RECORD OF THE YEAR

I started writing this puzzle a long time ago and was convinced that 1) with 12 rebus squares, I couldn't fill it, and 2) couldn't keep it resembling a calendar while keeping the word count beneath 140 words. So I set it aside for a better part of a year before picking it up again. When I did, I found a layout where each month was relatively the same size, relatively rectangular, and not-too closed off from one-another, and forged ahead.

There ended up being three major challenges to filling this puzzle. First, the many straight edges. Constructors know that it is way easier to fill puzzle sections that have diagonal edges rather than straight edges. With straight-edges, it becomes difficult to avoid words that have terminal non-RSTLNE letters, so one ends up with a lot of fill that is, shall we say, STRUCTURAL. Second, there are very few three-letter entries in the grid, which are a constructor's best friend in otherwise hard-to-fill sections. Finally, I didn't want any of the rebused entries to share a common root with their corresponding month. This was particularly hard for July, which comes from Julius, as do just about every word and name you can think of that starts with JUL besides JULEP. BANJUL, then was my only other JUL-containing option, which as a world capital, is fully legit, but as a city of only 30,000 people, is pretty small potatoes. But so it goes!

I hope this Sunday puzzle — my first! — kicks off your new year on the right foot.

Normally I don't care for grids that segment puzzles into sections, but this design was so appropriate for the theme. Cool idea to have twelve "boxes" — just like a calendar! Great a-ha moment when I first uncovered (JAN)GLE and TRO(JAN)S, turning my initial grousing about the puzzle's (quasi-)segmentation into compliments.

We've seen a lot of rebus puzzles now, one with the exact same rebus strings and another one still. But David's calendar-looking layout gives it a unique execution. Even though I knew what the rebus square would be in each box, I still had to work to uncover it. Nice balance of deviousness and solvability.

A puzzle like this which uses so many short themers is hard to build — due to word count maximums, if your themers aren't long, your fill has to be. David did well here, with DIETITIAN, RAW SCORE, BEER STEIN, CROSSBAR, etc.

Usually I prefer entries with rebus strings to be long — LIFEBLOOD, SOTOMAYOR, CARL JUNG — but I enjoyed BAN(JUL)'s NYT debut, as I traveled to the Gambia in 2009 with a non-profit org. I learned a ton traveling upcountry for two weeks. The Gambia is a little snake-like country that the British carved out of (French) Senegal, specifically to claim the Gambia River. Crazy colonialists.

I didn't enjoy I(MAR)ET as much. I find it so much more satisfying to uncover something even like (DEC)OCT than I(MAR)ET, wondering if the latter is a valid word. I happen to know it from crosswords, but I liked the WI(N OV)ER / CASA(NOV)AS discovery much better, for example. Personal taste.

ONEK is such an odd entry. The first time I encountered it, I was sure it had to be wrong. It does parse to ONE K, but 1.) it's rare to have a 1-K race, and 2.) it's even more rare to see a race written out as a Five K or a Ten K. Personal preference, but I much prefer entries that one sees in real life. To me, that one sticks out much more so than INTRA or ETATS or even IAL or BRRR.

Very neat effect today with the calendar-looking grid.