Verbal Zendo

post by abstractapplic · 2018-10-21T21:51:11.643Z · score: 10 (7 votes) · LW · GW · 5 comments

You're probably familiar with the 2-4-6 test. Well, I made an automated version of the verbal version of the Buddhist version of that. In essence: there's a rule that declares some phrases "have the buddha-nature" while others don't, and you get to keep testing ideas against it until you're pretty sure you've figured out the rule; then, you can check to see if you were right. If that sounds interesting to you, it can be played in-browser here.

Warning: Don't let yourself get nerd-sniped. My puzzles are not optimised for fun or meaning; if you find yourself spending more than an hour or so on this weird toy of mine, just look at the answers and move on.


Comments sorted by top scores.

comment by Said Achmiz (SaidAchmiz) · 2018-10-22T00:07:16.996Z · score: 10 (4 votes) · LW · GW

Usability feedback: please make the Enter key also submit, not just clicking the Test button!

comment by MakoYass · 2018-10-22T04:43:04.649Z · score: 6 (3 votes) · LW · GW

Here's something we can run in our url bars to get this now


javascript:document.getElementById("theInput").addEventListener('change', ()=> test())


comment by MakoYass · 2018-10-22T04:46:05.174Z · score: 4 (2 votes) · LW · GW

Relatedly there's a nice little singleplayer zendolike for android here

I'd recommend it

One of the things you might like to steal for future iterations of this is its testing process. To demonstrate that you know the rule, before you're given the green light to move on, it quizzes you on a large set of examples and you have to be able to categorize them all correctly.

comment by Nebu · 2018-11-06T06:20:16.293Z · score: 1 (1 votes) · LW · GW

I feel like this game has the opposite problem of 2-4-6. In 2-4-6, it's very easy to come up with a hypothesis that appear to work with every set of test cases you come up with, and thus become overconfident in your hypothesis.

In your game, I had trouble coming up with any hypothesis that would fit the test cases.

comment by Zack_M_Davis · 2018-10-22T04:48:33.927Z · score: 1 (1 votes) · LW · GW

Another fun programming exercise is to do the other direction: have the user come up with a rule, and make the program come up with examples to try to test its hypotheses. (You want the program to generate examples that falsify half the remaining probability-mass.)