Re: OpenDKIM 2.3.0 Release = stable!

From: Murray S. Kucherawy <msk_at_blackops.org>
Date: Thu, 3 Mar 2011 09:46:06 -0800 (PST)

On Thu, 3 Mar 2011, Mark Martinec wrote:
> Murray is right, both are just filters. Each of them is capable of
> signing and verifying DKIM signatures on its own. I don't see any
> opportunity or need for integration, but I do see it as a benefit of
> having two independent DKIM implementations: in the past one or the
> other helped reveal problems in the other implementation, benefiting us
> all. Also a little competition does not hurt :)

I don't even see them as competing. I think they serve clearly distinct
"markets".

But one place where they might be able to co-operate would be a mechanism
by which they can share statistics. It might be useful, for example, to
be able to correlate sources of spam and/or viruses either by IP address
or by signing domain. OpenDKIM can do that now using the stats
extensions. I've been able to correlate signing domains with high spam
ratios, for example, by having OpenDKIM use Spamassassin results. Such a
thing could also be done for viruses.

In one of the upcoming OpenDKIM releases I'm hoping to start posting
regular updates to this data, when I can get reliable feeds from more
sources.
Received on Thu Mar 03 2011 - 17:46:20 PST