On Wed, 2004-09-15 at 11:20 -0400, terry(_at_)ashtonwoodshomes(_dot_)com wrote:
[ Tony Finch wrote: ]
You do not understand the problem. The difficulty with
forwarding occurs at the receiving sites, not the sending
sites that publish Sender-ID records. If you use -all you
must never send email to an alias-fowarding address --
but you have no way of knowing which addresses these are.
Valid point.
With more thought, you are correct, sorry.
Good. Some are rather disingenuously still claiming not even to see
that.
I still support -all, because:
Another way of looking at it is, if the receiver wants to forward
other (valid) email address(es) to his/her actual email address, the
onus is on the recipient to ensure that the MTA doing the forward
on his/her behalf has the necessary mechanisms.
Even assuming that the deployment of such mechanisms is acceptable, it
will take _time_ for them to be deployed after the publication of the
RFC which mandates them.
It is the responsibility of this working group to enable
interoperability between sites; we must not _simultaneously_ publish a
requirement for such new mechanisms and permit sites to assume that the
same mechanisms are _already_ deployed. That would be extremely
irresponsible and foolish.
For this reason, I believe that use of the '-all' record on domains
which actually send mail could be described, but must be _temporarily_
prohibited, until such time as it is determined that a large enough
proportion of the world has implemented these mechanisms of which you
speak.
--
dwmw2