Spamming the universe

Already feel like you're drowning in a sea of spam? Experts predict it will only get worse, more than doubling by 2007.

According to spamfilterreview.com, more than 31 billion e-mails are sent daily, with 40% of those qualifying as spam. It's estimated that by 2007, the amount of spam will increase by 63% over current amounts. Annually, individual users receive 2,200 pieces of spam and 16% of all e-mail address changes arise in an effort to escape spam.

The Can-Spam Act took effect January 1 after Congress responded to the public outcry spawned by abuses taken by commercial organizations at the expense of unsuspecting e-mail users. The primary provisions of the act (enforceable by the Federal Trade Commission and the Department of Justice) ensure that commercial e-mailings do not contain misleading or fictional header information, including the "To" and "From" lines and routing information. Other provisions require that e-mails do not contain deceptive subject lines and provide recipients with an "opt-out" method.

In addition to the support provided via the Can-Spam Act, organizations are also seeking in-house solutions to the problem of spam. With large businesses (more than 1,000 employees) spammed an estimated 2.1 million times in a year, the financial incentives are estimable. For each piece of spam received at work, it's estimated that employees lose 4-5 seconds of productive work time. When multiplied by the millions of spam messages received by businesses in a year, the resulting billions of dollars are an indisputable impetus for change.

Using statistical techniques to examine key words and phrases, filters can classify incoming mail as spam or non-spam. Known as content filtering, this technique is used to sort e-mail for users. A drawback to this approach: Filters sometimes incorrectly deem valid e-mail to be spam. For those organizations that refuse to miss even one legitimate e-mail this means extra time must be taken to see to it that e-mail identified as spam is not deleted outright. The e-mails labeled spam can be collected for periodic review so that the legitimate e-mail can be pulled out. But this gives rise to the question: Isn't this defeating the purpose of content filtering?

There are a variety of proactive approaches for reducing the influx of spam. If spam should make its way into your users' mailboxes, make sure they know not to follow any links in the text. Just replying to an address or even simply clicking on a link within the message will confirm that your address is valid. Remember this when presented with an "unsubscribe" option, as these are often another means for spammers to collect valid addresses. Another tactic is to disable the automatic downloading of graphics in HTML mail, because again, spammers sometimes send these in the hopes your e-mail client will download the graphic from their Web server, thereby verifying that your address is valid and active. You can achieve the same result by disabling HTML mail entirely and viewing messages in plaintext only.