Long time ago... :) a more or less effective method of fighting spam was found by network people - it is a public database of well-known spam-sending hosts and networks. But only few of such projects continuing to feed more or less adequate information. You can find some of them on our links page.
As "black" spam sending market grows, antispam tools becomes more and more effective, but spammers smarter and smarter. As of today, this process is not stopped, and we think, that the only right way here is to develop fully automated, intelligent on-line process of detecting and recoginsing attempts to push the spam.
We have a big installation base of a anti -spam and -virus systems (project page). They are receiving around 15 connections per second. Big part of that connections are spam-sending (more than 99%). After collection, recognition and processing we put this information to RBL-style database. And all of that info becomes available to all of our systems and clients, which rises ability to detect and drop the spam connections.
We do not try to concurent with existing databases, but trying to analyse and process data not containing in such databases, rising ability to drop the spam.
Our database have a dynamic nature, which assumes not only on-line additions, but also dynamic information expiration. For example, some host sends a spam, information about this fact enters to our database (while this host is spam-active), then host is cured and stopped sending spams. In that case after 26 hours since last activity was detected, database expires and removes information about that host. So, RURBL database is fully dynamic and precisely follows spam-activity state changes in covered address ranges.
Because most of our systems DCC://Box are installed in RUNet, a russian portion of Internet, and, as of this, collected information effectively reflects attempts to send a spam in russian language. Definetly ALL information is counted. One more important fact, that our systems installed in many russian AS address blocks, that gives very fast spammer detection in case, when IP addresses are scanned by AS blocks.
You can use information slowly, but free (with standard RBL client, using db.rurbl.ru as RBL domain), or at full speed, paying some compensation.