Aren't all distributed databases basically just clever wrappers around write a head logs?

From my small experience in writing an event database, I am inclined to agree with this.