The Apache Software Foundation is taking the spam fight to a new level — literally — with the promotion of its Spam Assassin project to top-level status.
Prior to Thursday’s vote by the ASF to move it to top level project (TLP), Spam Assassin had been in incubation.
Spam Assassin (SA) is a mature, widely deployed open source project that acts as an e-mail filter in order to identify and block Spam (UCE) by using a diverse array of tactics.
The spam identification tests include header and text analysis, as well as support for blacklists and the Razor spam tracking database (http://razor.sf.net).
“I think spam filtering is now a critical part of the network
infrastructure and Spam Assassin is a leader in the area,” said Daniel Quinlan, chairman of the Apache Spam Assassin Project Management Committee. “So, I think we bring another well-regarded project with a very large installed base into the ASF. Not to mention that as of earlier this week, the ASF is now running Spam Assassin on our mail servers,” he told internetnews.com.
“Spam Assassin has become the Sendmail of the anti-spam world,” said Jeremy Zawodny, an early supporter of the project and community member. “It’s easy for administrators to setup, very flexible, and it simply identifies more spam than just about anything else out there,” he told internetnews.com. “SpamAssassin is clearly a leader in the anti-spam fight.”
The actual “Spam Assassin” trademark was previously owned by Network Associates (NAI) and has now been assigned to the Apache Software Foundation, which, according to SA developer Michael Parker, is among the reasons the project moved up to full Apache project.
“We’ve been incubating as an ASF project for awhile. I don’t know all of the original reason for wishing to become an ASF project. I know one was resources. SourceForge just wasn’t cutting it as a hosting provider,” Parker said.
The other reason cited by Parker and Quinlan for SA becoming a TLP at this time is that the project is almost ready to release its next version. “Spam Assassin 3.0 is a major milestone in our development,” Quinlan said.
Version 3.0, which is expected to be officially released as a stable version in about a month, includes a number of improvements such as a new architecture for plugin modules. It features both Bayes and autowhitelist SQL database support, as well as support for the Sender Policy Framework (SPF).
“SPF is another tool in our toolbox that we haven’t had before and it lets us better know if a particular message is really from the sender,” Quinlan explained. “The main thing is that it helps make identification of forged email addresses easier and that is a common spam tactic.”
The vote comes one day before Microsoft joined with the author of the Sender Policy Framework (SPF), Meng Weng Wong, to submit their merged e-mail authentication standards to the Internet Engineering Task Force. The protocol is called Sender ID.
According to May statistics on Spam, the majority of e-mail traffic remains unsolicted. Brightmail reported that 64 percent of all e-mail was spam while Postini measured the volume at 78 percent.
