I've been writing Rust code since before the 1.0 days, and I still can't understand lifetimes in practice.
When the compiler starts complaining about lifetimes issues, I tend to make everything clone()able (either using Rc, or Arc, or Arc+Mutex, or full clones).
Because if you start introducing explicit lifetimes somewhere, these changes are going to cascade, and tons of annotations will need to be added to everything using these types, and their dependent types.
I think you're really intended to do the latter rather than the former. I mean, Rust lets you do either–it gives you the choice if performance isn't your concern–but usually it's better to not clone everything.
I find that lifetimes are ok, albeit annoying sometimes, especially the cascade part, as you mention. The one thing I can't get to stick in my brain is variance. Every time, I need to go back to https://doc.rust-lang.org/nomicon/subtyping.html#variance
When the compiler starts complaining about lifetimes issues, I tend to make everything clone()able (either using Rc, or Arc, or Arc+Mutex, or full clones).
Because if you start introducing explicit lifetimes somewhere, these changes are going to cascade, and tons of annotations will need to be added to everything using these types, and their dependent types.