Sample size and power for rare events
ArsTechnica posts a flowchart illustrating a thought process used by many modern feminists which shows they are not rational Bayesians (except under the most charitable possible interpretation). So I'll take the opportunity to link to Eliezer Yudkowsky's excellent post Conservation of Expected Evidence.
The consolidated tape will now display odd lots. About time.
A MongoDB proponent unintentionally explains it's problems.
Unit tests as penance. Upon reading it I realize I think similarly.