On complexity, rare events and the systemic underestimation of risk.
skeltoac says it's fine to use with other blog platforms, and for all the talk about "hard-core hackers" it's not at all difficult. The reliance on hated XML-RPC keeps me away, but eh.
I'd thought it was inscrutable mathematical voodoo, so never bothered reading further, but it amounts to determining the value of a particular piece of evidence by taking into account both false positives and false negatives. Simple but powerful.
Statistical correlations between random xor questions.
[T]he results showed said it was possible to discover clues about what people were like simply from the music they liked.
Almost 38 percent of hip hop devotees and 29 percent of dance music fans were more likely to have had more than one sexual partner in the last five years compared to just 1.5 percent of country music fans.
Not such a surprise, given that most people I've seen "dancing" to hip-hop are already having sex by some definitions, and a night out for a dance music fan involves a seething mass of drunk/drugged friends jumping up and down together in a darkened room. An opera fan, by contrast, will be sitting next to Great Aunt Millicent and watching a fat man in a penguin suit; that this is less likely to lead to a passionate encounter is actually something of a relief.