can you deliver the same message attachments through 2 SMTP services (AOL and another) ...
> >  
> I can use 3 SMTP servers: AOL, Yahoo, and GMail.  The problem occurs with
> AOL, not Yahoo or GMail.  The behavior does not depend on which IMAP server
> I use with a viewer.  I blame AOL's SMTP.
> 
I think you are correct - going back to your original post, the
trashed characters were those which could not be associated with an
ASCII (or perhaps 8859-1) glyph.  So what I assume were cyrilliC A,
Je (J), Dze (S) were rendered.

What I personally found interesting was that in my UTF-8 locale, all
of the cyrillic glyphs in the 'good' version rendered fine (in urxvt)
which caused me to wonder if it had been converted to UTF-8 on the way.

The replacement (U+FFFD) is often rendered as a question-mark in
reverse-video (e.g. in a diamond or other shape) in fonts, although
some fonts are stupid enough to not do that.

ĸen
-- 
With a few red lights, a few old bits, we made the place to sweat.
No matter what we get out of this, I know, I know we'll never forget.
Smoke on the water, a fire in the sky.  Smoke, on the water.