On Friday, Sept 10, 2010, Microsoft finally released an update to its Microsoft Outlook Junk Mail Reporting tool.  This is an update for Outlook 2007 SP2 or later, or Outlook 2010.  This is not an upgrade for the existing version, it is an entirely new version written from the ground up.  This plugin is designed to be used by customers of Microsoft Forefront Online.  So, if your mx record points to mail.messaging.microosft.com or mail.global.frontbridge.com, or you sign up through the BPOS service (various MX records but it points through us), this plugin is for you.

The plugin is a one-click stop for spam in your inbox – if you have spam, highlight the message and then click “Report Spam”.  The message disappears from your inbox and is deposited into your junk mail folder.  If you have multiple messages, highlight them all and click “Report Spam”.  All of them will disappear from your inbox.  What happens is when you click the message, it is packaged up by attaching the message and inserting the message headers inline the message.  It then submits it to our spam abuse reporting alias.  Our spam analysts then take a look at the message and take action, or some of our auto-processing jobs handle it and take care of it.

Anyone can download and install this plugin.  However, if you are not a customer of ours (ie, your mail does not flow through our service), while your spam is still interesting to us, we probably won’t take action upon it.  That is because we focus primarily on spam flowing to our customers.  While there may be some spam that overlaps, there is some that doesn’t.  We can use it for trending, analysis and comparing to existing spam to our own users, but even if we did take action it wouldn’t affect you.  Of course, one use is that if we get one submission from our users and hundreds from others, this is indicative of a trend.

The advantage of our customer base using this plugin is that, as I say above, the mail is packed up including the headers.  The headers of the message are important because it tells us what our own filters did with the message and preserves other useful data, like patterns that spam bots might insert.  Having everything the same allows us to extract tokens, features, do auto-classification and get through submissions much quicker.  With everything in a predictable format, it makes it possible to update our filters faster as well as do customer requests that have been in the pipeline for a long time.  There’s no guesswork in the extraction of data.

You can get the update here from the Microsoft Download Center.

Finally, congratulations to the team here in Forefront Online that was working on this (excludes me).  Well done!