A distributed monolith is an application split across tightly coupled services, combining disadvantages of monolithic and distributed systems. Microservices need independent deployability and loose coupling to deliver their promised benefits. section 7.3 section 7.7
Unstable service boundaries and APIs can create this risk. For a new application, boundary and API changes are often easier in a monolith; start with one unless multiple technologies or compelling segregation needs justify services. section 1.4.1
Service-spanning operations also bring network latency, partial failure, and eventual consistency. Treat those costs as design constraints when deciding whether to distribute an application. section 1.3.4 section 7.7

