Very few areas of databases in general, but SQL Server in particular, suffer as many misconceptions and outright misinformation as these. Developers seem to maintain these nagging mental injuries, like row versioning isolation levels allow dirty reads, or locking isolation levels guarantee snapshot-type consistency of reads, that are simply wrong. In these modules, I'm going to show you the truth.