The approach is good, the results are encouraging as well. Not much effort and a very visible change. We need more such initiatives.

The situation is still complicated for maintainers... And companies benefiting from their free labor don't get it. This leads to really stupid situations.

This is a funny but interesting productivity tip.

This is an interesting and deeply buried optimization for the GNU C++ STL implementation. I didn't expect anything like this.

Clearly the error handling landscape still evolves in Rust and that's a good thing. The current solutions are too fragmented at the moment.