ietf-mxcomp
[Top] [All Lists]

Re: Choice of SMTP headers

2004-03-24 07:54:08

Dave Crocker <dcrocker(_at_)brandenburg(_dot_)com> wrote:
There is a long track-record with making changes to the running
Internet. The track record is that new services -- as opposed to changes
to an existing service -- take 3-5 years to gain widespread deployment,

  Which new services have achieved widespread deployment in less than
3-5 years?  What properties do they have?

  I can think of a number of new protocols deployed at the edge, and
in wide use, in significantly less than 3 years.  But none of them
had IETF involvement.  Perhaps that's the biggest issue here.

MCL> SPF is an example of a server-side solution, with minimum impact on the
MCL> client and transparent to the end-user.

Let's see.

It requires end-users to pre-register the MTAs that they will use.

  SPF says nothing about end-users registering MTA's.  It talks about
domains listing MTA's in DNS.  The two statements are very, very,
different.

  And if end-users wish to use an MTA not listed in DNS for a domain,
they can talk to the domain owner.  That owner can either include
their domain (end-user registration of MTA with the domain), or the
owner can tell them to use one of umpteen existing protocols to send
mail through a system controlled by the domain owner.

  That is, SPF allows end-users to register MTA's.  It doesn't
"require" that registration.  A domain can publish a record saying
"permitted from 0/0", which means no registration is required for
anyone.

Otherwise, it breaks third-party MUA usage, and quite a bit of mailing
list usage, and otherwise-legitimate mobility scenarios.

  I guess that means SUBMIT isn't part of the solution.

  And the alternative is worse.  Do we really intend to permit
end-users to use a domains name without the consent of the domain
owner?  Worse, do we intend to FORCE the domain owner to accept that
behaviour, even if they want to restrict use of their name?  Do we
want to prevent the domain owner from being able to make that choice,
or publish that decision?

  RMX (SPF and variants) are about choice: permitting people to
cooperatively and consensually communicate.  If we decide that we
don't want to permit people to make those choices, then we are
imposing a dictatorial solution on them.  IMHO, that makes us just as
bad as spammers, who want to force their garbage on us despite our
stated non-consent.

  Alan DeKok.


<Prev in Thread] Current Thread [Next in Thread>