27 karmaJoined


Hi Noah! Thank you for your questions and thoughts around my idea. And sorry for not replying earlier, I simply hadn't comment notifications turned on.

Your method is interesting. Have you thought about how you might implement it with a computer partner that helps you explore or develop knowledge? If applied faithfully, it might be an effective learning method.

That sounds interesting and no, haven't thought about it yet. I guess it would need some powerful AI behind it, that would be able to deeply parse the questions and get their meanings. Something I don't if it would be easily available at the moment.

On a different topic, one way to cheat this system ...

Regarding your thoughts about cheating in general: Yes, it is clearly possible in this format to act in a bad way and thus prevent it from being insightful. My pitch is not that this format will debunk "evil" actors.

It is rather for two people who genuinely believe what they believe but at the same time have Scout Mindset and honestly want to find out why and how someone else might believe something different.

Another way is to select "depends" conditions in some irrational way.

I see. My ad-hoc approach would be here to call them out once I notice such an inconsistency: "Do you agree that saying / agreeing to ... is inconsistent?"

Have you encountered those problems? If so, how do you handle them?

No, I haven't seen or had debates with such "unfaithful" debaters yet.

How do you use this with people who start refusing straightforward answers, but instead offer probabilities? For example, "Is X true?" "There's a 51% chance that X is true."

That's actually a nice idea! :) I'd totally "allow" such answers, if this is a way the debater wants to answer them and convey their beliefs. Regarding "who may ask next then?" I'd say: it's up to the 2 debaters to find a rule. A reasonable one could be "count > =50% as a Yes, thus keep the roles then".

what do you think demonstrates a good-faith debate partner, someone who will follow the rules and go through the process, that is, what evidence signals such a partner, before the debate begins?

Before the debate begins? Mhm, that's a hard question. But it's probably not one you need to answer because anytime during the debate, you can simply stop it once you notice your partner is not really helpful and not giving you enough insights.

To be clear: This format is not for "winning something", or for "playing well". It's a suggested framework that people can commit to and then use it as long (or as short) as they want to.

Thanks for taking interest in the project! About your question: I'm trying to think of an example to better answer it.

So let's say I've started a discussion with someone who disagrees with taking the Covid vaccine. And we get to a point where they don't believe the Covid stats because they don't trust the CDC.

Then at this point, I could either continue debating whether the CDC is trustworthy, or I simply leave it there because I realize that this is someone I do not share enough common ground with. (There would be still a benefit: My disagreement would be much more informed now.)

In any case I am sure that a moderator who would intervene and say "You have to trust the CDC!" will not be helpful to the disbeliever.

However, there are situation where I think a moderation could be necessary:

  • when the required conditions at Depends were not delivered,
  • or the false premise at False premise is missing.

Let's see if and how we get there.