I wish agent-based programming was more popular today. It's an extremely powerful way of thinking about complex problems and distributed solutions. And it has much more actual theory behind it than things like SOA (which is mostly a re-branding of good OOP practices with a lot of buzzwords added in).
What downsides of frameworks cause you to want a domain-specific language? Python is great at extending the language using frameworks, and I think the addition of async/await and event loop libraries makes it really fun to do agent-based programming using coroutines, especially with the Curio library[0].
Given async/await, coroutines, and a good library to drive the event loop, what else is missing that you would like language-level support for?
> What downsides of frameworks cause you to want a domain-specific language?
Because it is a kludge. The language has a different paradigm it was built on and there will always be mismatches. Coroutines are not agents and don't have all of the properties of agents specified in [1].
> Given async/await, coroutines, and a good library to drive the event loop, what else is missing that you would like language-level support for?
I don't want to program in a language that doesn't treat agents as the primary unit of organization with their own threads and communications. I don't want to have to think in terms of a pieced together system. Frankly, having objects that can act without agents is just a painful idea.
Complex system simulations (look up Net Logo). Distributed, fault-tolerant data processing systems. The latter is harder to back up, because all real-life "agent" systems I've worked with are rough approximations of "agents" made with commonplace technologies and don't use the concept to its fullest extent.
