New York Times, Thursday, May 23, 2019

I conceived of this puzzle five years ago this month, not long after the publication of two Tuesdays that would later factor into a decision that may have made the difference between rejection and acceptance (more on that to come.)

At first, I tried making the themers as long as possible to fill out a 21x, but I quickly abandoned that, as it became clear that the single-word themers wouldn't make for an entertaining Sunday solve. Shortly after, I asked my friend Jacob McDermott, whom I'd first contacted several months earlier about a themeless (and is getting married on Saturday!) if he could breathe new life into this concept.

We both ended up making two lists of themers, each of which had between 74 and 82 total letters, which felt like "no puzzle's land." That was too few for a Sunday even with the revealer but too many for a daily without many compromises in the fill. Moreover, there were 22 different themers among the four lists, but only six of them appeared in all four, and this fell into development hell.

Trying 15x14 and 15x16 grids about a year later did not revive it... not directly, that is. It was while I was playing with one such grid that I hit upon splitting FOURTEEN and POINTS, making a symmetrical 15x grid possible. The two-part revealer stayed in its current place from the very start, as I immediately noticed that BROWNIE would work nicely crossing it, and coming up with TIPPING was indeed a tipping point for the puzzle, as it allowed not only PRESSURE but also GRADE and STAND to form theme crossings. I am also glad that Will kept my reference to Gladwell, as he was one of my sister Maria's favorite authors at the time.

Still, by spring 2016, it had become clear that packing in all this theme would indeed force many compromises in the fill. I had not yet found a complete fill by April 20, when Tom McCoy's puzzle with the same revealer ran. Grudgingly, I shelved this even though it was a very different take on the phrase, and I suspected that it might not be fair game for the NYT when I was able to bring it back out. But remembering the two Tuesdays from early 2014 (January 14 and February 18), which both had over a dozen themers that nicely justified going over 78 words, I split up 2-Down and the then 39-Down. That made all the difference in the world when it came to filling the NW and SE, in both avoiding those nasty 3's and 4's and opening up a lot more and livelier possibilities for the two remaining 8's — which Will said he liked when he accepted it last September 13.

Thanks as always to Will and his team for polishing this, both with the clues (I got a real kick out of theirs for 35-Across) and fill, specifically removing the Roman numeral I'd had at 65-Across. Seeing GREECE as the new 48-Down made me remember my Pappou (Greek for grandfather), who died at 93 in 2013, and my Yiayia (grandmother), who turned 91 on May 10. She solved crosswords almost every day for many years, but since this may be my last chance to do so, I dedicate this puzzle to her. And thanks as always to my fellow constructors, especially Jacob (congratulations again on getting married!), and to you, the solver. Hope you enjoy this!

Wilson's FOURTEEN / POINTS played upon today. Can you imagine a president trying to lay out fourteen points today, the first worded as:

I. Open covenants of peace, openly arrived at, after which there shall be no private international understandings of any kind but diplomacy shall proceed always frankly and in the public view.

People would get about three words in before they'd tweet #TLDR.

The idea works for a Thursday, 14 grid entries missing their final POINT. PIN is really PIN POINT, NEEDLE is NEEDLE POINT, etc. I liked that Alex chose mostly ones that made no sense until adding the POINT. For example, WEST is much better than PLOT, since WEST is clearly not a school, while both PLOT and PLOT POINT can fit the story-related clue.

Laying out 14 themers — actually, 16, including FOURTEEN POINTS — is no joke. I started to highlight them to better help them stand out, but the grid began to look fugly.

Note how many themer intersections there are — PIN / NEEDLE, BROWNIE / FOURTEEN, TIPPING / STAND, etc. — as well as pairs in close proximity. It's a nightmare of a gridding task.

I was pleasantly surprised to get some goodies in the fill; GUT PUNCH a standout, but also DEAR GOD and LAY ODDS, CRECHE, LOUVRE. As a constructor, I think that's pretty good.

As a solver, though, I felt like there wasn't enough snazziness in my solving experience. With barely any long themers, there ought to be more juicy long bonuses to keep me interested. Such is the challenge of a daunting theme construction.

I'd have preferred something less audacious, maybe SEVEN POINTS as [Score for a touchdown + extra point], allowing for more breathing room in the grid. Also allowing for a smoother product, not needing gluey bits like ANE ASTA CLU ORNE (yuck!) SWAGS (?), SCI THU, etc. to hold everything together.

I'd have even been okay with FIVE POINTS — the old Manhattan neighborhood featured in "Gangs of New York" — which might have allowed for longer POINT examples, like EXCLAMATION, INFLECTION, VANISHING, etc.

Maybe that's beside the.