Re: Identities and Bcc:

[Jules Bean <jules jellybean co uk>]

On Thu, Aug 30, 2001 at 11:08:35PM +0200, Anders Larsen wrote:
> On 2001-08-29 15:04:52 +0200 Brian Stafford wrote:
> > Scream!!
> > 
> > On my last message on this topic, I forgot to remove the Bcc: I put in
> > to test the identity Bcc: stuff.
> > 
> > The message was indeed transmitted twice.  The Bcc:ed one was received
> > with one of the mail hops having helpfully stripped the Bcc: header.
> > 
> > If there are any MTA implementors listening, please, please, please
> > do not permit this non-conformant behaviour.
> 
> Oh, but that's indeed *conformant* behaviour!
> 
> The Bcc: header is *not* supposed to be sent - think about it:
> Bcc means *Blind* Carbon Copy - which would be worthless if any
> recipient could just do a "View Source" and see the supposedly
> hidden names of other recipients...

On the contrary: MTA must never, under any circumstance, modify a
message in *any* way except adding a Received header. You'll find it
in the relevant RFC.

You're right that Bcc: removal has to take place at some point, but
the correct point is the MUA: by the time the message has reached some 
intermediate MTA, it should certainly not be messing with the headers.

Jules