I have thought about this problem some and currently see it as an issue of scaling thresholds.
If it's literally just you, there's a lot of benefit to leveraging existing work so that you can focus on the differentiating part of the project(which is probably not a language innovation). It's not just the libraries but the whole ecosystem - example code, troubleshooting help, IDE support.
If you have an engineering team of even modest size, the picture can change very swiftly towards ensuring your result is built on a solid foundation, even if it means a lot of pioneering infrastructure has to be built and a lot of late nights spent debugging core toolchain issues. Team efforts have the necessary momentum to break free and do that ground work as the overhead gets swiftly absorbed in "person-year" budgetary terms. Individuals can only really justify the same as their core direction of research, sole hobby, or speculative investment - e.g. being the "first to implement" some hot new standard could be a well incentivized career move.
If it's literally just you, there's a lot of benefit to leveraging existing work so that you can focus on the differentiating part of the project(which is probably not a language innovation). It's not just the libraries but the whole ecosystem - example code, troubleshooting help, IDE support.
If you have an engineering team of even modest size, the picture can change very swiftly towards ensuring your result is built on a solid foundation, even if it means a lot of pioneering infrastructure has to be built and a lot of late nights spent debugging core toolchain issues. Team efforts have the necessary momentum to break free and do that ground work as the overhead gets swiftly absorbed in "person-year" budgetary terms. Individuals can only really justify the same as their core direction of research, sole hobby, or speculative investment - e.g. being the "first to implement" some hot new standard could be a well incentivized career move.