Puzzle-writing goals, part 2

I’ve been meaning to write this post for a long time, but every time I sat down to work I instead just wrote another puzzle. (Go figure.) But since it’s about the end of January, I figured I might as well do it now.

I’ve finished the January goal — in fact, I’ve pretty much blown it out of the water. With 8 puzzles posted to my website and 6 more kept private for various reasons, I’ve been able to write at a pace of around one puzzle every two days.

Puzzle-book

For the puzzle-book I discussed earlier this month, it might behoove me to reconsider which variants I include more of. Equations is particularly easy to set, while Consecutive/Dutch Neighbors is incredibly hard to set. Also, if I want to make this book entirely approachable (rather than targeting both beginners and experts), Clue Placement (i.e. Alphabet Soup + maybe other puzzles I’ll set later) might not be appropriate. Every other puzzle (so far) seems like it’ll be 2-3 stars of difficulty, so it’d be weird to force in a 5 star puzzle.

Also, I don’t know how milkable the 7-Eleven concept is. I tried setting a puzzle entitled “24-7” where the orthogonal neighbors of each 7 summed up to 11, but no such grid exists. (Exercise for the reader: why?) Instead, to preserve the theme and many of its ideas, I forced the 7 to have orthogonal neighbors of 2 and 4.1 The Averaging Circles variant seems much more tractable, since we don’t have to deal with global constraints, so global digit neighbor constraints will likely cover a broader range of puzzles in the book.

In general, I’m thinking that as long as my puzzles have unique, semi-related constraints, it will be fine. It is probably going to be difficult (and somewhat boring!) to literally write 7 puzzles with the exact same constraint, so the sections of puzzles in the book will likely be roughly grouped by theme, rather than strictly organized into constraints.

New Year Goals

As for the rest of my goals, I think for February I should probably up it to four new variants. I originally was thinking five, but between writing this post in my notebook and typing it up I managed to use Renban lines, X-Sum, and Little Killers (speaking of the latter two, I really ought to complete my sudoku rules intro). So the goal will probably be a bit harder than initially anticipated, particularly since some of the constraints I have left are annoying global constraints like Anti-knight.2 But there are a ton of other variants I didn’t even mention, like Kropi, Sandwich, Odd/Even, Diagonal, Anti-Diagonal, and Indexing.3

On the other and, my goals for March and April were grossly easy. (Which is a good ting, because if I set low standards I am virtually guaranteed to meet them and not get discouraged, etc.) As I’ve already discussed, I’ve managed to set two Classic Sudokus, and a few days ago I also set a Star Battle. So maybe I should up the ante a little bit and change March/April goals to themes. Try to set 4-ish classic sudokus in March and 4-ish pencil puzzles in April, and increase the focus in those genres for those months.

May Goal

I’ve only set one 5-star puzzle so far (Alphabet Soup), and I think it’s much easier for me to set 2-3 star puzzles than 4-5 star puzzles. (1-star puzzles I can do, but I don’t do it as often because the space to explore is smaller with easier puzzles.) So a new goal for May: publish puzzles on Logic Masters Germany such that I have at least one of each rating (1-5 stars). I say publish because I’m a pretty bad judge of difficulty (I originally thought Bird was a 2-star puzzle at most). So depth/range is one of my goals now.

(Of note: I’ve never written a 4-star puzzle.)

Cracking the Cryptic

I know it’s been a while, but Close Neighbors has been featured on Cracking the Cryptic too. I think the debut feature was a much bigger deal for me (since, y’know, first time), and probably I will get less surprised with each new puzzle featured.

Of course, it’s always going to be a nice thing to see my puzzles make it, but I think it does not make sense to make a post announcing each new feature. So I will probably mention features when they happen if I’m already writing a puzzle blog post, and otherwise, I’ll probably just silently update my puzzle page.

Taking a break?

Maybe I should take a break from setting or reduce the pace to get my life kind of together. (Right now I’m barely even a functional human being since I spend so much time setting.) I say this, but I know with pretty high probability I won’t intentionally do this. A couple of tings that will naturally make me slow down, though:

There are other things I want to do, like finally finishing my Git server4, pivoting back towards teaching for Math Advance, etc. So yeah, a steady pace of about 3-5 puzzles a month (and not 14) is probably what I’ll converge towards.

Thanks for reading, and thanks for trying my puzzles.


  1. This is one of the puzzles I’ve kept private — gotta leave something for the book! Instead of using cages, this time I used thermos, so if that interests you, keep an eye on the book!↩︎

  2. Don’t get me wrong, anti-knight is pretty fun to solve when done right, but trying to write it is absolute misery. Particularly because if I use the anti-knight constraint, I’ll have to center the puzzle around anti-knight deductions or at least base a large part of it off that, and that seems kind of annoying.↩︎

  3. My original draft of this post had Little Killers, X-Sums, and Renban. But I just had to make it harder on myself :D↩︎

  4. Fuck college apps. Seriously. They completely killed any momentum I had going.↩︎