We also have the ability to immediately discard any email identified by PureMessage as spam without it reaching your inbox. This saves you from processing the incoming spam yourself (i.e. with procmail) but could result in an incorrectly flagged email disappearing without anyone ever hearing about it.
CSLab feels that the automatic rejection of email is relatively safe and quite effective (and we use it on many of our own personal mailboxes), but that the automatic discarding of email is more dangerous, since if a message is mis-classified no one will ever hear about it. However, if you are going to discard such messages anyway, you might as well have the CSLab mail system do it for you.
In addition, the mail gateway can reject email to you that is from machines outside the department or from machines outside the university (machines in DGP, CDF, and KMDI are considered inside the department). Because we reject it, a person sending you email from an outside machine will find out about it. It is important to note the difference between machines and people; for example, someone in the university might send you email from a Google Mail account, and that would be considered to be outside the university.
If you want to enable any of these options, please contact your PoC. These options can be applied to your own mailbox and/or to CSLab mailing lists that you run (and a mailing list can use different options than your own mailbox).
All incoming email is now processed by Sophos PureMessage, a server-side spam solution that tags emails as "spam" and "virus", which means that you don't have to train a spam engine yourself. For most people, the first link below on PureMessage basics should be enough to get you up and going.
Some mail clients contain built-in spam classifiers. Below are instructions on how to configure some of them.