Pushed the first version of the last feature I wan...
# devlog-together
j
Pushed the first version of the last feature I want to add in 2022, relevance visualization in the fact editor. You fill in what you know. If you don't know, or there might be more values, you "unlock" that category or attribute. Run the test. It shows you the answers it finds, and any assumptions it used to find them. (e.g. bob won if jane threw scissors.) Any inputs relevant for finding new unconditional answers (e.g. what jane threw) are highlighted in green.
Still all ugly as sin, of course. 🙂 But I now have a usable platform for demonstrating explainable, declarative encodings of legislation, with hypothetical and reverse reasoning, etc.
🍰 2
i
Interesting! How do you get "jane threw scissors" by abduction? How do we know jane didn't throw paper?
Ah — I see what I missed. Jane throws scissors just in Answer #1. Presumably, Answer #2 is jane throwing paper (jane wins), and Answer #3 is jane throwing rock (tie)?
j
Close! 🙂
Answer 1 is bob wins. Only one possibility, Jane threw scissors. Answer 2 is jane wins. Three possibilities, there: jane beats jane with each of the three signs. (The rule is poorly encoded, because it doesn't require a disunity between the players.)
👍 1