This content was published by Andrew Tomazos and written by several hundred members of the former Internet Knowledge Base project.
Spam, Filters and Whitelists
Email is one of the oldest applications of the Internet.
It is practically free to send a message to anyone, anywhere in the world, instantly.
It is technically very easy to send many millions of emails at the same time.
The From address of an email is set on the honor system, which means that it is easy to make it look like it has come from anyone or no one.
Thus was born the worldwide epidemic of spam. Unwanted email.
Spam can be categorized according to
- whether or not it was a commercial offer
- whether the same message was sent to one person or many
- whether or not the author used their real From address to send it
It has been tackled by programmatically:
- analysing the message for typical textual patterns of unwanted mail. (bayesian filtering)
- comparing the from address against your address book (whitelist)
- encoding email in a special way to prove that the sender was real (cryptography and digital signing)
None of the current solutions are adequate because what constitutes unwanted mail cannot be specified very well to a computer.
In fact, ultimately you can only know that a mail message was unwanted by a paticular person by getting them to read it first, which defeats the purpose of a filter.
Making sure that you can identify the sender of an email is an important step though.
Back to Index