RE: draft-danisch-dns-rr-smtp-02.txt

["Hallam-Baker, Phillip" <pbaker@verisign.com>]

> is going to be worthy of some comment I suspect (and from a quick
> glance, probably more).

OK this group has decided no more compressed addresses, I think you could be
a little more constructive and point it out. This is a serious proposal
receiving serious attention. It is in the interests of everyone to make sure
it works as well as it can.

The use of RMX type concepts was mentioned by Microsoft at the FTC workshop
as an avenue they are working on with Yahoo and AOL. With the ASRG having
collapsed and over half the members left it would be a very good thing if
members of this group took a look.

I hope that here we can at least have a discussion on the issue without
having people called liars, described as 'snakes' etc and every constructive
proposal pulled apart by people whose real interest is killing competition
to their own schemes.


> I gave up on careful reading at ...
> 
>    The suggested method is to let the DNS server for the sender domain
>    provide informations about which IP addresses are authorized to use
>    the domain within a sender's address.
> 
> which I cannot find any possible way to implement.   That is, 
> I have no idea
> at all how I'd specify which IP addresses are authorised to 
> send e-mail
> claiming to be from munnari.oz.au - as that set is more or less, the
> IP address where I happen to be this week, and can be anything.

That is not an issue, the only clients that need to use this protocol 
are spam filters and they are not going to simply reject email for not
having an RMX.

What will happen is that the spam filter will change the estimate
of spam probability in response to the presence of valid RMX, the
presence of invalid RMX and abscence of RMX.

So if an email purports to be from hotmail.com we might have the following
probabilities:

A) Comes from RMX IP address           5% probability spam
B) Does not come from RMX IP address  95% probability spam
C) RMX not available                  50% probability spam

In case A the probability is probably low enough to simply whitelist (note
this is assuming that the spam filtering is stateful in the extreeme and the
probability for bozo.com with RMX might be 50%, exactly the same for no
RMX).

Case B is not quite high enough to reject (5% false positive rate is
unacceptable). But the threshold for rejection on content inspection would
be much lower, we only need to be about 90% sure it is spam to reject at an
acceptably low false positive rate, for case C we would have to be 99.5%
sure it is not spam to reject with the same level of accuracy.

We are not looking for perfect accuracy here, systems with one false
positive per 100 messages are usually acceptable. Some people are claiming 1
in a million false positive rates but I have not seen that substantiated.

> What's more, usually, wherever I am, I use whatever is the local mail
> relay (check this message, you don't find it coming from anywhere near
> munnari.oz.au - but it is also not forged, I promise...).   

OK, you are going to end up having problems sending your email reliably in
future as spam filters become steadily more aggressive.

RMX is only one authentication option. You could use SSL or S/MIME to
authenticate your email and your IP address of the week scheme will work.


The problem of forged sender addresses tends to be concentrated at the large
providers though. The spammy #$&#$(*&s choose addresses like hotmail or aol
because they are familiar to people and more likely to get replies. So a
scheme that only works for large ISPs but has very low deployment cost can
help reduce the problem.

[They are also hijacking the email addresses of people who are active trying
to stop spam, I got 15 bounces today from emails I never sent...]

> What's more, to assist in reducing, as the claim is that this will do,
> everyone has to do it - it is no good just some small 
> fraction of sites

Actually it does have an effect even if only 3 ISPs deploy if they are the
right ones. A huge proportion of the spam received at AOL purports to come
from hotmail and vice versa.


The issue that I am somewhat more concerned about is what might happen if
the spam senders start to react to RMX by trying to hijack used IP address
blocks. So far they have tended to hijack unused ones but RMX would create
an incentive to either hijack real IP addresses or attack the DNS system.

		Phill

--
to unsubscribe send a message to namedroppers-request@ops.ietf.org with
the word 'unsubscribe' in a single line as the message text body.
archive: <http://ops.ietf.org/lists/namedroppers/>