A couple of commenters expressed skepticism about Gevlon's "4 fun ppl" theory. I thought I'd elaborate on what I find most compelling about it, compared to Penny Arcade's GIFT theory.
The first different part is "normal people" versus "basement dwellers". In the GIFT formula, everyone is a potential bad guy. This means that there is no hope of removing the bad element, because you'd have to remove everyone. But if Gevlon is correct, it is a specific subgroup of people who are cancerous. That means that you can target that subgroup specifically. As well, instead of binding the entire player base with rules, you can set specific privileges for specific groups.
For example, let's take vote-kicking. Under GIFT, we have to hedge vote-kicks with lots of defensive rules, because any normal person might abuse it. But under 4FP, it's only a tiny subgroup of people who abuse vote-kicking. So a better solution might be to have a broadly available vote-kick, but certain people are simply not allowed to vote-kick at all.
The second different part is "anonymity" versus "lack of clear rules/authority". Under GIFT, to clean up the internet, we have to remove anonymity and link virtual identity with real world identity. But a lot of people like anonymity and even feel safer with it. It is very unlikely that we will get rid of anonymity on the Internet anytime soon.
Under 4FP, anonymity isn't an issue. 4FP is perfectly fine with pseudo-anonymity. As well, removing anonymity is no guarantee that people will behave. I think Gevlon massed enough evidence that some people behave badly even when they are not anonymous.
However, I do think there is a partial link between anonymity and lack of rules. Very often, anonymity signals that a lack of clear authority exists.
I admit that I like Gevlon's 4FP theory because I believe in Broken Windows theory and the idea that people respond to their environment and push the edges of that environment. By setting the bounds of acceptable behavior closer than the absolute maximum required by law, I think it's more likely that the resulting community will be acceptable.