For those interested, I've slowed work on the app in favor of refining the app's ontology in markdown form[1], building an ontology playground[2], and building a prototype[3] to show off features that the updated ontology enables. I had some major changes I wanted to make and making it work in the real app was going to be too much effort.
I'll likely be making a Show HN for one/some of these soon^TM (in the next month or two?) but they're currently still a bit early.
[1] https://github.com/keyserj/reasoning-tools/blob/main/amelior...
[2] https://keyserj.github.io/reasoning-tools/ontology-playgroun...
[3] https://github.com/keyserj/reasoning-tools/tree/main/amelior... (not yet deployed)
- mind explaining it like how you would do to a 5yr old kid
in any case: cool
I might need LLM help to figure out how to explain to a 5 year old haha.
Regarding styling edge types: Yes and I think it's a good idea. I think there are some things I haven't tried yet, but I added logic gates https://github.com/amelioro/ameliorate/issues/785 (More Actions Menu > Preferences > Edges > Use semantic arrow shapes) and there's an open ticket for using colors to convey good/bad https://github.com/amelioro/ameliorate/issues/836. Thickness is another thing that potentially could be played with, thicker being higher score? Hadn't gotten to that.
Ameliorate models questions and arguments, but they're more auxiliary (modeled as being _about_ a cause or effect) than in IBIS, where they're pretty central. Questions in Ameliorate are more "unknowns" than "core questions we are trying to answer".
If you see my root-level comment in this post, I mention refining the Ameliorate ontology (ameliorate-v2) which elevates questions to first-class (to address the issue where it's hard to know where to look in the diagram, without questions to identify motivation). You might also be interested in my ontology playground in that comment if you want to see an IBIS > mermaid pipeline.