As you probably read in Today at Berkeley Lab and noticed in your email inbox, the Lab has enlisted a new ally in the continuing battle against unwanted junk email, or spam. On Wednesday, Feb. 19, the Computing Infrastructure Technologies Group rolled out a service called Brightmail, which identifies and tags suspected spam, according to a continuously updated set of criteria. These messages are labeled with the words [SUSPECTED SPAM] in the subject line. Please do not forward or report tagged spam to spam@lbl.gov. You can cull out these messages for review and disposal by setting up your email system to either sort by subject or filter them into one folder. Here's how.

Now entering its second week, the service has prompted a number of employees to ask why the spam messages are only being tagged, rather than automatically deleted. The answer: During this initial phase, the Lab's central email managers wanted to err on the side of caution and not delete mail that wasn't truly spam. During the next month or so, the group will be evaluating the service, seeing how well it identifies -- and misses -- true spam, as well as mislabeling legitimate email as spam (known as false-positives).

Employees can help with this evaluation by alerting the Brightmail team of problems and inconsistencies. If you received a piece of spam without the [SUSPECTED SPAM] tag, forward the message to spam@lbl.gov. Remember, though, that the service will not stop email if you signed up to receive it, even inadvertently. If you want to stop such mail, you'll need to unsubscribe directly with the sender. However, the Labs' spam-fighters recommend that employees not reply nor try to unsubscribe from obvious spam as this can actually lead to even more unwanted email.

And just as the new system can't catch every piece of junk email, it can occasionally misidentify legitimate messages. If you get a legitimate message that has been tagged as spam, forward it to notspam@lbl.gov.

Read more about this and other features of the service.