Spam Filtering

Messages I send to people at Cornell are getting marked as spam. How can I stop this from happening?

PureMessage checks for a few hundred spam "clues," to see if your newsletter or other message is being labeled as spam.

If an e-mail newsletter or other mailing needed by several people is being mistakenly labeled as spam, please write to the e-mail system administrators,, to request that the filters on the central mail servers be modified to accommodate your needs. (This is called "whitelisting"—the opposite of "blacklisting.")

Why can't CIT just delete spam, instead of making everyone set up filters or delete spam messages one by one?

The risk of deleting legitimate messages is too high, especially considering that spammers try to make their messages look legitimate. E-mail newsletters that people choose to receive are particularly likely targets for misidentification as spam.

Also, the choice of how to respond to spam rightfully belongs to each individual. A message that is of interest to some people may look like junk mail to others.

Will all Cornell e-mail be screened by PureMessage?

PureMessage will only be implemented on CIT's Exchange mail servers. If you read your e-mail via the CIT Exchange servers you will benefit from this tool.

Does this new filtering policy mean CIT is reading my e-mail?

No. The filtering tool, PureMessage, enables CIT to continue its policy of not routinely monitoring an individual's communications. The tool can also be tailored to meet the community's needs.

Cornell's policy on Responsible Use of Electronic Communications supports spam filtering or blocking as an appropriate restriction on the university's network, in accordance with university policy prohibitions against harassment. However, given Cornell's commitments to freedom with responsibility and to free inquiry, the decision to begin spam and virus filtering was made with great care.

What else can I do to deal with spam and viruses?

PureMessage automatically deletes virus-infected messages, and messages that have a greater than 80% probability of being spam, based on the messages' spam-like characteristics. Read more about Cornell's spam and virus filtering.

In addition, Outlook and Entourage offer junk-mail filters that can help you weed out anything that makes it past those initial defenses. See how to set up junk mail filters: