When Microsoft first announced the Microsoft Exchange Intelligent Message Filter (IMF), the company said it would make the Exchange Server 2003 add-on available only to customers that were enrolled in the Software Assurance (SA) program, a licensing program that lets customers spread licensing costs across a 3-year period during which Microsoft provides free upgrades for the products covered by the license. (See http://www.microsoft.com/licensing/programs/sa/default.mspx for information about SA.) Many people speculated that this decision was a result of Microsoft's attempt to provide SA customers with more value as well as a not-so-subtle effort to encourage non-SA customers to join the program (and upgrade to Exchange 2003 in the process). Upon the add-on's release at Microsoft TechEd 2004, however, the company announced that IMF would be available as a free download to all Exchange 2003 customers.
The antispam industry provides value through third-party products and specialized companies that perform the hard work of tracking the most recent spam techniques and sources, as well as figuring out the most efficient way to suppress spam while generating the fewest possible false positives and without slowing down overall mail transmission and delivery. Now that Microsoft is providing a server-side antispam service, will it be the end of third-party Exchange antispam products? Hardly. Several factors will keep third-party products going strong.
First, there are many email systems other than Exchange 2003—including its biggest competitor, Exchange Server 5.5. Second, many enterprise administrators don't want to connect Microsoft servers directly to the Internet. (Microsoft plans to ship a specialized version of Exchange 2003, running the forthcoming Microsoft Exchange Edge Services, sometime in 2005, as the Web-exclusive article "Get Ready for Exchange Edge Services," March 2004, http://www.winnetmag.com, InstantDoc ID 42013, explains. Still, it will take time before Windows becomes the platform of choice for edge connectivity and before IMF proves its competence in production.) Third, it's highly unlikely that any one product will ever be able to block 100 percent of spam at one point in a messaging infrastructure. The best way to eliminate the most spam from user Inboxes is by deploying blocks at the network edge, mail server, and client. This practice leaves plenty of ground for companies such as Brightmail, GFI Software, Symantec, and Sybari Software to fight over. (If you want more information about third-party antispam alternatives, see "Beef Up Exchange with Add-On Software," May 15, 2004, InstantDoc ID 42405 and Buyer's Guide, "Enterprise Spam Filters," April 2003, InstantDoc ID 38277.)
Furthermore, antispam vendors will surely react to the new Exchange antispam standard that IMF introduces. To make themselves worthwhile, third-party products will need to offer extended features such as easier management (especially in distributed and heterogeneous infrastructures), better reporting (e.g., daily statistics for throughput and spam detection), better management of captured messages, regular updates to counter new spam techniques, integration with antivirus protection, and fewer false positives. I'm interested to see how this piece of the Exchange add-on market will evolve now that Microsoft has clearly decided that antispam protection is important enough to be a builtin part of every Exchange deployment.