On Tuesday 15 August 2006 14:54, Basil Chupin wrote:
William Gallafent wrote:
Message-Id: <200608150938.47561.william@gallaf.net>
To me this is not enough information to find the exact message you are referring to.
That's OK, I did it here too :)
But, have you looked at the header to your message (the one I am replying to)? For some inexplicable reason it has a Repy-To field, and it is "suse-ot@suse.com" which means that a SuSE server put in the Reply-To field.
Allow me to explicate: I (being the author) put that header in myself - it wasn't put there by the server. To quote our favourite RFC, 'When the "Reply-To:" field is present, it indicates the mailbox(es) to which the author of the message suggests that replies be sent.' So, I used that header to express my suggestion that this thread would live better in the suse-ot list. Admittedly, I didn't really expect that to happen :)
I said that a person's message should (and it really should be "must") not be interfered with during transmission and arrive at the destination as originally sent. (Proviso: all ISPs reserve the right I believe to delete a message if it contains a virus et al.)
However, this does not mean that after its arrival the server on which it has arrived cannot add some field(s) which do not interfere with the original message.
When has a message "arrived"? when it gets to the SuSE server, or when it gets to a subscriber to the list? (I say the latter, and that the former is really a sort of glorified intelligent relay of some kind ;)
If there is no Reply-To field in the original message then the server can add one (Reply-To: SuSE server - just like to your message).
Critical point being that _I_ put that header in, not the server! The server reproduced it faithfully, fulfilling the "no interference" rule you state.
If there is a Reply-To field already then the server honours it and does not insert its own field. QED.
Here we differ. I interpret the Reply-To header as the email address which the _author_ suggests I send my reply to. If the list server adds it, it's misleading me in to thinking the author intended the reply to go to the list, whereas in fact that was the intention of the list server, and the author's decision to omit the Reply-To header has been overridden.