Message queues have been widely used in networked systems for their ability to offload long-running tasks from critical paths, ensuring reliability through guaranteed delivery, persistence, and dead letter queues. However, building a system with message queues requires much more than just the queue itself, including chaining, concurrency, back-off, debouncing, state persistence, prioritization, error handling, idempotency, cancellation, observability, and recovery tooling. Durable execution is now considered the right abstraction for building modern systems, combining durability and flow control to provide reliability and guarantees to application workflows. Developers should consider using durable execution instead of directly implementing message queues in their applications.