Starts really good, liked it immediately and I’ll read it all.
I fully agree with the premise – we can get very fixated on languages alone and that’s a problem. What we should be paying attention to are IMO the runtimes. I like Elixir a lot and it got a lot of things right in terms of language design and minimal stdlib API but if it wasn’t for the runtime it would probably never take off as it did.
There are many other possible selling points of languages – the portability of pure Go code, the memory safety of Rust, the DSL-friendliness of Racket, and maybe in the future an OTP-like clone in Pony, who knows? Languages that formalize the work like Haskell, OCaml and Idris can be extremely useful as well (I keep wanting to learn OCaml and can’t find the time and I am sure I’ll regret that one day!). Also overall, Go, Rust and OCaml seem very fast. Nothing wrong with orchestrating nodes in these languages inside an OTP supervisor, for example.
As I work on more on more different systems with time I grow to appreciate the value of good flowcharts and proper RDBMS / SQL reporting tools – and data migration, and reverse-engineering tools as well. Also concepts like How to create beautiful pipelines on Elixir with Opus (more options linked here: Flow based programming in elixir - #15 by amacgregor) caught my eye lately and I am due to researching them properly.
I personally am less and less interested in arguing with anyone about Python, Java, PHP, Ruby, C++ or whatever the holy language war is about during that particular day. I want to get my work done well and to be proud of it.
…Who knows, one day I might like BPMN as well. 
