Probe HN: A place to argue about truth of a statement | Hacker News Reader

Probe HN: A place to argue about truth of a statement

4 pointsoxplot5 comments

I came up with this idea years ago and spent a good time writing about its philosophy, etc. But I'm a practical person and even though the idea intrigued me and still does, I still have doubts about its usefulness.

I'd like to know what HN readers think of it. Would you find it useful to have a place where you can pose a statement which may or may not be true, and argue for or against statements posed by others? If it helps, think of it as something along the lines of reddit, but without the generality.

The advantage over reddit is how other users can benefit from the discussion. By requiring the arguers to choose a side (ie for/against) before commenting, it is possible to quantify the truth of a statement based on the quality of evidence/arguments provided in the discussion (quality can be determined by number of upvotes on a comment for instance). This requirement further forces the participants to do some research on their own before engaging in the discussion (ie because they have to pick a side first), leading to a more relevant and higher quality discussion.

This model would naturally yield itself to controversial topics and not obvious facts (ie no fun in arguing for things you can find the answer to using a quick google search). Of course there are many websites that currently host such discussions, but it is time consuming to get an overall picture of such without reading every post.

I had a similar idea a while back but it didn't have as noble a goal. Really just another time waster but I could (and still do) see some potential.

It was going to be an extension of something like HN or Reddit and it would have been called "Internet Fight!". The premise is that arguments happen all the time on these forum sites and I thought it would be funny to highlight them as the primary content of an application.

I was going to promote it primarily by using it on Reddit first. I would make it easy for anyone to initiate a formal fight between any two reddit users. For example, I may happen across an argument taking place between two users (reddit_user_1 and reddit_user_2) and I would reply to one of them with a link to their fight - say, "www.internetfights.com/reddit_user_1_vs_reddit_user_2/ or something.

It would create highlights, declare a winner at some point, allow others to comment on a parallel thread about the fight itself. I still think it would be great. Someone make it. As many fights as there are on reddit, your go to market plan is basically you just setting around initiating fights til it catches on.

Interesting. When I first thought about it, the concept of controversy, trouble, fight stood out. I kept thinking how I could declare a side the winning side and stamp a "Fact/Fiction" label on the statement at the end of a set period (ie a week after the statement is posed provided the commenting activity slows down).

I also thought long and hard about people who just wanna stir things up by constantly posting new statements, based on recent news (I even thought of writing a bot initially to generate such statements off news sites).

So what you're describing sounds very familiar to me.

Regardless of the actual site/format the quantifying of truth part seems like a big, interesting problem. Any good resources on that? A nice proof of concept?

I'm trying to wrap my head around it. I don't know. If I was you I'd just research the whole quantifying part and try to design some conceptual algorithms for that.

http://skeptics.stackexchange.com/ is somewhat tangential to that. Just putting the link out there. There's also a bunch of QA sites that force your to take a side, but I don't remember them right now ;C

And couldn't you do that with existing QA sites? Like Klout for truth, haha.

So could you describe what the discussion for, say, "P = NP" would look like? What would it score? .01 truthiness?

Are entire comments upvoted, or can single statements be up/downvoted line-item-style? Can the result of one discussion automatically influence another (can I post a statement /foo that assumes /bar is true and /baz is false)?

I assume it would be full of attempts from both sides to provide proofs, counter examples and lot of external links to existing materials on the web (a good argument could for instance increase the rank of an external page if we cared to keep track of such metrics). You're right, ideally it would score something close to the center of true/false scale (whatever number we assign it) but things like imbalance of count of participants on each side could create a biased result.

I really like the idea of upvoting a part of a argument and not the whole. I have also thought about inter linking of statements which is useful to keep each discussion within some bound.

Reply on news.ycombinator.com