Go Zendo

You’re correct, in fact, it must be equivalent to a rule of this form.

It might not be the most easy way to describe the rule in that form though :stuck_out_tongue:

7 Likes

Could you elaborate on in what sense knowing the color of the empty board pins down the first quantifier? I thought I understood this before, but after thinking some more about it it doesn’t make sense to me.

If all rules referred only to stones, I could see how a red empty board would imply the rule being of the form “there exists stones such that…”, while a green empty board would imply “for all stones it holds that…”. But what happens with a rule of the form “for all empty intersections…”?

8 Likes

Counterexamples:
image image

For the following rule guesses I would suggest a rule similar to the original Zendo: You must provide two koans, one that you believe is green and one you believe is red. If you are correct you are able to make a rule guess where I then have to give counterexamples. Otherwise you’ve already generated your own counterexample. That way you have to test first with koans whether your rule might be correct.

7 Likes

image

1 Like

Yeah, you’re completely right :slight_smile:

6 Likes

I noticed that all four red koans revealed so far have either EWEBEWE or EWEEBWE on the middle row. Let’s explore this further:

imageimageimageimage

2 Likes

All red
imageimageimageimage

2 Likes

Hm, what about moving the White stones’ positions from koan 488?
(The previous 4 koans all contain the initial stones from koan 488.)

image
image
image
image

1 Like

imageimageimageimage

2 Likes

Koan_513

Koan_514
​

​
If 513 is green, the rule could be something like

All maximum-size white chains see all other white chains,
where at least the black stones have non-zero size

.
​
​
508 is what would force the black stones to have non-zero size.
514 is my attempt at inferring what that size might be.
(However, if the rule is something like chains seeing chains, then
it might be using rectangles instead of Euclidean lines and radii.)

Variations of #510:

imageimageimageimageimage

I sort of assumed that the rule is invariant under isometries while crafting these koans, but we don’t know for sure.

Indeed 513 is green and 514 is red, so you can now guess a rule :slight_smile:
imageimage

But I’m not sure I understand what you wrote, can you clarify?

1 Like

imageimageimageimageimageimage

2 Likes

I do not yet have a rule guess. ​ I was just putting my thoughts out.

I think your idea about lines of sight is good. I haven’t checked carefully, but in all green ones with more than one white chain I think each white chain can see at least one other white chain.

3 Likes

This seems to apply horizontally, but not vertically, and perhaps for both colours.

image
image
image

Edit: And this koan contradicts koan 528 in that:

image

1 Like

If it is about line of sight along rows / columns, then I’m wondering if the following Koan is Red:

Koan_541

2 Likes

image

2 Likes

I have two rule guesses:

  1. If there are at least two white groups, then every white group can “see” a different white group.
  2. If there are at least two white groups, then there exists a white group that can “see” all white groups.

In the sense that group X sees group Y if there is an orthogonal straight line of empty intersections between two stones of the two groups.

Under rule 1, I expect the following to be both green, while under rule 2, they should be green and red.

imageimage

3 Likes

imageimage

4 Likes