Spam - What is the problem

What is SPAM?

Like everybody else discussing the problem, it is essential that we define what we mean by spam, and why it is a problem. There are many sites on the Internet discussing this issue, and they have several good definitions. For our purposes, I will give you my opinion as to what is, and what is not, spam.

Spam is unsolicited email, generally, but not always, of a commercial nature, which is sent to large numbers of recipients.

  • A single mail from someone you don't know is not spam.
  • Mail from a commercial concern with whom you have a relationship (even if the mail itself was unsolicited) is not spam.
  • Anonymous email is not necessarily spam.
  • Mail that comes from a different address than the reply-to is not necessarily spam.

    It becomes spam when it is sent in bulk, and when it comes to addresses that do not exist or are not advertised. It is probably spam if there is no clear reason why the sender thought the recipient might be interested in the message.

    Spam senders demonstrate that they are not providing legitimate advertising by the fact that they go to such lengths to hide the sending address. A legitimate advertiser wants you to know who they are. They are proudly advertising their own products, and want you to be able to respond. A responsible advertiser even wants to hear from you if the ad is unwanted. Responsible sellers realize that harrassing people will not get them to buy your product, it will instead cause them to tell their friends to avoid your product.

    Why is it a problem?

    There are a few main reasons that spam is a problem.

    First, there are legitimate web sites which are being swamped by the sheer volume of the email. Charity sites and non-profit organizations are sites that we want on the web, but if they start receiving tons of spam, they have a major problem. Most of these sites are not in a financial position to hire somebody just to sort through the email. These sites wind up making choices; they either spend all their time sifting through the junk mail looking for the legitimate enquiries, or they apply broad filters and hope that not too many of the legitimate enquiries are getting thrown away, or they just throw up their hands and stop accepting email. Some sites will simply shut down because they do not have the resources to handle the incoming mail.

    Second, even if there were not the problem of somebody having to go through all that mail, there is still the problem that the mail is being sent over the network. The bandwidth available on the network is neither infinite, nor free. Spammers take advantage of this, and happily use the bandwidth that other people are paying for. In essence, with the constant increase in the volume of spam, the spammers are launching a denial of service attack on the entire Internet. For smaller sites, even filtering at the mail server may not suffice if enough mail is coming at them.

    Third, spam is a receiver pay technology. In most other media, these have been banned. With spam, the recipient must pay for their online time, which is not always free. The recipient must also spend their time sorting through the deluge of unwanted messages. When the spammers say that it only takes a second to delete their unwanted message, we should be asking why they don't do it themselves then. One second per message is about two weeks, 24 hours a day, of deleting for a single spam sent to a typical mailing list (one million listed addresses).

    Why is it hard to solve

    The Internet is a vast network of interconnected computers. They run numerous different operating systems. They run numerous different mail servers. They are not all always maintained with the latest software. There are many subtle interactions in the way the Internet works now that allow desirable features such as anonyomous mail, redundancy, multi-ISP hosting, and many others.

    This means that any solution to the problem should be an iterative one. The receiver should be able to move towards a solution without destroying any of the features we've come to value.

    While I intend to discuss even solutions which require massive changes to the way the Internet works, I prefer those solutions which introduce small iterative changes, and acheive the results without risking the benefits we have gained so far.


    This page maintained by Rob (at ewan dot com, of course).