Like an inner conversation? That seems a lot like how I think when I consider a challenging problem.

Thank you!

Currently it is just a bit of pattern matching and polymorphism that pulls in various libraries according to entity type. If it picks up any interest I'll put some thought into how could be a bit more elegant / modular.

From the article: "70% of the soft plastic that reached a known destination was burnt, not recycled".

That doesn't seem too bad. That means 30% got recycled, which is more than I expected given the general cynicism around plastic recycling. 30% definitely seems worthwhile.

For the plastic that got incinerated, it is presumably more efficient to burn products for energy reclamation that are already separated by type.

Having these recycling pathways set up is a necessary step towards improving recycling rates. We should be able to improve on that 30% over time.

A small, cute CLI app that gives overviews about files and other entities. I've called it `what`. It's inspired by being able to right-click on stuff in a GUI file manager to get an overview about what it is.

https://github.com/richsmith/what/

> This could probably slip up a human at first too [...] > breaks the illusion that there's real human-like logical reasoning happening

This does seem like the sort of error a human might make. Isn't the problem here that the model is using reasoning that is too human-like? I.e. error-prone pattern matching rather than formal logic?