Hey,
It might just be my impression, but it feels like some kind of intervention is needed regarding the external crates.
- There are far too many crates out there which are essentially incomplete and they are neither complements or substitutes. I am sure I am not the only one who have ended up spending days jumping from crate to crate, their source code and incomplete documentation, just to find out that crate A supports only one half, crate B the other and they don't complement each other.
- The "survival of the fittest" seems to have extended to the adaption of names which should have been reserved (we can find every possible iteration of the words "google cloud storage" out there, while none of them is essentially a complete API of google storage)
- Rust's liberalism through traits and generics seems to have become an issue with some APIs, where essential functionalities are not provided out of the box transforming them to something more like "DYI APIs" (tracing's json fmt is a great example, where you need to play finding Nemo to rename a json field).
What do you think on a process with strict criteria on the quality of documentation, code and completeness to shorten the period of open source "survival of the fittest" where many scratch the surface of the same problem but nobody actually solves it?