That hasn't stopped software developers from trying. The onslaught of unwanted email has inspired many types of filter tools, including email forwarding services, software plug-ins, and built-in filters for Web-based mail such as Yahoo! Mail or for applications such as Microsoft Outlook. Emailias, launched last fall, is designed to shield a consumer's primary email address from spammers. Emailias or other services, such as SpamMotel or Mailshell, allot an unlimited number of fake, or alias, addresses for the consumer to use when filling out forms, posting to newsgroups or signing up to mailing lists, where they can subsequently be "harvested" by spammers. For $4.95 per month or $19.95 per year, Emailias' plug-in sits in a browser's "favourite links" or on its task bar. When consumers are asked for an email address, they can click on the link to retrieve a pop-up window with an address specialised for that page. Email sent to that address is forwarded to the consumer's primary account. Subscribers can discontinue the address at any time -- for example, when an e-commerce company sends unsolicited mail from "partners." Another tool, Novasoft's SpamKiller, costs $29.95 and is one of the most popular tools at Download.com, a site run by CNET Networks, publisher of ZDNet UK. Among other filtering techniques, the software lets consumers block messages by the sender's address, message subject or headers, and message text. For example, customers can dump all email with the words "make money from home" within it. SpamCop, Kilgallen's choice, costs $3 monthly, with a free service for reporting spammers. It filters mailboxes based on "whitelists," or a list of acceptable addresses to receive mail from, and "blacklists," unacceptable sources of mail. The service filters the IP addresses used by rogue marketers in real time so complaints may help improve the filters. With the account, subscribers also get an alias address. Even with regular filter updates or new blocking inventions for consumers, however, spammers often find a way to infiltrate the most guarded in-boxes. A method called "harvesting" involves scraping email addresses posted in newsgroups or message boards, from which the spammer compiles a bulk-mailing list. "Nefarious people have created robots to go and harvest your email address from discussion groups and then spam you," said Paul MacIntosh, chief technology officer of New Jersey-based Emailias. "Normally, an address will get tainted, and there's no way to take back that address or stop the spam other than changing that address." Spammers may also use what's known as a "dictionary attack" in which they guess every possible user name in a domain. On the opposite extreme, spam filters are frequently accused of being overly zealous in weeding out email, capturing good messages along with the bad. Filters have been known to redirect email from a company's help desk from the in-box into a "killed" email box, for example. Thor Ivar Ekle, creator of SpamKiller, admitted that his system is designed to catch 97 percent of mass emails, including help-desk mail. Some consumers say that this is reason enough to declare spam filters a failure. "I have plenty of client filters, and I still see lots of spam slip right through...and lots get trashed. It's a losing battle from the consumer side. It's in the hands of the ISPs," said one woman who is a self-professed spam fighter. A higher-level solution
Three years ago, most ISPs saw spam filters as dangerous or censorious because they could block valid email. But in the last two years a dramatic rise in spam and complaints from customers has prompted a shift. Now, behind the scenes, many Net access providers and anti-spam agents are labouring to block spam from moving through Internet pipelines. The all-hours battle is costing ISPs an enormous amount of time and resources. Last year, the European Union estimated the global cost of spam at $8bn annually. "The ISP industry attitude changed from 'We won't filter spam' to 'Which filters shall we use?'" Spamhaus' Linford said. America Online calls junk email "public enemy No. 1" on behalf of its 34 million subscribers. Despite its in-house spam team working to block known bulk mailers and the plethora of filtering options it gives consumers to manage email, AOL spokesman Nicholas Graham said commercial email still manages to creep into mailboxes. ISPs such as EarthLink, MSN, AT&T WorldNet and Verizon Communications have enlisted spam-filtering software from San Francisco-based BrightMail to help shield consumers from bulk mail. About a third of ISPs also use block lists based on the worldwide DNS (domain name system) to refuse spam at their mail servers before it gets into subscriber mailboxes. Well-known blocklists from groups such as the Mail Abuse Prevention System and Spamhaus work to keep track of IP addresses used to send spam, in an effort to block them altogether. BrightMail's service, which operates a spam-detection center called BLOC, works by updating "mail rules," or filtering guidelines for the newest spam senders, every five minutes to seven minutes and sending them to customers. Such systems are focused on trying to pinpoint patterns in incoming mail and filter based on repetitions and keywords. But filtering systems that let consumers block email based on the wording contained within the message often fail because spammers are always tweaking language. For example, a consumer may set up a filter on "win a free car." But after using that terminology, the spammer might tweak the language to say "won a free car." Some state laws, including those in California and Washington, give consumers some legal recourse against junk mailers, but many anti-spam advocates say they don't root out the problem. Because the laws require consumers to "opt out" of receiving junk mail, advocates say the action costs people more time than they have. In some states, marketers are required by law to add the prefix "ADV:" to commercial e-mail. But spammers are learning to beat the system. They get around filters by using variations such as "[Ad V]" or "." Marketers use such tactics because email is quickly becoming the lifeblood of sales. Jerry Cerasale, senior vice president of government affairs at the Direct Marketing Association, said his organisation is trying to define spam and set guidelines for its 5,000 members to avoid bringing government regulations into the fold. "It wouldn't be spam if the sender has had a prior business relationship with the consumer or he has joined a list" to receive sales pitches, Cerasale said. Still, anti-spam advocates say such policies won't solve the problem anytime soon. "The filter war is an arms race which neither spammers nor consumers can win and which can only be stopped by outlawing spam," Spamhaus' Linford said.
For everything Internet-related, from the latest legal and policy-related news, to domain name updates, see ZDNet UK's Internet News Section.
Have your say instantly, and see what others have said. Go to the Telecoms forum.
Let the editors know what you think in the Mailroom.
