Idea: break up Alda components into discrete services

[daveyarwood]

As discussed in my blog post here: https://blog.djy.io/a-history-of-the-architecture-of-alda/

I'm interested in further breaking up the components of Alda into separate processes, to make the system more modular. This would allow developers (myself included) to experiment with rewriting parts of Alda in other languages which might be better suited for what that component needs to do.

I'm interested in other developers' opinions about whether this is a good idea, possible pros and cons, etc. Please weigh in!

[daveyarwood]

I've been doing a lot of planning and design for Alda 2.0, and I've decided to head in the opposite direction and do as much as possible in the Alda client, i.e. simplifying the architecture instead of making it more complicated.