[jbclements]

Yes! I wrote about this in a widely-discussed ^H^H^H^H^H^H^H^H^H^H^H^H^H^H^H^H^H totally-unread blog post back in 2014: https://www.brinckerhoff.org/blog/2014/12/12/the-why-button/

[Edmond]

Hmmmm

You're making quite a bold claim, I don't see you make any reference to ML, if I were to draw inference from your post I would conclude you were talking about better ways to instrument software so you can track what it is doing, not a particularly novel idea :)

The paper referenced in this post talks about a technique for teasing out the "logic" for a learning model's decision.