Getting Started with Respite 2

Introduction
What is a "trap" or "spam trap"?
Expected Results
Is Respite for me?
I think I'm a good candidate. What settings should i use?
How do i improve the filtering of Respite?

Introduction

This document is designed to give you a very brief overview on how to use Respite, best practices for maintaining your spam trap, and some basic guidelines on tweaking your individual setup. For more detailed information on specific Respite options, please see Respite Users Guide.

What is a "trap" or "spam trap"?

Throughout the various Respite documents you will see the documentation referencing "traps" or "spam traps". These terms are used interchangeably and mean the same thing;Your Respite account. Since the pending folder is where all the messages that require decisions are held, this is where most of your activity in Respite will be. For example, the phrase "Check your spam trap" is referring to checking your pending folder in your Respite account for trapped messages.

Trap names within Respite are the sames name as the address they are associated with. For example, "foo@rpi.edu" would be the trap name for the email address "foo@rpi.edu". More advanced users may have access to several traps (if you are an owner or moderator on a Respite-enabled listproc list, for example), so this method of trap naming is most convenient for ease of use.

Expected Results

Respite, like all anti-spam software, can only guess as to what you would consider spam and what you would consider non-spam (also called 'ham'). A common analogy to the spam-filtering problem is that expecting the mail server to block all the spam and none of the non-spam is like asking your postal carrier to deliver only the mail you want. As a result, there is no Respite configuration that will get rid of all of your spam and allow through all of your non-spam. You can, however, greatly reduce the amount of spam you get to just a few per day or less

Is Respite for me?

Everyone has a different interpretation of what a "spam problem" is. Some people receive one or two spam messages and consider it to be a problem, while others are willing to wade through hundreds of messages before considering it a problem. Depending on your situation, your mileage with Respite will vary, so we have compiled a checklist for the users who would benefit most from using Respite. You would be a good candidate for Respite if:

You receive more than just a few messages per day (the more, the better) AND
A large percentage of the email you receive is spam (the more, the better)

You will not get good results from Respite if you only receive a few messages per day, or if you receive just a few spam messages per day. Rather, you will be at higher risk for "false positives" (trapping messages that are really non-spam) than your spam-swamped counterparts. If you don't fall into the above categories, your best option is still the delete key.

I think I'm a good candidate. What settings should i use?

When you first sign into Respite, you will be presented with a Disclaimer with some basic notes about spam scanning. After reading the disclaimer, click the "CONTINUE" link, and proceed to the "Defaults Setup" page where you will be presented with a set of options for the initial configuration of your spam trap. Read through the entire page to understand what options are being set. The page is set with the recommended settings for new users, so it is recommended you select the defaults given.

If you have previously used Respite, you will not get the default setup pages. If this is the case, or you want to reset your settings with different defaults, you can get to the Setup page by choosing "Options" from the left menu, then selecting "Restore Defaults".

Get a feel for your settings for at least a week or two before changing them. This will give you a feel for how Respite works, and whether you really need to tweak anything, and if so in which direction. Often, the default settings works well enough to suit the user. Remember - no spam filtering is perfect, and a few missed spam messages are normal.

Finally, make sure you check your trap daily when you first start using Respite, to make sure your settings are adequate and you are not trapping a significant number of false positives. After you get comfortable you can check less frequently, however it is recommended that you check at least once every week.

How do i improve the filtering of Respite?

If, after several weeks, you decide you need to fine-tune Respite to your needs, keep several things in mind.

There is a point at which continued tuning of your Respite settings will not yield any significant results. If you have an acceptable number of false-positives, and only a few spam messages are getting through, further tweaking may actually hurt your spam catching performance.
Spammers are constantly changing their techniques to get around spam filters. A tweak to get rid of that last 1% of spam getting through may work today, but may not work tomorrow.
The more special case rules you use, the higher the chance of catching a false-positive.
NEVER use simple word-based matches to automatically reject spam. There are 600,426,974,379,824,381,952 ways to spell "viagra", and they will use all of them to try and get their message to you. Its best to leave the munged word decryption to the spam-scanner. (See http://cockeyed.com/lessons/viagra/viagra.html if you're interested in this statistic)

The easiest way to adjust your spam scanning is to adjust the "Spam Threshold" line, located under the "Options" menu, in "Stream Options". When adjusting this score, keep in mind that a lower score will catch more spam and potentially more false-positives, while a higher score will catch less spam, however with fewer false-positives. Keep in mind what you are setting here - this is a score for which all scores above it will be trapped. You can adjust the score in increments of .1.

More information on tuning your spam trap can be found in Tips, Tricks, and Hints for using RESPITE and the Respite Users Guide.

Last modified: February 3, 2007