Experienced Clojure developers want a job where they can "do Clojure"
This seems like a common problem with functional languages in general. I've heard the same thing about Haskell, repeatedly. Such devs search out a place where a Haskeller/Clojurist got into a tech lead position and then all pile in, before you know it the team is spending half its time writing DSL compilers instead of adding features to the product.
I look forward to functional methods becoming mainstream (again?)
Has it ever been mainstream? AFAIK it's always been this basically strange offshoot of the family tree that split off a long time in the past.
The core problems you highlight don't have any obvious solution, unfortunately. Learning materials can be improved but ultimately Lisp is very old and very unlike more modern languages. Libraries not getting to 1.0 is a symptom of a small community that derives more satisfaction from an intellectual exploration than having lots of users, combined with lack of commercially driven outcomes. Startup time I guess can be solved with technical solutions (and is being solved, as a side-effect of other projects elsewhere in the JVM ecosystem).
This seems like a common problem with functional languages in general. I've heard the same thing about Haskell, repeatedly. Such devs search out a place where a Haskeller/Clojurist got into a tech lead position and then all pile in, before you know it the team is spending half its time writing DSL compilers instead of adding features to the product.
I look forward to functional methods becoming mainstream (again?)
Has it ever been mainstream? AFAIK it's always been this basically strange offshoot of the family tree that split off a long time in the past.
The core problems you highlight don't have any obvious solution, unfortunately. Learning materials can be improved but ultimately Lisp is very old and very unlike more modern languages. Libraries not getting to 1.0 is a symptom of a small community that derives more satisfaction from an intellectual exploration than having lots of users, combined with lack of commercially driven outcomes. Startup time I guess can be solved with technical solutions (and is being solved, as a side-effect of other projects elsewhere in the JVM ecosystem).