How do we know what spam is? Somehow it is easy to say if something is spam and if something is not, but it is difficult to describe how you do it.
In a way, it is your individual choice, or so the marketing droids keep telling us, before sending us all the ads and all the spam to offer us some "choice". Nevertheless, a community will quickly recognize what is spam and what is not, without having to argue about it. Therefore, let us not waste time trying to define the obvious. The reaction is always the same: Aggravation and a waste of time. Which is why people try to fight spam.
See also: WhatIsaTroll
Ways to fight community spam:
- Fix it; e.g. PeerReview, ReversibleChange -- pick up their litter
- DissuadeReputation; e.g. ForestFire -- just ignore it, eventually delete it
- Earn Trust; e.g. TrustMetric, WebOfTrust -- disallow people from speaking up until they have somehow proven their merit
- Lose Trust; e.g. moderation, RealTimeBlackholeList, UseNet cancelbots -- Although afforded a PrincipleOfFirstTrust, when proven untrustworthy, crush them
Hardcore spammers often use automation: both to send large quantities of spam, and to bypass simplistic TuringTests that attempt to seperate computer-generated and human-generated content. Communities must be prepared to MeetForceWithForce.
The ways to fight spam are the follow the exact same structure that you use to fight any enemy that you cannot control. You can either be like a rock and try to build fences (HardSecurity), or you be like water and bend around the attacker (SoftSecurity).
When I was at MIT watching HowardRheingold talk about SmartMobs, I met one half of the gestalt UseMod:AdamBraite. While Howard was talking about the problem of spam, the half of Adam Braite suggested that I knew the answer--as in SoftSecurity. I've been wondering what he meant, but maybe it's basically what's written above. -- SunirShah
Fighting email spam
- Wiki:SpamAssassin
- Define 'spam' to a bayesian filter by feeding it lots of messages you consider spam, and lots of messages you consider 'ham'. Let the filter sort out the new incoming messages (e.g.: http://spambayes.sf.net)
- Illegalize (?) sending spam; see http://www.spamlaws.com/ (LegalSolution).
- Mail Abuse Prevention System RealTimeBlackholeList http://mail-abuse.org/rbl/, as well as the defunct Open Relay Behaviour-modification System [WebArchive:orbs.org]
- [EmailImmuneSystem] - a system where you setup honeypot addresses to trap spam and then use that catched spam to identify spam coming to your normal address.
Note that no CommunitySolutions are considered: "A TechnologySolution may be better than a CommunitySolution in dealing with strong disagreements or disruptive people.".''
That ignores the two realtime blackhole lists which were social for the most part.