I've found when people discuss spam they really have no idea what they are talking about. There are as many different definitions of spam as there are people. In point of fact, this factor alone (not being able to define what spam is and what it consists of) makes it virtually impossible to control.

In order to control spam, a useful definition is necessary. Why? Simple. In order to control something, you must know what you are controlling. When you understand the basic facts, then you can take whatever course of action is necessary. Until you achieve that understanding, you will be shooting blindly at an undefined target. This makes it very difficult to actually do anything useful.

So on that note, what kind of definitions for spam work and don't work?

Commonly spam is defined as unsolicited email. Unfortunately, this definition by itself is NOT spam and means absolutely nothing. What's wrong with it? This definition does not help you solve the problem, and thus is incorrect. If this definition was true, then to prevent spam you would have to somehow contact a person to ask them if you could send them an email.

For example, I don't generally call someone on the phone and ask them for an email message. That would be silly. In fact, by definition most email is unsolicited; I don't, for example, expect my wife or a friend to ask me if it's okay to send me a message.

Sometimes spam is defined as emails that are from unknown sources. Hmm. This really doesn't work well either. I'll get emails from my website from people I don't know - these are not spam. Also, sometimes my friends will pass my email address to their friends, who send me email. These are also not spam, even though they were from an unknown source and were unsolicited.

How about just plain annoying emails? That seems to be the definition that most people have in mind when they mention spam. If the email is annoying in some manner, and especially if it was unsolicited, it is spam. This definition probably gets a little closer to the heart of the matter, but it really doesn't define spam well.

What about unsolicited bulk email? This definition gets a little bit closer but it still doesn't really define spam well. I mean I give my email address to my bank and I really didn't ask them to send me emails (although I didn't ask them not to as well). Yet I would not call this spam as I do business with the bank. Their emails might be annoying, but since I have a business relationship with the bank I expect them to communicate with me occasionally.

Okay, so what is spam?

I like to think of spam as "unethical mass email". By this I mean emails which violate the netiquette standards of the majority of users of the internet.

Note that by this definition, an individual email sent to a person is not spam. A commercial email, however, is another matter. Even a single commercial email might be unethical if it does not follow the rules below.

Ethical emails are targeted well towards their audience. Unethical emails are mass mailings sent out blindly to a large number of people.

These are emails that are sent to thousands, tens of thousands, even millions of people, hoping against hope that a few dozen will be stupid or greedy enough to respond. These emails are untargeted and will not pertain to the majority of the recipients. Since the majority of the people reading the message (usually upwards of 99%) will simply delete it immediately, this makes the mailing unethical.

Ethical email messages include valid email header information. This information properly identifies the sender of the message. In addition, all of the other header data in the message is correct.

Spam messages often have forged or invalid email headers. This means it is difficult (if not virtually impossible) to trace the source of the email based upon the header information within the email message. Since the sender of the message cannot be identified the message is unethical. In this case, even a single email message would count as spam.

Ethical mailings include a method for opting out which actually works.

If you run a newsletter or do any kind of mass mailing, you must include at least one method of removal in the email message itself. This removal method (and more than one is preferable) MUST WORK. Some things which I often see in opt-out schemes which ARE NOT VALID include the following:

• Any email message which states that the reader must go to a web site, log in and then modify his email preferences is UNETHICAL. This requires too much information from the user and forces him to do too much work.

• If the email message includes an unsubscribe link (or other means) which does not work, then it is UNETHICAL.

• Messages which validly allow for opt-out but then say "you will be removed in a week" or some other long period of time are UNETHICAL. These are computers people, and there is no reason to include these long delays. Remove the person immediately.

Ethical mass mailings are double-opt-in. This means after a person signs up for the mailing list, he receives a confirmation message. He must either reply to this message or click a link to activate the mailings to him. Any other form of opt-in is UNETHICAL as it allows people to be subscribed by others or by accident.

Ethical mailings do not include webbugs, set cookies or perform any kind of involuntary tracking.

Email messages are often opened up by the recipient before he knows anything about the message. This means if you are doing any kind of tracking, the person has no way to stop it, short of blocking the receipt of the message entirely. This lack of a choice on the part of the recipient makes this kind of tracking UNETHICAL. The only time this would be ethical is if it was clearly stated when the user signed up for the mailing. In that instance, this behavior is known and this makes it ethical. Note that while the web site privacy policy should state this fact, it must also be stated clearly on the page where the person actually signs up for the mailing.

Ethical mailers do not use email harvesters. Using special robots to gather email addresses from web sites is UNETHICAL. These email addresses are generally included on web sites to allow individuals to communicate with individuals. Rarely is the intention to join a mass mailing list distributed on CD. ALL USES OF EMAIL HARVESTING IS UNETHICAL.

Ethical mailers do not take advantage of open relays or use other "spammer tricks". If you are legitimate, then there is no need to attempt to hide your whereabouts or cover your tracks. Using a relay without permission or sending millions of emails through an unprotected formmail script is simply bad manners.

Get the idea? Spamming is NOT sending someone one or more email messages without their express permission. Spamming is simply ignoring the rights of others (your audience, system administrators and even the users of the internet as a whole). That's all it is.