> !   3.4.9.  BACKSPACE CHARACTERS
> !
> !      ASCII BS characters (Backspace, decimal 8) may be included  in
> !      texts and quoted-strings to effect overstriking.  However, any
> !      use of backspaces which effects an overstrike to the  left  of
> !      the beginning of the text or quoted-string is prohibited.
> !
> !   August 13, 1982              - 15 -                      RFC #822
>
>   I understand this to mean that a backspace may not occur as the
>   first character of a line of text.
 
The proposed backspace was not intended to effect an overstrike to the left,
but rather to delete the preceeding newline element.
 
A serious question is if the above RFC822 item would have led mailer designers
to scan the mail looking for "prohibited backspaces" and assume that if a
backspace immediately followed a newline then it was intended to backspace
to the left of column 1 and is thus prohibited and the sender should be
flogged.  It seems to me that what Victor has done is identify a sequence
of elements (e.g. newline then backspace) that has no pre-existing useful
purpose, and used them to solve and existing problem, in a logical way.
 
Since the problem primarily exists in environments restricted to 80 character
line lengths, I would say that such enviroments should implement this as a
"transparent layer" in which the used of the backspace on continued lines
REPRESENTS a single longer line to which RFC822 is the applicable protocol.
On mail ENTERING the 80 character environment, this layer is added, and on
mail exiting the 80 character environment, this layer is removed.  No one
outside the 80 character environment ever need see this, so the "serious"
question about other implementations having a problem with this is drastically
reduced.