Recycle your Computers & Technology with us.

Message Rules Not Working For Spam (Part 2)

Posted By : of Data Doctors on April 27, 2005

Follow us on Facebook   Follow us on Twitter   Follow us on LinkedIn

Let Data Doctors be your personal IT department today

In my last column, I explained why Message Rule “keyword filtering” was useless against large scale spammers.

This question was answered on April 27, 2005. Much of the information contained herein may have changed since posting.


Spam filtering has become a major battlefield between spammers and the companies that develop tools for fighting the problem Unfortunately, the spammers have the upper hand because they can take any of the tools and methodologies being used successfully and figure out ways to get around them.

Since the problem is not getting any better (and won’t anytime soon) we as users must take on the task of reducing the amount that we receive through filtering.

There is no such thing as the perfect spam filter; they generally filter out some good mail (ham) or let in some junk mail (spam) or in some cases both Here is an overview of some of the current options:

ISP based filtering – This is a filtering system that is developed and maintained by your Internet Service Provider This removes the burden of choice from the user, but it also makes it difficult or in some cases impossible to know if desired mail is getting filtered Depending upon the technology used by the ISP, this method can be fairly effective or overbearing Before activating an ISP filter, be sure you have a way to see messages that were tagged as spam

White Lists – This filtering can be created in most e-mail programs (look in the Help section) and only allows messages from addresses that you have listed as acceptable on your “white list” The biggest problem with this method is that you have to know in advance that someone, not on your list, is trying to send you a message or you have to constantly search through all your filtered messages looking for those that you want to add to your white list.

Challenge/Response – This method has gained much popularity because it requires the sender to verify themselves (only the first time) by responding to a challenge Automated systems don’t have a way to respond to the challenge, therefore they don’t get through The down side to these systems is that unless you can add a white list of approved senders, you will never get airline confirmations, eBay auction updates, newsletters, hotel confirmations or virtually any response to an online transaction Systems that have both are available as trial downloads at www.qurb.com and www.mailfrontier.com

Bayesian filtering – These tools use mathematics to calculate the probability of a message being spam based on its contents Unlike simple content-based filters, Bayesian spam filters learn from what you consider spam and good mail, resulting in an adaptive anti-spam tool that gets better over time Free Bayesian filter downloads are available at www.spampal.com, http://popfile.sourceforge.net and www.spamihilator.com and a trial download of SpamCombat is available at www.glocksoft.com

Whatever you use as your method of filtering, no matter how high the effectiveness is today, it’s likely to be less effective over time

The systems in the last two groups allow you to tweak the filtering system and will be useful for a longer period of time because you can adjust as the spammers methodologies change AND they will.

By no means have I covered every product or process available, so if you have found a product or service that is particularly effective, let me know.

About the author

Posted by of Data Doctors on April 27, 2005

Need Help with this Issue?

We help people with technology! It's what we do.
Contact or Schedule an Appointment with a location for help!