Hi again (and again and again, some will say..).
Since my message of a few hours ago, I have re- and reread both the draft
and RFC822.
The expression 'body part' never appears in RFC822. 'message body',
'(a) message's body' or 'the body of a message' are used.
There are not many occurence where 'body part' is used deliberately to
mean 'the part of message that is not the header' in the current draft.
But there are some, and some very ambiguous cases.
So another solution is to suppress those, and keep 'body part' to mean
'a section of the body of a multipart message.
In all cases, I still think that the term used in the BNF should match
the one in the text (it could be body-part in the BNF).
But the most important thing is to fix the many occurences of 'body or
body part', 'message body of body part', because they are really misleading.
At places, the use of this expression can lead to thing that a 'body part'
may be encoded (which is of course false : only the body of a body part
may be encoded, the header must stay in clear).
/AF