Jumpy [SPOILERS]

[Kory Heath]

Actually, this message doesn't contain any real spoilers for Jumpy, but its follow-ups might.

Last night I played a solitaire version of Jumpy (difficulty level 1). I won the game, but I didn't figure out the pattern. I haven't been able to figure it out even after looking at the whole pattern for a while. Anyone else have any luck? Ryan, wanna fill me in? (Maybe offline?)

-- Kory

[Ryan McGuire]

Since, the subject points out that this is a SPOILER thread, I don't
feel too bad giving away the answer.

[SPOILER ALERT, WOOP WOOP]

Of course you noticed that the tiles are colored in pairs:
blue-blue-pink-pink-green-green-etc. I also would have to believe that
you realized that the curved lines between rows meant this was one long
sequence as opposed to a 2-D grid.

The symbols at either end of a single-color pair shows how far away the
next pair of the same color is. For instance, a star as the second
tile in a blue pair means that there were six tiles between the star
tile and the next blue tile, and a circle as the first tile in a pink
pair means that there are four tile between this and the PREVIOUS pink
tile.

Square - 0
Triangle - 2
Circle - 4
Star - 6
Quarter circle - 8
Diamond - 10

I feel like a first-time Zendo master. "What??? You didn't get my
rule? But it's so obvious!"

[Ryan McGuire]

By the way, the name "Jumpy", hints that the symbols show how far you
have to "jump" to the next tile of the same color. Of course, that's
too obscure to be useful in solving the pattern.

[Kory Heath]

Hi Ryan,

Your pattern reminds me a lot of one of my own early attempts entitled "ESP". I've attached it in case you want to take a look at it. Don't bother playing it, though - it's too hard. The "rule" in my pattern is that each tile is colored based on how many shapes there are in its row and column which match it. For instance, a star tile will be red if there are no other stars in that row or column, yellow if there's exactly one other star in its row and column, etc.

Jumpy is a neat pattern, but it's too hard for Paradigm. Like ESP, it would mesh better with an "experiment-game" like Zendo or Eleusis. Imagine if you just started with a couple of tiles on the board, and you could try to add any colored shape you liked at the end of the sequence, but the Master would only allow correct additions. Through experimentation you could figure out (for instance) that any colored pair that ended in a square is always followed by another pair of the same color, and from there you could probably figure out what to try next. As it is, there are only two squares in the entire pattern (besides the end-caps), so it's more difficult to make this connection. Zendo and Eleusis allow you to home in on sub-patterns by creating many instances of the concept you're exploring.

All of this suggests a new general principle: a pattern should *not* function as a puzzle when it's completely revealed. In other words, if you show me a completely revealed pattern, I should almost instantly be able to figure out what's going on in it. If I can't, it's almost guaranteed to be too difficult for Paradigm. This is an important finding, because many people's initial intuitions (including my own) will be to design puzzles.

I notice that Jumpy can be viewed as a "group-based" pattern or as a "tile-based" pattern (depending on how you state the rules). So I'm beginning to doubt that my concept of "grain-size" really captures the exact feature I'm looking for, although I still suspect there's a strong correlation between the two. I'm now tending to say the following two things:

1. Good Paradigm patterns are "exploratory" rather than "experimental". If a pattern would work well with an experiment-based game like Zendo or Eleusis, then it probably won't work well with Paradigm, and if a pattern works well with Paradigm, it probably won't work well with a game like Zendo.
2. Good Paradigm patterns are completely obvious when revealed. If a pattern provides any kind of a puzzle for someone looking at the entire thing, then it's almost definitely too difficult for those who can't see the whole thing.

-- Kory

[carlto...@gmail.com]

I suppose I could maybe gloat now. I figured out the pattern about half
way through. I was in a puxxle solving frame of mind though. I probably
wouldnt have gotten it otherwise. Kinda Like Cryptic crosswords. I have
to be in the right frame of mind for those. Some days theee clues just
jump right out at me and other times I can stare at them for hours and
they make no sense.