Something I'm immediately thinking about with this is change management and inertia at the early stages of a new, underdefined project. Less code is great, the big question is how such a system compares to the usual hack-and-slash method of getting a v1 up and running as you search for PMF from the perspectives of ops, cost, data migrations, rapid deployments, and so on. Presumably, the idea here is to start from the beginning with Rama, skipping over the usual "monolith fetches from RDBMS" happy paths, even for your basic prototype, this way you don't slip into a situation like Twitter did where that grew slowly into an unscalable monstrosity requiring a rewrite. So an article focused on the "easy" part that's required in the beginning of rapid change, as much as it's not as important as the "simple" part that shines later at scale, seems useful.
The basic operation Rama provides for evolving an application over time is "module update". This lets you update the code for an existing module, including adding new depots, PStates, and topologies.
