After discussions on this list and in person at the IETF, both (A)
and (B) seem more problematic than the 'delsp' attribute (which was
dismissed early on as being too ugly). Since it avoids the problems
with the other proposals, there seems to be general agreement that
it's the way to go (despite the ugliness).
At 11:43 PM -0600 3/20/02, Randall Gellens wrote:
we could add a new attribute: delsp, which is either true or false
(false by default of course). When true, the space as well as the
CRLF is to be deleted on reception.