Previous work to cull data should have removed everything stored prior to December 2015, however an investigation of compaction (T126221) has made it obvious that some of this data still exists. Since a considerable amount of data was deleted, anything remaining is going to be difficult to reason about. It would be interesting to know how much data remains, and why it was excluded from the previous deletions, and it would be good to see it cleaned up so that we can reason about what we have stored.