I like the author's division of the issues into soil, surface, and atmosphere. I've definitely made project choices on all three. Ruby is not great, not terrib

I like the author's division of the issues into soil, surface, and atmosphere. I've definitely made project choices on all three. Ruby is not great, not terrible. I enjoyed learning the language. I didn't find much in its runtime or its constructs to recommend it over Python. But I haven't used it in years because the first medium-scale project I wrote in it I put down for two years, came back and found that Ruby on Rails had changed so much that my project wouldn't build; every dependency had gone through a major revision and Rails' preferred package manager had changed. The prospect of doing a lot of work to re-tool a build environment that had worked perfectly fine soured me on further activity in the space. This was nothing wrong with the language and everything to do with the ecosystem being so young that "the adults were not (yet) in charge." Contrast C++. One of the most supported languages on the planet; adequate performance for my needs. I can't stand the constructs in the language itself. The templating system couples badly with other language features and they "fixed" it by cutting firebreaks in the parse process via newly-required keywords and template parse rules (all taxonomies are broken, and there's a touch of atmosphere here too; I got to ignore all those changes for a decade because MSVC basically told the standards committee to pound sand until 2017). It's a Turing-complete language, you can do everything with it any other language can do... But doing it by having a template resolution hinge on the implicit typecasting of an integer constant `0` makes my skin absolutely crawl.