[Standards-JIG] anti-SPIM advantages

Tony Finch dot at dotat.at
Mon Sep 25 13:50:58 UTC 2006


On Fri, 22 Sep 2006, Paul Johnson wrote:
> On Friday 22 September 2006 09:32, Tony Finch wrote:
> > On Fri, 22 Sep 2006, Hal Rottenberg wrote:
> > > On 9/22/06, Tony Finch <dot at dotat.at> wrote:
> > > > DNS blacklists are a counter-example.
> > >
> > > They can be abused certainly, but I disagree that they are a
> > > "counter-example".
> >
> > What do you mean? You seem to agree that they are a widely deployed,
> > interoperable, and effective anti-spam standard, albeit de facto.
> > Ian was saying that such a thing does not exist.
>
> SPEWS, SORBS...

The existence of badly run blacklists does not mean that well-run
blacklists do not exist, and it does not mean that well-run blacklists are
not an excellent first line of defence, and it does not mean that there is
not a standard way of querying blacklists, and it does not mean that there
are not multiple interoperating implementations of this protocol.

The fact that badly-run blacklists exist and are sometimes widely used is
unfortunate, especially since they are the only working example of a
reputation system that we have. The strong high-level identities provided
by DKIM and XMPP are supposed to make it easier to run a good reputation
system, but the best blacklists are generally based on low-level
properties, such as automatic detection of infected machines or machines
running open proxies or relays. So there's a lot of work still to be done
here.

Tony.
-- 
f.a.n.finch  <dot at dotat.at>  http://dotat.at/
FORTIES CROMARTY FORTH: SOUTHERLY 6 TO GALE 8, DECREASING 5 OR 6 LATER. RAIN
OR SHOWERS. MODERATE OR GOOD.



More information about the Standards mailing list