It seems that somewhere between VM 8.0.9 and 8.0.11, MIME (RFC 1522) encoding
of headers broke:
VM 8.0.9 would encode the word "Rückstoßdämpfung" in the Subject line as
follows:
=?iso-8859-1?Q?R=FCcksto=DFd=E4mpfung?=
which is still somewhat readable to the unaided eye. In VM 8.0.11 this now
becomes the following:
R=?iso-8859-1?B?/A==?=cksto=?iso-8859-1?B?3w==?=d=?iso-8859-1?B?5A==?=mpfung
Setting vm-mime-encode-headers-type to Q doesn't help much:
R=?iso-8859-1?Q?=FC?=cksto=?iso-8859-1?Q?=DF?=d=?iso-8859-1?Q?=E4?=mpfung
If I interpret RFC 1522 correctly, the method of VM 8.0.11 is plain wrong here,
since an encoded-word must be surrounded by white space and is not allowed to
start in the middle of a word (RFC 1522 section 5):
| Ordinary ASCII text and encoded-words may appear together in the same
| header field. However, an encoded-word that appears in a header field
| defined as "*text" MUST be separated from any adjacent encoded-word or
| "text" by linear-white-space.
And of course the MIME decoder shouldn't look for "=?" in the middle of a word,
but display the above as plain ASCII text (RFC 1522 section 6.1).
