From: Alex van den Bogaerdt (alexergens.op.het.net)
Date: Mon Jan 12 2004 - 08:57:19 CST

• Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]

On Mon, Jan 12, 2004 at 08:13:02AM -0500, John Peach wrote:

>>>> It will stop the acceptance of forgery between domains and MTA that deploy
>>>> the capability. Those that choose not to participate will remain sources
>>>
>>> It is broken because it binds envelope sender addresses to IP
>>> addresses. This is obviously bogus and works only for very simple,
>>> restricted cases.
>>>
>>> Does the post office require you to only send mail if the sender address
>>> has a zip code that matches the area where you submit the mail?
>>
>> No, the post office adds its own indication of the source of the mail,
>> in the form of the postmark. With certain exceptions in large cities,
>> it identifies the 5-digit ZIP code of injection.
>>
>
> OK - so deals with the US; what about the rest of the world?

I don't know about the rest of the world, but where I live it is
unacceptable to take the envelope, rewrite "rcpt to" and put the
letter back into the mail as done in forwarding.

It is acceptable to bounce the mail and advice the postman about
the new address if available (relocated). This is different from
forwarding.

Forwarding the mail would go like this:

a) I take another envelope, fill it with the original package
   (letter + envelope) and mail it
or

b) I take the letter out of the envelope, stick it in a new
   envelope and mail it.

In both cases a new envelope is introduced and I, the forwarding
entity, will have to deal with bounces.

If such a forwarded letter would bounce, I have the option of
bouncing the original envelope.

If this is going to be used as an analogy, forwarding email as
currently done is _not_ how forwarding "normal" mail is done.

just another 2c.

Alex van den Bogaerdt

• Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]