The blog is older than you know. I prefer counting from the emergence of one integral anomaly to the emergence of the next, in which case this is the forth version.