This issue is to track work on using time series similarity to locate outliers, in contrast to some of the other probabilistic approaches.