This is a "me too" message so, but since I add one suggestion at the end,
I am passing it to the whole list, though I would like to see this thread
die.
When I first came across the description of exim's bcc behavior, I thought
it was odd. But a little bit of thinking about it convinced me that that
PH is absolutely correct. The MTA should do nothing (with the exception
of header rewrites for locally or near-by generated messages) to
originator headers. The Bcc, like and Fcc, is a pseudo-compose header and
should not go into the 821 headers. It should give information to the MUA
about what to do with the message, but shouldn't be among headers passed
to the MTA.
If the MTA gets a Bcc, or an Fcc, then it should assume that that is what
the user wanted to appear in the headers and pass it on unmodified. The
MTA has the *responsibility* to pass it on. Only when the MTA is being
used as an MUA (with the -t flag), should it strip the Bcc.
So, PH, please don't change the behaviour of exim here. It is correct,
but it did take a little bit of thinking about.
I wasn't aware until this thread started up that there were broken MUAs
which passed on Bcc headers as headers. It might be worthwile to identify
such MUAs and list them in some auxilliary documentation (along with MUAs
that are prone to generating syntactically incorrect headers).
-j
--
Jeffrey Goldberg +44 (0)1234 750 111 x 2826
Cranfield Computer Centre FAX 751 814
J.Goldberg@??? http://WWW.Cranfield.ac.uk/public/cc/cc047/
Relativism is the triumph of authority over truth, convention over justice.
--
*** Exim information can be found at
http://www.exim.org/ ***
