Skip site navigation (1)Skip section navigation (2)
Date:      Mon, 16 Nov 2020 19:10:24 +0000
From:      RW <>
Subject:   Re: Booting multiple BSDs.
Message-ID:  <>
In-Reply-To: <>
References:  <> <cmu-lmtpd-1514294-1605257963-2@sloti22d2t03> <> <> <>

Next in thread | Previous in thread | Raw E-Mail | Index | Archive | Help
On Mon, 16 Nov 2020 12:49:00 +0100
Polytropon wrote:

> On Mon, 16 Nov 2020 09:45:01 +0000, Thomas Mueller wrote:
> > > > > Harry Weaver =20
> > =C3=A2=C2=A2=C2=80=C3=A2=C2=A3=C2=B4=C3=A2=C2=A0=C2=BE=C3=A2=C2=A0=C2=
> > =C3=A2=C2=A3=C2=BE=C3=A2=C2=A0=C2=81=C3=A2=C2=A2=C2=A0=C3=A2=C2=A0=C2=
=92=C3=A2=C2=A0=C2=80=C3=A2=C2=A3=C2=BF=C3=A2=C2=A1=C2=81 Debian - The univ=
ersal operating system
> > =C3=A2=C2=A2=C2=BF=C3=A2=C2=A1=C2=84=C3=A2=C2=A0=C2=98=C3=A2=C2=A0=C2=
=B7=C3=A2=C2=A0=C2=9A=C3=A2=C2=A0=C2=8B=C3=A2=C2=A0=C2=80 https://www.debia=
> > =C3=A2=C2=A0=C2=88
> > which is not what I'm running now. =20
> Looks like there is still more than one 2-to-1 byte mangling
> involved, possibly crossing encoding borders. ...

What you saw was misleading, and has nothing to do with why Thomas
Mueller didn't see the Debian thing.

His client likely just has broken or missing UTF-8 and/or MIME support.
There's no excuse for that in 2020.

> The original message declares UTF-8 correctly, and also _uses_
> UFT-8 encoded symbols in the signature. The following drawing
> should be visible:
> 	=E2=A2=80=E2=A3=B4=E2=A0=BE=E2=A0=BB=E2=A2=B6=E2=A3=A6
> 	=E2=A3=BE=E2=A0=81=E2=A2=A0=E2=A0=92=E2=A0=80=E2=A3=BF=E2=A1=81
> 	=E2=A2=BF=E2=A1=84=E2=A0=98=E2=A0=B7=E2=A0=9A=E2=A0=8B
> 	=E2=A0=88=E2=A0=B3=E2=A3=84

That was visible to me in the post that you quoted from.

What happened was that Thomas Mueller pasted UTF-8 into a text/plain
body without an explicit character set. The relevant RFC says that this
should default to US-ASCII, which means that byte values above 7f are
undefined. This include all of the UTF-8 Braille.

My client, Claws Mail, identified the text as UTF-8 and displayed it as
such. UTF-8 can be identified very reliably is there are enough
multibyte characters.

Your client, Sylpheed, appears to have assumed the undefined bytes were
in ISO 8859-1, a common fallback. On replying, Sylpheed then converted
the quoted bogus text to UTF-8 which hopefully this reply preserves.=20

There may be a bit more to this, but any remaining weirdness is most
likely caused by Sylpheed. Note that some of the whitespace in the
quoted garbage above is from A0 bytes in the original UTF-8 being
treated as a "no break space".=20

Want to link to this message? Use this URL: <>