Ask HN: What is your “I don't care if this succeeds” project?

https://closedcaptain.com/

[0] https://wallsmash.com

v2 on master is as simple as it gets, but still incredibly functional; my team is dogfooding the hell out of it at work.

v3 on the "commandbased" branch is a total rehaul on the works, hoping to make this a more traditional/complete package, with a command-based interface (i.e. similar to how git works)

First is a site that has existed for around 25 years now, although it was shut off for several years. It came back to life when the pandemic started, and is behind private auth now. Basically, it lets me and other creative writers collaborate on branching fiction novels together. I write a chapter and put a couple of choices at the end, which serve as writing prompts. A reader comes along, clicks a choice, and is faced with the need to write that chapter, and so on. Our favorite story is currently 288 chapters, and is projected to reach about 1000 pages when it's done, which will probably happen in the next few months. We're all good enough writers to pay attention to characterization, dialogue, and theme, so all the threads tend to reach natural narrative conclusions, and I'm even able to export it to a printed "branching fiction" format and publish the books through Amazon KDP. Someday I'll find a lawyer and figure out how to open it to the public (it supports multiple in-progress stories) but for now I'm quite enjoying having weekly Discord meetings with my writer friends, writing a handful of chapters per week. As for the tech side, probably the most fun recently was writing a react component that lets you graph out the animated chapter map in all sorts of fun ways.

Second actually has some technology in common with the above. I'm enamored with using propositional logic (not predicate logic) to make and verify argument graphs. Meaning, you create some nodes with sentences and truth values, and build up actual visual graphs with sentences in bubbles that you can derive from other premises, and then collaboratively judge whether the truth value is propagating throughout the graph. It would enable people to browse to someone's conclusion, and then explore down through their entire argument down to their facts (which should have the same truth value for all people) and values (which should have truth values that can differ from person to person), and then people could pick points in the argument to disagree with or challenge. Kind of my own attempt to turn the current age's destructive argument habit on its head, into something constructive and collaborative, in a journey to discover shared truths, and opposing perspectives that we can respect. Like I said, ambitious.