Trace: Goals and Principles

📖 Description

In terms of research, I decided to devote the month of February mainly to foundations and tools. One project was to come up with a notation/language/framework which matches the way I've been thinking about computation - i.e. [DAGs with symmetry](https://www.lesswrong.com/posts/mZy6AMgCw9CPjNCoK/computational-model-causal-diagrams-with-symmetry) and ["clouds" representing DAG-structure-as-data](https://www.lesswrong.com/posts/G25RBnBk5BNpv3KyF/a-greater-than-b-greater-than-a-in-causal-dags). The tool I've been building - a Python library tentatively called Trace - isn't stable enough that I want to show it off yet, but I do think I've nailed down the core goals and principles, so it's time to write them up.

📊 Game Impacts

💭 Reactions

🔗 Sources

• [1] https://www.alignmentforum.org/posts/rt5X74Az3mXwTubRA/trace-goals-and-principles