Another reason why it is easier to integrate and use STM in Haskell than in most other languages, is the prevalence of immutable types. When only a tiny fractio

Another reason why it is easier to integrate and use STM in Haskell than in most other languages, is the prevalence of immutable types. When only a tiny fraction of all memory structures needs to be tracked by STM as TVars, it becomes orders of magnitudes cheaper than in languages with predominantly mutable data structures.