the Pros and
Cons of Spam Filters
By
Carrie Harrison
To outwit the tricks used by spammers, software developers are coming
up with equally resourceful spam filters. If you’re an email marketer
or communicator wanting to know what you’re up against, here
are a few tips about the latest batch of spam filters in use today.
A password
filter is fairly airtight. It
will only accept email that contains a pre-set password
in the
subject
line. Unless you know it, you can’t get through.
A password filter is an effective method for blocking
spam, but it can
block desirable
email too.
A challenge/response
filter,
such as MailFrontier Matador or SpamArrest, sends an automated
message that asks you to provide return confirmation of your
email address. The system is verifying that you are an individual
sender,
not a machine generating spam. But many “requested” newsletters
may also be blocked by these filters, if a company or email
solutions provider is not equipped to respond manually to verification
challenges.
Whitelist filters only
allow email through from approved senders. The sender’s address
must be in the recipient’s mailbox or the email will bounce back.
There is no early warning system that seeks to verify the sender’s
legitimacy, so be aware that you may have no indication your email didn’t
get to its destination.
Spam Assassin is
one of the more common rules-based filters available. It blocks email
based on how it conforms to pre-determined rules, and catches spam by
finding tell-tale signs like odd punctuation.
The good news is that many organizations are using this type of filter
and, if you are using all the rules of good permission-based email marketing,
you should have no fear of being filtered out. Unfortunately, the rules
keep changing and companies are able set filters to be so sensitive that
even innocuous daily correspondence can get tossed as spam.
A community-based
filter can
be an effective tool. Cloudmark Spamnet blocks email
based on a group’s determination as to what is
junk. Since the group decides what constitutes spam,
personal and other
material does
get through. As a con, this type of filter may not catch
the newest messages outside the group criteria and can
admit too
much spam.
An adaptive
Inboxer is a filter that learns by analyzing
examples of what is discarded as spam and what isn’t.
These systems require training, especially at the start.
The
system learns the user’s preferences, keeping out
spam and accepting messages shown to be desirable to the
user. Inboxer
from Audiotrieve has been
given high marks for effective filtering using Bayesian statistical techniques.
To work within the confines of each of these filters, ground
rules for email marketers include: researching and installing
spam filters for in-house pre-testing,
proactively ensuring good relationships with ISPs, carefully monitoring
campaign results, and meticulously ensuring that lists used are
based on recipients
who have opted-in and given you permission to send your messages.
It’s the only way to keep off the blacklists, stay on the
whitelists, and avoid being relegated to the junk bin.
| Types of Spam Filters |
Method |
Pros |
Cons |
| Password eg: Inboxer |
Blocks mail without password in the subject line |
Only those who know the code get their messages through |
Messages you actually want may be blocked |
| Blacklist eg: Geekmail.com |
Blocks mail from banned senders |
Knocks out known Spammers as opposed to legitimate
messages
|
Misses a lot of Spam, including ads you may not want |
| Whitelist eg: MailFrontier Matador |
Allows mail only from approved senders
|
Blocks mail from unknown sources
|
May hide messages you did not know you wanted |
| Challenge/Response eg: MailFrontier Matador |
Blocks unapproved sender's mail until response to
challenge arrives
|
Keeps out machine-generated mail |
May block mail you want, also may annoy legitimate
correspondents |
| Rules-Based eg: Spam Assassin |
Blocks mail based on how it fits with predetermined
rules |
Catches Spam by finding tell-tale signs such as odd
punctuation
|
Spammers can develop clever ways to fool the system |
| Community-Based eg: Cloudmark Spamnet |
Blocks mail based on community agreement that it
is junk |
Since the group decides what is Spam, personal material
gets through
|
May not catch newest messages, can admit too much
Spam |
| Adaptive eg: Inboxer |
Learns by analyzing examples of what you call Spam
and what you don't |
Learns your preferences, may allow for special purpose
folders |
Takes time to train, especially at the start |
Legislative
|
Laws to regulate unsolicited commercial email
|
Makes some Spam illegal, sends some people to jail |
Spammers already going offshore, some Spam may remain
legal |
© Copyright 2003 Galt Western Personnel Ltd. Unless otherwise specified,
you may reprint this article, quote from it, use it in research or projects,
duplicate it or distribute it. Credit of authorship and source MUST be
given to galtglobalreview.com. Ownership of Copyright remains with Galt
Western Personnel Ltd.
