Spam isn't exclusive to email, it shows up in multiple areas. While this means it annoys us in many areas, it also means we can analyse multiple situations spam is used and find out what is effective and what is not.
Chatrooms are losing the fight on spam. Robots login to a chatroom, post their advertising method and log out before the chatroom moderators even have time to register them, let alone spam them. Why is it so easy for them to do this? Simply put the spammer can login from nearly anywhere(using anonymous access interfaces if necessary) and can change their nickname at will. If one name is blocked from too many chats they simply revert to another. Some success was encountered with IP based blocking, but the proliferation of Anonymizer sites and software has effectively rendered this defense ineffective. The same robots can also use chatrooms whisper functionality to send private messages to genuine users, unbeknownst to the moderator that anything untoward is happening.
Mailing Lists are descendents of newsgroups and also do fairly well against spam. This is because they are moderated and the moderators can easily block an infringing email address. The spammer can sign up again, but must often wait to be approved before they can begin posting. Spammers like instant gratification, and approval processes often deter them. Subscription mailing lists are even more effective, as spammers dislike having to actually pay to reach an audience.
Online messengers are another medium that was targeted by spammers. They provide two methods of attack. Firstly there is address harvesting. The robot messenger requests authorisation from multiple addresses. Many come back as address does not exist, and are subsequently wiped off the spammers list. All that don't are added to the spam list. Furthermore anyone actually replying to the message - whether authorizing or denying the spammer, is confirming the address is active. If they are authorised, the spammer is free to bombard the user with messages. Due to the nature of the messengers these are often sent in a pre-ordered sequence leading up to a sales pitch. The messenger programs are reasonably proactive in eliminating these robots. The real advantage they have is that the spammer can be blocked at the address level. Certainly they can create a new address, but this puts them back to square one.
So what we have learnt from this quick study is not all that surprising. Mediums where anonymity is offered(such as chatrooms) are easy targets for spammers and are being swamped by the weight of spam messages. Methods which require some form of accountability in the users, even if it is only an email address which can be re-signed for, do much better. Also, denying instant gratification seems quite effective - that is - requiring some waiting period between sign up and being able to use a service.
Applying this back to email we can see some obvious parellels. The messenger method of requiring authentication before whitelisting an address is ported to the challenge-response method of anti-spam protection, and the open anonymous access problem that chatrooms face is mirrored by the spammers ability to anonymize their sender information in e-mail. The long term answer to spam then seems to be to make it difficult(or time consuming, or costly) to attain an email address, and impossible to send mail without one.
