3 ms·
I have an early prototype of a site that allows people to chain together reasoning in a visual sense, which also propagates truth so that people can know if a c
by tunesmith 6y ago
I have an early prototype of a site that allows people to chain together reasoning in a visual sense, which also propagates truth so that people can know if a conclusion is asserted to be valid/proven. This contrasts from the current definition of "argument mapping", and platforms like Kialo, in that it's more constructive, with logical machinery underpinning it. For instance, if someone has a conclusion that "Earth's atmosphere should stay below x anthropogenic GTCO2", or "SpaceX will be able to reach Mars by 2030", other people can then drill down through the argument until they get to simple axiomatic facts and values. The idea is that by doing so, people would either educate themselves if the argument is valid, or have the opportunity to specify exactly where they disagree (which would in turn propagate truth/proof state to the therefores that depend on that part of the argument), which is a respectful way of engaging with and perhaps strengthening the argument. In a social sense this could be useful for polarized communities to better understand each other over time. In a private sense (revenue opportunity) it could help organizations document the reasons behind processes, or make collaborative arguments for change that rely on deep first-principles reasoning.
This project is a little insane, as it relies on some optimistic assumptions about human nature that might just prove to be false - at worst, it is an opportunity to deeply explore an idea that requires some fun distributed-systems technology. I'm very much a tech/backend guy but it could evolve in a direction where I'd have to hire phd-level people. Anyway, I have a long list of features to develop before I get to real business planning, assuming I don't just eventually release it for free wikipedia-style, but what I'm mostly looking for now is a small group of people that are highly interested in following along (please explain your background and why you are interested if you contact me). I have a read-only prototype already pushed live, which allows exploration of existing arguments/conclusions, so as I make releases and update the changelog, I would like people that would like to follow along and make comments, and also perhaps give free U/UI advice, since there are a lot of challenges there in making a "new form of communication" feel intuitive. I'd imagine any possible cofounder arrangement would inevitably rise from that.
- 0x1F8B 6y agoJust want to say that this is a wild idea and I really like it. I don't know how viable it is but it's definitely interesting.
- tunesmith 6y agoThanks. Believe me, I share the uncertainty so I regularly have to hold myself back from getting too excited about it. It is really fun exploring the tech in the meantime though.
- jbms 6y agoThanks for letting me know about Kialo. It takes a pro-cons approach breaking things down into modular claims. What's not enough? A claim could be A->C Supported by separate claims: Since A; A->B; B->C, etc?
- tunesmith 6y agoMore like, if it's that A and B are collectively sufficient to guarantee C, and A and B are true, then you know, visually, that C is proven. If one of A or B are contested and (validly) claimed false, then the truth/proof state propagates and you know, visually, that C is not proven. And then yes, C can be used to support something else, so you can chain multiple together. Kialo - and really, the whole warrant/claim method of argument-mapping - doesn't have the concept of truth (or sufficiency, or necessity) built into the machinery. It goes deeper than that, there's a distinction between syntax and semantics (or form and content) - all of these design decisions are driving exactly how the project evolves.