eWEEK content and product recommendations are editorially independent. We may make money when you click on links to our partners. Learn More.
Ive been a confirmed pessimist on the future of spam for some time now, so it was with a Spocklike upturned eyebrow that I read AOLs announcement last week that it has turned the corner against spam delivered to its users.
Its quite a bold announcement. They claim that the amount of spam complaints from their users is way down, as measured by the number of times users hit the “Report Spam” button in their AOL client. Specifically, the number of such reports went from 11 million daily in November 2003 to 2.2 million daily in November 2004.
At the same time they claim that the amount of mail being diverted into users spam folders—in other words, the spam caught by AOLs filters—is down from 100 million messages per day to 40 million.
Its easy to get reflexively cynical about such numbers. There have been many reports of declining membership at AOL, so maybe that accounts for some of the decline in users spam folders and a declining number of complaints. My sister is an AOL user and was so bothered by false positives in AOLs spam filtering that she turned it off for most of 2004. Perhaps such dissatisfaction with their filtering is responsible for some of the decrease. The language of the press release is suspicious in this regard, in that it talks about “declines in the amount of mail being diverted to AOL members Spam Folder” as opposed to the amount of spam blocked. (My sister recently re-enabled the spam filtering and decided to be assiduous about whitelisting people whose mail she needed to receive.)
On the other hand, AOL blocks a certain amount of mail at the boundary whether the user has spam filtering on or not—partly worm traffic, probes, mail fragments and things like that, but a lot of plain spam as well. An increase in the effectiveness of such techniques would increase the other effectiveness measures they report.
Its reasonable to believe that the quality of this filtering and of other filtering has gotten better over the last year. But they actually claim just the opposite: “…the average daily amount of Internet spam e-mails that are blocked at the gateway by AOL antispam filters has declined sharply—a 50 percent drop—from a peak of about 2.4 billion in 2003, to an average daily volume of just 1.2 billion blocked spam e-mails in late 2004.”
The point of the announcement is actually not so much that AOLs anti-spam technology is getting better, but that spammers know not to mess with AOL and are targeting its users less. The marketing appeal of this is plain to see, especially at a time when ads from NetZero and others treat Internet service, including add-ons such as spam protection, as a commodity to be sought on price alone. AOL would have us believe that youll get less spam on their service irrespective of the technology.
Next page: So why is everyone else getting more spam?
So why is everyone
else getting more spam?”>
As they point out, AOL is at the forefront of civil actions against spammers, along with Microsoft and some other large ISPs. I always figured that to the extent these actions are effective we all benefit from them, but AOL would have us believe that the spammers arent so much chased out of business as much as off of AOL. I suppose its easy to scrub a list of e-mails of those in the aol.com domain, but I still have trouble believing that it really makes a difference.
The main reason Im skeptical is that nobody else is claiming a similar trend, and lots of other sources are claiming just the opposite. After the AOL announcement, I sent out questions to many ISPs and other vendors and only got a few responses (it was a holiday weekend). The only major AOL competitor to respond was Microsoft, who said that they couldnt make an apples-to-apples comparison to AOLs claims because they track numbers differently. This does make sense, since AOL gave the funny numbers about messages in users spam folders and other such vagueness.
Microsoft did say that “…over the last year MSN customers are receiving more than 60 percent less spam, mostly due to Microsofts “Smartscreen” filtering technology, which stops more then 3.2 billion spam messages a day worldwide.” Now this is more of a plain and credible number, although it says nothing about the amount of spam coming into the MSN and Hotmail networks. Is this number better or worse than the AOL numbers? Its impossible to say; youd need to know more about how much actual spam was coming in before the alleged declines.
Given that all other sources I follow claim that the amount of spam out there is going up, its hard to take AOLs numbers at face value. For instance, according to MessageLabs, the average percentage of e-mail identified as spam in 2004 was 73 percent; in 2003, the average was 40 percent. I see similar numbers from other sources. Are spammers really avoiding AOL to such an extent that their trend is a polar opposite of the industry?
Apart from AOLs claim that theyve scared the spammers off, theres just one other explanation, for which Ill thank Stephen Canale of OnlyMyEmail. They had speculated that the protections at consumer ISPs were being hardened against spam more effectively and more quickly than business domains, and that user addresses at consumer ISPs are more volatile than at business domains. Therefore, eventually spammers might migrate to focusing on business addresses and the quality of their lists would improve. Its a theory and you can make a case for it, but I still think AOLs numbers are a bit too happy.
Id love to be wrong, because if I am it means that aggressive filtering and aggressive legal action can be effective against spammers. We should know more by this time in 2006. Ill believe it more when I start to see other vendors making similar claims.