Tim Croce author page

Usually, when I start with a formation of black squares, it's not particularly close to what I end up with. But this is an exception. I usually like to start with stacks of 8, 9, or 10 and intersect them with stacks of 5 or 6, because that's what usually works. Whenever you intersect stacks of 7 or more with other stacks of 7 or more, that's when it gets challenging.

Did you find yourself solving five mini-puzzles instead of one puzzle? I hope not too much. This puzzle is a bit compartmentalized — there are only two ways into each corner. So I tried to make enough entries at the entry points into the corners so that (a) solvers wouldn't get stuck from entering a corner by two unfamiliar entries, and (b) the entries would be recognizable from the letters in the middle. That is, if you were working from the middle out, or from one corner to the middle then into another corner, you wouldn't (I hope) get shut out of a corner due to unfamiliarity.

A constructor's personality, it's said, comes through in their grids. One may surmise that, with two video game references, one of which being a seed entry, I am a big gamer — but I've never owned a video game console. Nor have I or would I ever use BARQ'S or anything like it. However, I did include 60-Across (PGA TOUR) and 1-Down (PAR FIVE) and did clue 4-Down (FOOTE) with two of my favorite things in mind.

By the by, for 52-Across... my original clue ("One starting a movement?") would definitely be better served for my website than for the Times. I appreciate that Will wanted to keep the clue at least a little palatable and — pun intended — didn't want to push it.

Yes, it's another one of those 7x7 freestyles. I construct 7x7s when I feel like a challenge; the shorter the entry, the lower number of "fresh" or "exciting" possibilities. The reason it's harder to build 7x7s is that not only do you have to find as many exciting 7s as possible, but you have to intersect them, and you have to do it in all four corners. With six-letter entries and shorter, you can get away with having any entry, as long as it's not ugly; I feel like 7 and above is where you have to start finding great entries to fill the space lest it feel like a "wasted opportunity".

When I take on a puzzle of this shape, what I try to do is to load up the grid with as many "seed" entries as possible and see how many of those I can keep and how many I have to sacrifice during the fill. (In puzzles with 8, 9, 10, 11, or 15-stacks, I can get away with three or two seeds, or even one.) This is why I was very fortunate in building this puzzle: I don't remember ever being able to keep this many seeds that I started with. I started with 1A (CARLS JR), 8D, 20A, 33A, 24D, and 38A (among some others I don't remember that I had to throw away). I didn't even care that 8D (HOOK ME UP) and 20A (ME LIKEY) shared a word.

I'm very happy that Will kept my clue for 63-Across (REC ROOM); I had the toughest time out of all the entries thinking up a good clue for it. I'm glad that effort wasn't for naught.

Given recent events (that is, since I constructed the puzzle), I now consider 31A (ASSAD) off-limits when constructing a puzzle, up there with HITLER or IDI. I put it in then because it fit; now, I'll reconstruct a section completely rather than include it.

You'll notice that I included two pairs of "cheater squares". I can sorta see the appeal of not having them — the aesthetic value of the grid is reduced, I guess — but I never have qualms about using them. I value having that great entry without the expense of a bad one, or having a passable entry without the expense of a REALLY bad one, over pure aesthetic appeal. I understand that aesthetic appeal has value occasionally; but, given the choice, I favor content over look.

I did this one the (sort-of) good old-fashioned way — by hand. I say "sort of" because I didn't quite use pencil and graph paper, but Microsoft Excel. I think I might have singlehandedly crashed onelook.com at one point during the construction process, if not worn out the "?" key on my keyboard. This had been one of the odder crossword goals of mine, to get a freestyle hand-constructed puzzle published. I'm glad I did it, because I have gained a completely new respect for those folks who did this before computers were a thing … and for the few of those who still don't use them. It's hard to keep away from crosswordese when doing it this way.

I'd be pointing out the obvious by saying that the whole thing started with 1-Across; when you're building these things manually, you really have to have a definitive seed answer at a definitive position. Happily, Will loved the answer I put there. It progressed (slowly) from NW to NE, through the center to the SW, and then to the SE.

As the hand-constructing process goes, the black square arrangement went through myriad changes to make this thing work; I wasn't even originally planning on having a grid-spanning entry in the middle, but it just kinda worked out that way after building the NE. The SE was the painstaking part — I had to do it twice. There was one particular entry (BADASS) that I thought may have been OK for the Times, but Will didn't quite feel the same way. (There were a few other iffy entries in that corner that pushed it over the edge, too.) So, true to form, I revised it the same way (it took another four weeks, on and off) until I got it into its final form.

(Shameless plug: I'm gonna be doing my own crosswordy thing at least once a week. It's like I've given myself my own playground now!)

This is the result of collaboration between members of two parties long thought to be incapable of mutual agreement. Those parties are, of course, Red Sox Nation and Yankees Universe.

The meeting was by chance: we had puzzles published back-to-back (11/10/11 and 11/11/11), and we were thus able to see each other's email addresses. You must know that I (Tim) am very shy by nature, so the thought had never entered my mind to contact anyone to collaborate with. Despite that, we got in touch over nine months later... I (Alex) wrote to Tim on August 31, 2012, seeing as he was the constructor for the next day, the day of my mother's second wedding, at which I was best man.

In summer 2012, I (Alex) came up with the combination of TAPPAN ZEE BRIDGE and LIBATION BEARERS in rows 3 and 4 and CUTLETS and LIP BALM (the latter not yet having been used in the NYT back then) crossing them. At one point, Tim and I tried 4-5-4 across the top and bottom two rows, but sticking with 3-7-3 made it much easier to work with those 7's. We agreed on either SEMIGLOSS PAINTS or SEMISOFT CHEESES for the third span. The top left and middle fell into place because there were few other options for those sections; once we discovered that –SES presented more options than –NTS, CHEESES it was. There was some debate over the fill in the NE, but, once we found ORESTES, a tie-in to 20A, that was it.

The bottom was harder. The age-old debate arose: pangram or smoother fill? We tried a number of stacks: STREAK FREE SHINE / IT'S THE LATEST FAD, LEAVES A BAD TASTE / INCOME TAX RETURN, SPACE SATELLITES / AEROBIC EXERCISE — here we abandoned Scrabbly fill — FAST EDDIE FELSON / IMPERIAL MEASURE, then this one. There was some hang-up over SPANG, but we thought the fill in the bottom center to be much better with than without it. And I (Alex) feel fortunate to have still been able to use one Scrabbly letter in the bottom half, as I thought the Z in TAPPAN ZEE BRIDGE might feel lonely with no other Scrabbly letters in the grid.

As for cluing, we agreed to both write full sets and chat over Gmail, going clue by clue and deciding which one was better. We found that to be better than splitting the clues in half, because deciding between two clues for each entry would make a better product. Of particular note is the one for ELBE, not being able to decide between referring to the Vltava and the border between West and East Germany (which the Elbe formed part of) and thus putting both in our manuscript.

Thanks, Will, for the terrific job you've done editing this and all our puzzles. And thank you, reader, for enjoying our puzzle!

Once the inspiration came for this theme (no, I wasn't at the zoo when the idea hit), I knew that this was the kind of theme that was just begging for interlock. I honestly wasn't planning for this much interlock, though. My original plan was to have the two vertical theme answers intersect the A two columns closer to the center than what you see here, but I decided for the heck of it to see if I could wedge a seventh in there. (Well, "wedge" is the wrong term, I hope... that implies it fits where it shouldn't have. I suppose you all will be the judges of that.) It was a happy accident that it turned out to have a themeless word count, because the interlock forced my hand here; I needed to keep the sevens in the top left and the sixes in the top right (the unsightly clump of five blocks on the top and bottom were necessary too).

It was also a happy accident that I was able to include 1A and 16A in the same stack; I'd like to say that I meant to do that. With the multitude of 3s and 4s that resulted, my concentration was on the sixes and sevens. I spent more time than usual trying to get as much good fill in the top left and lower right as I could.

Now, this may come as a surprise, but I rather like all the colorful ways in which the English language allows its speakers to express themselves. This puzzle, to me, is a prime example of not only the diverse nature of English, but also the old truism that language is a reflection of culture. Namely, in the case of this puzzle, humanity has seen nonsense in everyday life so often that they have to keep making up words for it to keep from being bored of describing it.

I know it seems that the fill is a little constrained because of the sheer amount of theme content, but I put a lot of constraints on myself from the outset and, I think, minimized the junk given the number of theme squares. I moved the various theme answers around a lot — a LOT — and tried a few other words that just didn't make the cut (TRASH, RUBBISH, GARBAGE, BOSH being some of them) because of the effect they had on the fill. (8 pairs of theme answers mean a lot of permutations!) The only words that I didn't move around much were JIVE and JAZZ, because terminal J's and V's severely limit the options in a grid where my options were already considerably limited. Those two stayed where they were from the start of construction.

On the whole, I was so enamored of all these synonyms for "nonsense" that I felt like I had to wedge as many as I could into this puzzle ... even at the expense of going over the max word count significantly. It was a bit of a gamble to do that, but I'm glad that Will didn't see past all that nonsense.

By the time this reaches you, I hope all who celebrate have had a Merry Christmas.

It's unusual to see theme answers arranged like the second and fourth ones in this puzzle — with a "cheater"* and not spanning the rest of the grid. However, I had to do that. I tried not having the cheaters and using the 10's and 15's in all four combinations (I didn't want the revealer in the middle; I didn't want the top and bottom 15's only one row apart from the 10's), and found that the fill went much smoother with this arrangement. (That's even with IERI, which I definitely didn't want, but had to use.)

Two things about the low word count for a themed puzzle. One, I confess that I tried to go lower than that (tried making 9D or 10D 7 letters, tried making 3D 8 letters... but not at the same time!). Two, I didn't want to clutter the top with too many blocks, considering that I had to use a lot of them in the middle to make the themers work. Once I found that 6/8 worked in the top two rows, I didn't hesitate to use it. (Plus, I'm a themeless guy at heart, as you may be able to tell.)

About the theme answers — I didn't set out to make this kind of theme (I know we've seen these themes before), but I was fortunate to have my crossword sense (like "Spidey-sense", but a lot geekier) activated when hearing this phrase that would become the revealer. I thought that finding a mixed-up seven-letter sequence in crossworthy phrases would be harder, but I still had to think a bit to get the symmetrical pairs. I lucked upon the 15s, honestly — the 10s were much easier.

*Cheater = black square which doesn't affect the total word count. Today's has four cheaters, the short ends of the four L blocks of black squares.

The inspiration for this puzzle was simple: I wanted to take a common (timeworn, in this case) theme, the "double talk" theme, and give it some variation. I didn't know how exactly I'd do it until I was doing a crossword (don't remember which) and saw one of those infamous "When repeated" clues. I didn't want to do it that way, because I didn't want a million clues to begin with "When repeated". I came up with this way instead because I wanted to give the early-week solver just the briefest bit of pause as to how to parse the answer. I wanted to give the solver at least a small "aha moment", however slight it would be.

The construction of this puzzle was a prime example of an exercise in balance. How many theme entries could I fit until I started to feel like I was "wedging" them in, until the sacrifices in the fill and the grid structure became too great for the amount of theme material? Truthfully, when I started, I didn't expect to get sixteen. Believe it or not, after I had the sixteen in place, I actually tried to fit another two at 32A/43A, but that's where the "balance" principle kicked in. I had already put in a pair of cheater squares to get this far; that was enough of a compromise. With the addition of another pair of theme answers, I would've lost too much in fill quality for it to be worth it.

So how did I do it? Well, I started with some graph paper, several mechanical pencils, a LOT of erasers, and a caffeine IV...

...just kidding. I used pen.

No, honestly, I didn't set out to make a sub-60 grid. I went for an eye-catching grid pattern, and it happened to come out at 58 words. This grid started life as a 66-worder with the stair-step block pattern extended out diagonally one more. It was on a whim that I took out those four blocks to see if I could fill it.

I started with a power-fill to see if it'd work, not expecting much. I knew there'd be compromises (47/49A, 4D in the final product). Not only did it find a fill, but it came in only a couple minutes, and the center actually came out like this the first time I filled. So I had my framework (26-29-30-31-35A/14-17-20-23-27D) MUCH quicker than I expected. I don't know if this was the easiest sub-60 pattern to fill, but I've certainly not found any "easier" ones.

From there, the northeast most frustrated me. At first, I didn't like 8D because it seemed awkward. I was this close to scrapping the whole framework and restarting. But I searched Google and found it to be legit. With that, I found my fill and came upon 6D, which was, as of then, unused in the NYT (to my surprise). As a college football fan, I'm happy to see it go!