Banning something at the slightest reasons seems to be a popular idea, at least among the administrators. It is a rather strange that most of the time medium or the format through which something "seemingly" malicious has been put up is the one under attack, but seldom the actual content. For me, I very rarely do take any notice or associate a value to the things that are put up in community sites. One of simple tools of statistics is that you take a average over all observation (our brain does a more complex analysis than this). The point I am getting at is that if the so called hate communities are isolated outlier entries, their impact factor on people will be highly marginal. The community is just a representation of a group of people who are already vehemently opposed or irked by the particular entity. On the contrary if these are not isolated entries, it is best that the group/person under attack realise they are non-populist and do something to figure out reason for their lack of popularity (among the netizens), rather than blinding themselves.
Trolls, hate messages, spam and insensitivenesses are going to stay. You don't have to look further than rediff message boards to get a good concrete example. The earlier people learn to ignore them the better (which I do believe most people to a certain extent have done so).
