Terms of Service | FAQ | Settings

SPAM REDUCTION FAQ

Table of Contents:

What is Spam?
What does your service do with the SPAM
Does this service quarantine the SPAM?
How does the system decide what is SPAM and what is not?
How accurate is your system?
How do I set up my email program to filter out the marked messages?
Can I create my own rules?
Can I just have the SPAM deleted before it goes to my inbox?
A SPAM message got through.  What can I do to stop this message from coming through?
An email from one of my mailing lists has been marked as SPAM.  How can I stop this from being tagged?
How do I change the default settings for my whole domain?
How are default settings applied?
How do I adjust the settings for a specific user/email address?
One user in my domain doesn't want their email filtered.  Can I exempt one address?
Does the system send a notice to the SPAM sender?
Is the ISP of the spammer notified that they have a spammer using their network?
Does your system do content (adult material) filtering?
 

The Answers

What is Spam?

Spam is a nickname given to bulk email trying to sell you something. Other terms include Unsolicited Commercial Email (UCE) or Unsolicited Bulk Email (UBE). The nickname SPAM originates from the famous Monty Python SPAM skit. SPAM is a registered trademark of Hormel Foods Corporation.

[top]

What does your service do with the SPAM?

Once the system determines that the message is most likely SPAM, it will tag the email using either X-headers or by adding *****SPAM***** to the subject line letting you know that it is SPAM.  This allows email client programs to apply rules to the incoming message to delete the message or move it to a junk mail folder for later review.  Messages with exceedingly high scores may be discarded to reduce the burden on the Internet.

[top]

Does this service quarantine the SPAM?

No.  It is a waste of time to have to open a web browser, login, and review your SPAM online when a simple rule in your email program can keep the SPAM out of your way and make it much easier to review by simply opening a folder.  Spam reduction services should reduce the amount of wasted time, not increase it!

[top]

How does the system decide what is SPAM and what is not?

Incoming messages run a gamut of over 250 tests to determine what is SPAM and what is not.  Each rule has a score, either a positive value for SPAM-ish tests and negative value for non-SPAM-ish characteristics.  The higher the score, the more likely the message is SPAM.  The system can also "learn" what may or may not be SPAM by watching the emails that come through and training on emails with very high or very low scores.  A special "Bayesian Classifier" looks at key areas of the messages and assigns a probability of being SPAM which then adds/subtracts from the score.  If the score is above the score threshold, the email is tagged as SPAM.

[top]

How accurate is your system?

The service starts out near 90% effective and gets better as it learns from your emails! Current levels are over 97%.

[top]

How do I set up my email program to filter out the marked messages?

Instructions for Outlook 2002 are available here

Can I create my own rules?

At this time there are no user-definable rules.

[top]

Can I just have the SPAM deleted before it goes to my inbox?

No.  False positives are a possibility and it would be unwise just to turf everything.  By setting rules in your client, you have much better control over any mistakes the system may make.  That being said, there may be an option to discard high-scoring emails coming in the future.

[top]

A SPAM message got through.  What can I do to stop this message from coming through?

There are 2 actions to take:

  1. You can add the email address of spammers to the black list in the control panel under "Reject Mail From" in the "add rules" section. 
  2. Train your Bayesian database to recognize the message as SPAM by using the footer link or by entering the special header X-Filter-ID: which will have a value like: 0xB04B10E64B2B11DBB0A08EAF28E2B2ED.  This number can be copied and pasted into the control panel and trained.

[top]

An email from one of my mailing lists has been marked as SPAM.  How can I stop this from being tagged?

You can add the email address of your mailing lists and other addresses that are not SPAM in the control panel under "Accept Mail From".  Addresses in this list will pass through without being marked as SPAM. As above you can train the Bayesian filter to recognize the emails as non-SPAM

[top]

How do I change the default settings for my whole domain?

As the domain administrator, you will see a user named "@somedomain.com" where somedomain.com is the name of your domain.  When you edit the settings for that account you are creating the default settings for you domain.  Please note that individual settings such as score will override the domain settings. see below.

[top]

How are default settings applied?

The settings made here will override the global system defaults and the user settings will override the domain defaults.  The allow and deny lists will actually add together as opposed to override  creating a more comprehensive list.

[top]

How do I adjust the settings for a specific user/email address?

As the domain administrator you can create "accounts" for each email address.  Once created, you can set any of the settings independently of the domain or other users.

[top]

One user in my domain doesn't want their email filtered.  Can I exempt one address?

As the domain administrator you will have an extra field in the "@somedomain.com" account called "Accept Mail Sent To:".  Addresses placed in this list will not have their email tagged as SPAM even if it scores above the threshold.

[top]

Does the system send a notice to the SPAM sender?

No.  Actually finding the true originator of a SPAM email is almost impossible as spammers use fake or random user accounts.  Sending messages back to a bogus or innocent email user contributes to wasted bandwidth on the Internet.

[top]

Is the ISP of the spammer notified that they have a spammer using their network?

No.  It is difficult to ascertain the true origin of the SPAM in many cases and it clogs the Internet's arteries with yet more mail! That said, if you can figure out the source, go get 'em!

[top]

Does your system do content (adult material) filtering?

Many adult themes are tagged, the systems is not designed to specifically deal with that content.

[top]