Thumbnails

Happy 2014 indeed! Good way to kick off the new year. Pete uses the digits 2 0 1 4 in the horizontal direction, and homonyms of them in the vertical. At first I was stuck around that area, but it sure provided a nice moment when things clicked. The only pause I had was "oh" didn't seem right for "zero" (sorry, Pete!), but that might be a dialect or usage issue, so we'll let that pass.

Pete wasn't kidding — this is a very tough construction. There are four parallel downs which must work together to form coherent across entries, plus Pete chose to go with a 74-word puzzle. The acrosses around HAPPY 2014 are nice and smooth, with just an IANS as a blip; very impressive. I didn't know WATUTSI but was glad to look it up. And the NW was beautiful! All sorts of fancy entries up there, EGBERT actually being my favorite. I can overlook an ENV and and ESO to get NIRVANA and GEISHA.

Some issues arose in the SW corner, though. The high constraints in the middle propagate through the puzzle, and since that SW corner has a lot of white space, it becomes even tougher to fill. CAVER made me pause (I used to cave, and we called it caving, but never called ourselves cavers), and it crosses the long partial ARE WE. Our old crossword frenemy ETUI also makes an appearance, along with the outdated ARTE Johnson. It's really tough after WATUTSI is fixed into place, which constrains the number of fill options and forces some compromises.

Nice theme concept, fun solve, with a bit of crunchiness.

– puzzle by Peter A. Collins

Man oh man, it took me forever to cotton to this very innovative theme. Five theme answers are quasi-literal interpretations of phrases where the first word ends in ENDING. For example, the first theme answer is VENDING MACHINES, written out as MACHINES V (machine with V-ending). Tough, huh?

The hardest part was finishing the middle, where I didn't expect there to be a fifth theme answer (MENDING WALL written as a WALL with a M-ending). Will has said in the past that he prefers to have the theme answers stand out in some way, usually by making them the longest answers in the puzzle (or at least the longest answers in the across direction). That would have helped me greatly here, as I found it very confusing to have a shorty smack dab in the middle. But perhaps a little changeroo is good every now and then to keep everyone on their toes.

Note the wide-open nature of the grid. At just 70 words, it could qualify as a themeless puzzle. But any time you work with themeless-level word counts plus four (or more) theme answers, there are bound to be trade-offs. Todd has some nice long fill in here: LOVE CHILD and MATTRESS with its fun clue are really nice, but overall, there were enough compromises to affect the quality of my solve. Seeing too much of AMATIVE, UNLOOSES, GROFE, ARACE, etc. took me out of the experience somewhat.

I do wish there had been more wordplay type of cluing to help keep the solve a little more playful, especially as I was banging my head against the (non-mending) wall, attempting to figure out what the heck was going on. Overall though, I applaud the decision to push the boundaries of construction.

– puzzle by Todd Gross

Another fun solve from two of the themeless masters. This one played extremely hard for me, more like a Saturday, given the opaque cluing. A real brain-buster of a workout. I think it's more typical to see WYCH ELM and NICOLAI Ghiaurov on a Saturday, as they're both difficult pieces of trivia.

I really enjoy misdirectional or wordplay type cluing, so FOREVER STAMP was both my favorite entry and clue in the puzzle. It had me thinking about public utilities, then apartment rent caps, and the real answer took forever to make itself known. Great stuff.

Being on the TWITTERVERSE (follow me at @JeffChenWrites if you don't mind inappropriate/insane/inane observations) I loved the entry, but it would have been great to get a less direct clue. (Perhaps TWITTERVERSE isn't well-known amongst different generations to do that, though.) And it's a pretty cool feat to have EXPEDIA right next door, as well as an up-to-date clue for SEPIA involving Instagram. I liked the modern vibe.

Finally, I've been thinking more and more about Ian Livengood's comment regarding seeding a themeless with tiny words (three and four letter entries) that have easy letter patterns, and seeing what you can get for the long entries after that. I really enjoyed today's puzzle, but as a constructor, it's difficult to overlook entries like ERI, OFT, SYL, THA, NAE. Tyler Hinman, a former ACPT champ and excellent constructor, has said there's no such thing as an excellent themeless with a few ugly entries. I wouldn't quite go that far, but I think he has a point. Anyway, just one guy's musings. Enjoyable Friday.

– puzzle by Barry C. Silk and Brad Wilber

Very fun puzzle today, aside from my willful ignorance of "Jersey Shore". I couldn't finish the NE corner of the puzzle and am proud of it. Don't tread on me, DJ PAULY D!

What a beautiful SW corner. I loved the crossing of STONEWALL and PATDOWN with its fun clue, "Touching scene at an airport?". Brings back hilariously unpleasant memories of going through security at a Netherlands airport back when I had my devil-length goatee. Good times. And I find CODON to be a great entry, too infrequently used given its crossword-friendly alternation of common consonants and vowels. I couldn't totally recall what a CODON was, but it came back easily enough with four letters in place. Okay fine, with five letters in place!

Overall, I really appreciated the smoothness of (most of) the grid. The NE corner was the roughest bit, with ATH and the awkward LEND A partial. If I had paid the slightest bit of attention to pop culture, I might have rolled through it. I wonder if DJ PAULY D will persist in American culture? Get on that; make yourself honestly famous, mister! Or is Pauly D a missus? Wow, am I out of touch with "popular" celebs.

Nice work today from Kyle. There were a few bits I didn't care for, PREFIGURE in particular, but I really liked all the snappy short fill accompanying the marquee answers: JET SET, PACK RAT, RUB OUT, BIG TOP all added snazz to the puzzle.

– puzzle by Kyle Dolan

One of the aspects I love most about the NYT crossword is the constant desire to stretch the limits of the art form; to create something never seen before. I enjoy xws from many different sources, but what other daily xw pushes the boundaries like this? Neat idea to incorporate the game "Clue" today. It's been made into a movie, why not a crossword as well?

Interesting layout: four "rooms", three of which give clues to the fourth one, which reveals the SUSPECT, the ROOM, and the WEAPON. Each of the three "rooms" give hints in a different way, i.e. the first room contains three words which can be preceded by the word SCARLET. Ah, our old crossword friend the SCARLET TANAGER makes itself useful in a fun way, giving an insider's nod to all the times it's been (over)used in triple-stack creations due to its very friendly crossword nature of common letters and mostly alternating consonants and vowels.

I enjoyed the solve, but I'm not sold on the total segregation of the four rooms. Small diagonal "doors" three blocks long could have been a fun way to keep the sense of a house layout while maintaining the usual crossword rule of "no isolated sections". It could also have made for a natural progression, moving from one room to the next, gathering clues as you went. It also would have been really cool to have multiple possible suspects, rooms, and/or weapons, using a Schrödinger-type puzzle. Not sure how that would have gone over though; perhaps too tricky for many.

I was amazed to see that this was Alan's first construction for the NYT (though third to be published). The quality of fill is not at all what I would expect from a beginner, with such long goodness as PTERODACTYL, DIET SODA, ABANDON SHiP, ALIEN ATTACK, etc. Yes, there are instances of the dreaded (and not really a thing) ENOTES and the old-timey ISTLE in close proximity to ARACE, but I'm willing to wager that Alan will work hard to avoid these types of entries in the future. Themeless grids are hard to fill, and this xw has effectively four themeless-style mini-puzzles. Good job of filling them with some nice entries and a minimum of ugly stuff; looking forward to more from Alan.

– puzzle by Alan DerKazarian

Really nice to have a Monday with an additional layer of complexity. During my solve I couldn't tell what was going on, and even after I filled in ROMAN, still wasn't sure. WARNING: SLEDGEHAMMER ALERT A pretty cool moment to realize MMXIV was a little shout-out to the new year, 2014. Nice concept!

As always, C.C. does a nice job of choosing her theme answers. They're all strong phrases, and I appreciated the nod to Sue Grafton and her very useful "_ IS FOR _" series, which has been either a godsend or a bane, depending on how you see the A IS, B IS partials. I personally consider them more evil than tofurky sausage (don't ask), but I can see how some solvers would appreciate having a gimme to start them off, especially with harder puzzles.

I like C.C.'s layout too. She allows for four long downs, all nice entries, although I could see how people might react negatively to KIRK GIBSON (if you aren't into baseball, or if you're a diehard A's fan STUPID KIRK GIBSON AND THE SLEDGEHAMMER WAYS YOU RUIN PEOPLE'S DREAMS). Ahem. YOGA MATS and THE ALAMO are very nice.

I considered this puzzle for the POW! because it's innovative and well-constructed, but my personal preference for ultra-smooth Mondays made me hesitate. ELOI is on my personal Do Not Call list (although it's gotten me out of jams in the past), and the smattering of ANIM (awkward abbr), our old xw friend ISAO Aoki, along with a SCH EER ONCLE felt like a touch too much. I love GOKART as an entry, but ONCLE and TNUT feel like a high price to pay on a Monday.

Starting the daily constructor comments has been an incredible privilege, and seeing the human side to puzzles has made me appreciate puzzles even more. Hearing about C.C.'s troubles with the SE made me want to investigate further. And sure enough, once you lock in the last theme answer and THE ALAMO and decide you're not going to use cheater squares, that little chunk becomes tough to fill cleanly.

I realize I'm extremely picky about smoothness on Mondays, and many constructors/solvers don't share my values, so I'll stop there. Overall, I really enjoyed today's offering EXCEPT CURSE YOU KIRK GIBSON.

– puzzle by Zhouqin Burnikel

Ah! It took me a while to figure out what the theme really meant, and it was a nice moment when I realized each phrase started with a slang synonym for "good". I like the overall concept and the huge theme density, although the clue for the revealer was so long I didn't bother reading it. I wonder if the impact would have been stronger if that had been concatenated. Or if the revealer had simply been GOOD?

Puzzles, especially early-week ones, often don't need a revealer to explain the theme. But ones like this, where the theme isn't overtly clear, definitely benefit. IS THIS GOOD feels a touch arbitrary to me however, not really a lexical chunk to my ear. I sometimes pull sniff-test duty in my household, and I think OH GOD THIS IS BURNING MY NOSE is more accurate. Seriously though, IS THIS STILL ALL RIGHT or IS THIS BAD feels more correct to me.

I'm fully agreed with Will that IT RIP is likely the worst piece of fill today. I'm also not a fan of several other entries though, especially A SCAR, OOOLA, and OLIO. A SCAR bugs me because I have a personal preference against five-letter partials (see IT RIP). OLIO is on my personal Do Not Call list, just like ELOI yesterday, because it's 1.) seen heavily in xws but not in real life and 2.) composed of common letters. None of these entries are BAD per se, but they're not BAD. Er, GOOD.

I do appreciate the effort to incorporate more long fill (DRESS SIZE and INCUBATES are both nice) but can't help wondering if it would have been worth it to break up ENDWISE and VIRTUAL to achieve a smoother solve. Any time you leave a five-letter word space (NTEST and WAHOO) right in between two themers, fill is going to be rough around the edges, because of all the parallel constraints you impose on the grid. VIRTUAL is a pretty good answer, but sacrificing it to get rid of some of the shorter ugly stuff would have been sick.

– puzzle by Mark Bickham

What a nice construction today. This type of grid arrangement, with an 11-letter entry in the middle, can be difficult to fill because that central entry (sort of) divides the puzzle in two, creating big corners of white space to fill. But Sam and Vic do an admirable job in their fill.

Aside from the main theme answers (which lead to SPONGEBOB SQUAREPANTS), there is SO much goodness all throughout the grid. Any big corner like the SW or NE is bound to have drab and or unpleasant fill, but look what's packed in the SW: NEW ISSUE (I have an MBA* so this totally appeals to me), IRKSOME, NOSE BAG all in parallel, with ORWELL crossing it. There are a few compromises, but ESSA and ASSN are pretty minor offenders.

In the NE, RAN OUT ON and SIGN FOR are very good. And call me crazy, but I don't agree when people gripe about ETAIL (or ETAILER). It's a legitimate term, seen frequently in business. Aside from that can of worms, everything is relatively smooth and clean. Nice work.

I wasn't a huge fan of the theme itself, as we've seen SPONGEBOB a few times recently in the NYT xw, giving him a bit of an overused feel. I am glad they changed FANCY PANTS to BOSSYPANTS, since theme consistency is preserved (THINGAMABOB and BOSSYPANTS are both one word, while BATH SPONGE and TIMES SQUARE are two).

All in all, not a standout theme, but a very nice job of making a snazzy and enjoyable puzzle.

*Mostly worthless but two very fun years of debauchery

– puzzle by Sam Ezersky and Victor Fleming

The day to day administration of XWord Info has affected my solving experience in strange and unexpected ways. One of the aspects that's changed the most is regarding "trick" puzzles. Jim and I want to keep the database as clean and useful as possible (we fix up all the screwy answers), so every time we get something like today's, I either appreciate the trick even more as I perform the fix-ups, or grumble as I wade through C# coding syntax. Ah, the highs and lows.

I liked the concept today. I generally try not to link to old puzzles (because I don't need to remind people that most everything has been done in some way or another) but this one so heavily reminded me of Andrea Carla Michaels's very first NYT publication, taken to a new level of difficulty. I was a little frustrated as I was solving, but the trick seemed pretty neat once I cottoned to it (the entire right half of the puzzle "slipped" down one row). And who doesn't like seeing ATMSBROAD in the puzzle? Don't answer that. And don't worry, we fixed up that entry in the database.

I would have loved perhaps one or two more theme answers. It's pretty neat that the SAN ANDREAS FAULT is a "site of slippage" and BANANA PEEL / PATCH OF ICE fit the theme (and they intersect SAN ANDREAS FAULT!), but it felt slightly thin to me. It could have been really fun to open it up to other meanings of "slip", like giving someone the slip or a slip of the tongue. Even one more theme entry would have been enough for me.

There's some really nice stuff in the fill, especially PEACH FUZZ and MACH ONE. SUBDIVIDE isn't that snazzy in itself, but the clue did a great job in making it a strong entry. Overall though, with just three them entries, I would have liked more long, snappy fill. Yes, the theme entries intersecting makes it harder to work in good fill, but not that much harder. I would have liked to have seen more 7+ letter fill, even if that meant having more 3-letter fill in exchange. There's so much in the 4-, 5- and 6-letter range today, and any fill under seven letters is hard to make memorable. (You can press the "Analyze" button at the bottom of the page to get the exact distribution.)

Overall, nice concept with some unfulfilled potential.

– puzzle by Caleb Emmons

★ Patrick Berry earns his second POW! with this beaut. Why do I have a feeling that I'll be handing out many more POWs to him in the future? Given his propensity to make innovative grids incorporating very nice fill and little to no dreck, I'm looking forward to it. And when POWs rise to be worth one beeeellion lek (I'll pay everyone when we meet up in Albania), he'll be laughing all the way to the bank.

At XWord Info we have a page displaying common grid patterns. Scrolling down, you'll see that several of them are of the themeless variety. It's totally fine for a constructor to take a grid from a grid library and fill it, but there's an artfulness, a higher skill to crafting a grid from scratch. That might sound hoity-toity, but the big advantage it brings to the table is the ability to fit in exactly the lengths of entries you desire. Not only that, but you can adjust block placement as you construct, allowing for snappier or smooth fill. Often times I find moving a few blocks here or there solves a lot of problems.

There's something awesome about hugely wide-open grids with patterns I've never seen before. I had a moment of awe with my first glance of the giant tilted football in the middle of the grid. You might ask, why don't more constructors attempt grids like this? Having tried it myself, I can answer that: because it's insanely hard to do with beautiful execution. There are so many pitfalls, so many times when you work yourself into an ugly entry and have to reboot ad nauseam.

As with every puzzle, not everything came up roses. I wasn't a fan of THENCEFORTH, which at the best felt like a "glue" entry to hold things together, and MACHOS feels iffy. Also, today's Scrabble score is pretty low, just counting one V and a smattering of Ks. But that's all a very small price to pay for the amazing nature of today's grid.

Yadda yadda yadda great entries like CRACK SHOT, CHEAT SHEETS, SCRATCH PADS, COPACABANA, BATH PILLOW (all in the center section!) blah blah blah nary a bad entry etc etc etc wonderful clues like "Cribs" for CHEAT SHEETS and "One who wants in on the deal" for CARD PLAYER. Seriously though, I joke to assuage my grief that I'll likely never be as good as Patrick. But that's okay; being around greatness inspires me to work harder. I'll be studying this one further to see what I can learn.

– puzzle by Patrick Berry

I don't know how MAS does it. I've tried my hand at quad-stacks in order to expand my construction skills, but I'm nowhere even close yet. Coming up with the four stacked grid-spanners is hard enough, but throw in the fact that you need the section to tie to something else in order to make it a full puzzle (and have everything be symmetrical!), and you have a real feat on your hands. There's a reason why very few people are making grids like this.

There's not much room for snazzy fill outside of the 15s, so they must be snappy in order to give the puzzle kick. MAS does a pretty good job here, OBSCENE GESTURES being my favorite (no surprise given my fourth-grade maturity level). Some people might disparage TRADITIONAL IRAS, but financial and investment planning is a hobby of mine, so I really liked seeing it there. DENTAL ASSISTANT isn't great in itself, but a nice wordplay clue makes it shine.

Let's look at just one interlock issue as a demonstration of how difficult this construction is. In the bottom stack we have *ESHA in the fifth column. WAUKESHA isn't a great answer (unless you're from Wisconsin and/or love cheese curds, yum!) but what else is going to go there? Sure, you could shorten the entry to MOESHA or AYESHA, but that means you'd have to do the same in the symmetrical spot (GLEASONS might become GLEANS or GLEAMS). And then what happens to the surrounding areas? Maybe you could get something to work where KERT and where SOSPAD are, but with all the other constraints locking you in, that's highly doubtful.

Sure would be nice to do something about the KERT/MENSAL/REINA subsection though.

Yes, there is a good amount of subpar fill in here, but that's typically the price to pay for an audacious construction like this. Solvers' preferences are all over the map, some loving this sort of jaw-dropping grid, some preferring smooth and silky but less ambitious. I wouldn't want every themeless to push the boundaries like this, but I appreciate both the variety and the feat of construction. Fun solve.

– puzzle by Martin Ashwood-Smith

Before I started this gig last year, doing the NYT xw was fun in itself. But I always wanted more, specifically to hear Will's thought process regarding why he chose each puzzle and how he changed it to his liking. There have been many highlights in the past six months, but having Will write and send daily comments on the puzzles is among the top.

One observation that's been eye-opening is the diversity within his target audience. Before we started communicating, I always thought Will strove for the jaw-dropping, mind-blowing puzzles and filled in around them with "ordinary" stuff because he just couldn't get enough of the former. But now I've come to see that a significant portion (probably a large majority, actually) of his constituency shies away from anything too new, being very appreciative of puzzles that are familiar or push the envelope just slightly.

All of this is a long way of saying that even though today's puzzle doesn't blaze entirely new ground, it probably meets the needs of more solvers than the rule-breaking, game-changing puzzles. We've seen many puzzles before where theme answers use only one vowel, but it's a nice addition to have all the clues use only that same vowel. Some of them get a little tortured for my taste (the clue for RASTA MAN brought back unpleasant memories of Jar Jar Binks*) but overall the extra layer was appreciated.

A nice debut. There are definitely some compromises because of the high theme density and the way the themers interlock (the SO I/ON UP/ID DO section is so heavily constrained by the placement of the theme answers, for example), but also some nice long fill (RAW ONION, PYROMANIA, LAB ANIMAL are all great stuff). To start a construction career on a 21x is no small feat. Looking forward to more from Andrew.

*meesa still not forgiven you, Georgie Lucas

– puzzle by Andrew Chaikin

★ An absolutely beauty of a Monday puzzle. It has everything I personally look for: 1.) innovative theme, 2.) spicy long fill and 3.) fill accessible to the novice (little to no ugly stuff). There's more good stuff coming up this week, but Monday constructions are perhaps the hardest type to pull off well, and Lynn's nailed it.

The theme isn't necessarily ground-breaking, but it contains an added level of complexity which helps makes it more interesting for the experienced solver. We see a lot of puzzles where a certain word is hidden within theme phrases, but having two words hidden helps to distinguish it.

And with five long theme entries, I'd expect to see a lack of good long fill and a lot of short ugly stuff. Lynn has done something unusual, shoving her top two entries into one region of the grid (instead of spreading them out, one to the left and one to the right). That usually creates difficulties in filling such a dense area, but she makes it look easy. Having a five-letter word sandwiched by theme entries (VILLA between PHONE CALL and RHONE VALLEY, and TOMEI between STONEWALLED and ONE AND ALL) often will result in bad fill due to the severe constraints, but she uses a set of cheaters in the very NE and SW to smooth things out. I fully approve — as Patrick Berry has said many a time, he'd always choose to use cheaters if it results in better fill.

The grid arrangement allows Lynn huge flexibility in the NW and SE corners (the theme answers barely constrain her at all in these regions), and look at all the goodness she's packed in there. ON THE GO / FOR A SONG / FLEW SOLO all in parallel, with only ESS as a (very minor) blight? Yes, please! I didn't notice any really bad fill as I solved, nor did I notice anything even with a careful scan afterward. A stonker of a job.

Uber-professional work, a total pleasure to solve; a near-perfect Monday puzzle. Beautiful start to the week.

– puzzle by Lynn Lempel

Perhaps the hardest Tuesday NYT I've ever done. It's pretty neat, not at all BUSHWA (my favorite themer by a mile!) to see 16 different synonyms for "Nonsense" — I had no idea there were so many of them, all colorful — but the opaque cluing sure made for a difficult solve. The 16 essentially unclued entries made it almost like solving the puzzle with just down clues, an exercise some of the speedsters sometimes engage in.

It's rare that Will allows a higher word count than the maximum of 78 (for a 15x), but there are definitely examples. Something like today's puzzle definitely pushes the envelope of construction, almost necessitating the push to 82 words.

It's also unusual to have so many unsavory short answers, but today we see ANIGH, DERAT, ROW B, AT YA, STELA, etc. With so much theme density, it's almost impossible to avoid these. I've highlighted all the themers so you can better appreciate how much their placement constrained the construction. Consider the east section, where TRIPE, BUSHWA, and HOGWASH sit, just as one example. There are limited options for where STRUNG sits, and once you place that word, that east region becomes very difficult to cleanly fill. Perhaps one of ANIGH and STELA would be fine for an early-week puzzle, but it's a bit unappetizing to have them in such close proximity.

I personally would have preferred slightly less theme density in order to clean up the fill, but I'm sure there will be solvers arguing that their favorite "Nonsense" phrase was left out. Overall, I always appreciate seeing someone push the boundaries, and perhaps this will stimulate other ideas for future disestablishmentarianism. Plus, BUSHWA!

– puzzle by Tim Croce

Bernice's 100th NYT crossword in our database! How appropriate for her recent 100th birthday, yes? She actually has more than 140 NYT puzzles, a fact not reflected in our database since David Steinberg's Pre-Shortzian Puzzle Project is only back to 1979 so far. I'd love a chance to pick her brain when it comes to secrets of longevity, as I personally have a goal to set the world record for the 400m run in the 95 and older category (currently 2:17.45).

What, that seems like a weird goal? Don't get me started on all the other strange goals I have...

Anyway, nice Wednesday, consisting of four long answers ending with a homonym of a single letter (AFTERNOON T, for example) plus two bonus theme answers smack dab in the middle (ONLY U and WELL G!). Bernice uses the "pinwheel" construction (look at how the four longest answers are arranged) but adds a layer of difficulty with those two extra themers. Typically I think it's inelegant to have short theme answers hidden like this, because they're not the longest entries in the puzzle or don't have asterisks to distinguish them, but the theme concept was easy enough that I thought it was fair, even lending a positive "added bonus" feel. GEOGRAPHY B did throw me off since spelling bees seem to be much more common (or at least get more media coverage), but it's a fair term.

Nice job balancing the usual trade-off of snazziness vs. cleanness in the fill. I love that Bernice includes OMG (I'm very curious if she's ever used it), and SHOEBOX / LIPSYNCS / VOLTAIRE / CAPE HORN add a lot of zing to the puzzle. I did notice the COS / ON IN / OSIER type entries during my solve, though. Adding the two middle themers added constraints, so it's not a surprise to see a smattering of subpar entries.

It's great to see the diversity within the constructor community, each person bringing something different to the table. Here's to 100 more, Bernice! (both puzzles and years)

– puzzle by Bernice Gordon

What an enjoyable solve from Elizabeth today. Not a huge amount of overt theme material, so it felt more like a themeless as I was going. But what a nice surprise when I sat back to figure out what was really going on. At first I thought it was simply a O representing a dot over each I, but no, it's a full DOT. Very cool!

I personally tried a similar idea a year ago, but I just used Os to "dot" every I. I also wanted to pull off zero extraneous Is or Os (no dot-like Os floating around the grid), so during my solve today I was slightly put off by what felt like me to be superfluous Os — they seemed like orphaned "dots" that didn't belong to any Is. That's my own issue, though.

And the more I thought about my own (failed) experience, I liked the trade-off here. I still think there's a certain inelegance in having an O where there is no matching I (yes, I know, I can be super picky), but I really appreciated the cleanliness of Elizabeth's grid. And Will brings up a good point, that no I is left unDOTted.

That brings us to the revealer. DOTTED I is a brief touch, right to the point. I think DOTTED THE IS (11) or DOTTED ALL THE IS (14, maybe using a 14 x 16 grid) would have been more elegant an entry, but adding more theme material will almost always make the construction harder. Plus, a seven-letter entry like DOTTED I can go right in the middle without causing construction problems. Notice how HOE and REA (instead of banks of black squares) flank DOTTED I? That means the sides of the grid can be broken up in many, many different ways. A 9 or 11 or 13-letter revealer in the dead center makes construction more difficult, because it places a few black squares automatically

The fill is nice and smooth, although more marquee long fill would have been nice. Almost all the longest answers (all eight letters long) are single words, and aren't super snazzy. "Speaker's aid" helps to make ANECDOTE a better answer, and "Passed on, in a way" does the same for REPEATED — that sort of misdirection could have helped spice up the other eight-letter fill too.

And saving the best for last: "What some waiters never see?" as in "Waiting for GODOT"; that's absolutely brilliant cluing. More of that, please!

– puzzle by Elizabeth A. Long

A nice construction from one of the best in the business, Joon. Er, Jeff. Darn it, I mean Kevin!*.

One aspect I like best about Kevin's work is that he's always pushing the envelope, trying to figure out what he can do new that others haven't done before. Look at the NE and the SW today: beautiful chunky white sections, smoothly filled. Typically themeless constructors stay with triple-stacked areas, because moving to quad stacks exponentially increases the difficulty. It's definitely been done many times before, but check out how snazzy and clean he makes them. HOTEL BARS / ANIMAL FAT / AS I RECALL all crossed by BLACK OPS? Must have taken days, perhaps weeks to figure out that one subsection.

Usually going to 66 words in a themeless means you're going to have some compromises. But Kevin takes so much care in his work, leaving us with only ONE LS and KIEL as misdemeanors. And it was great to hear him say he struggled to see if he could eliminate KIEL somehow. But given the giant ell of white space up in that corner, it would be very difficult to do while still retaining BLACK OPS crossing all the long stacked answers. That NE corner has so many great answers though; they're easily worth the price of KIEL.

The one knock on today's puzzle might be that there isn't a marquee answer that jumped out at me with that hurts-so-good slap upside the head. I smiled when I finally pieced together DIME STORES — very cool clue — and I really liked the colloquial NO BOTHER, but I really appreciate getting at least one whiz-bang answer (usually themeless constructors place their best answer at 1-across).

Kevin took a stretch off of constructing for a while during his stint helping Obama get reelected among other pedestrian activities; very glad to see that he's in the kitchen again, mixing up more magic.

*I can joke about this, you can't. Ha.

– puzzle by Kevin G. Der

David continues his NYT crossword tear today. I'll set the over/under for how many he has in 2014 at... 14.5. Let the betting begin!

Slightly unusual today is David's use of triple-stacks of 11-letter entries. Typically most themeless constructors shy away from this approach, choosing to use triple-stacks of 8-10 letter entries, because 1.) it creates more three-letter words in the puzzle (USM / GEE / GAG in the NE, for example) and 2.) makes for that many more crossings to deal with. Eight parallel downs through a triple-stack is hard to work with, and 11 can be very hard to do cleanly. We see traces of that difficulty in ARNEL at the top and ITES / AJA at the bottom, but otherwise they're clean.

It's particularly interesting to me that while David's 11-letter stacks are nice and sparkly, I wasn't as big a fan of his stacks of 8s (in the NE and SW). After all, neither LAVATORY nor SEA ROUTE seem terribly exciting. And seeing IVAR / the stilted IT IS I / TYS ... what's going on, I wondered? It should be easier to fill that SW corner than one of the big NW or SE corners, shouldn't it?

I believe it has to do with where a constructor spends his/her black squares. With a maximum of 72 words allowed, the black squares must be used sparingly, and David does a great job of applying them so that the NW and SE are easier to cleanly fill. But that doesn't leave as many to facilitate filling the SW and NE corners. I wonder if the fact that both a 7-letter and a 8-letter answer (I LOVE LA / FACEPALM) cross the SW stack has something to do with the relative roughness. That's a lot of white space in one area.

Finally, I loved the clue for TAVERNS ("Round houses?") and TAXONOMISTS ("Professional organizers?") and would have liked a few more of those in my Saturday puzzle. Many of the entries today are great in the own right, but they're difficult to clue in a playful way (MEGAN FOX and BANANAGRAMS for example, you either know or you don't so the cluing must be made straightforward). It's a tough trade-off constructors must weigh: lean more on entries snappy in their own right but must have a straightforward clue, or ones that make for clever cluing? Anyway, a fun puzzle with some very nice entries.

– puzzle by David Steinberg

Quite a challenge for me today, as my knowledge of popular music is sorely lacking. Still though, a fun puzzle, "spoonerizing" famous(-ish) songs into wacky results. I laughed when I hit MAD BOONE RISING — brilliantly hilarious resulting phrase, plus the base song (Bad Moon Rising) is actually something I know. A perfect entry for this theme.

It can be quite a trick to build a theme around "oldies", as a chunk of your solving constituency might be left in the lurch. I felt a bit left out when I couldn't uncover SHES SO HIGH or HES SO SHY, same with I CITE THE WRONGS and I WRITE THE SONGS. Hard to feel deeply engaged when neither the base phrase nor the "wacky" ending phrase makes any sense. But might be solely my personal issue of being an oldies dunce; I could easily be in the minute minority here.

Now that I got that out of the way, I really appreciated Dan's construction. Once I saw those big open corners in the SW and NE, I braced myself for ugliness. Each one of those — a 6x7 chunk, for goodness sake! — is as difficult to fill as a subsection in a themeless. But both of those corners are absolutely beautiful, packed with nice stuff like PIEBALD and LETS EAT. Same goes for the west and east sections, where I cringed at first sight of the huge white spaces and their potential for disaster. But again, quality results, ENCODER reminding me of "A Christmas Story" and NON WORD with its great Colbert clue.

Bear with me as I delve into some numbers. Typically most 21x constructions lean heavily on three, four, and five-letter entries — they're typically necessary to hold grid regions together. Unfortunately, these shorties aren't usually "exciting", because most of them have been used so many times that solvers are used to seeing them. Debuting JUBA (the capital of South Sudan) is a rare example where a shorty is cool.

So the real genius of today's puzzle: Dan did a great job of laying out his grid, incorporating a lot of six- and seven-letter entries (a total of 49!). That's very hard to construct, since you're purposefully eschewing the gluey type short answers, forcing yourself to make grid sections connect through long answers. Very, very difficult, akin to building a network of bridges across the Grand Canyon. And look at the great result: SHOULDA, HARPOON, GI JANE, SCARAB, etc. all make for a chewy puzzle.

Great workout, the type I usually only get in themeless constructions.

– puzzle by Dan Schoenholz

Tribute puzzle to a great man. Surprising that we haven't gotten more MLK tributes on MLK Day. Maybe it has to do with the conspiracy put on by The Man, who also somehow makes good people of the world work on a such an important day.

*shaking fist at The Man*

Tribute puzzles are hard to do. I think they should celebrate the person by giving us a reminder of their achievements, but since it's a puzzle, it ought to entertain as well. My personal preference is to avoid tribute puzzles unless there's a way to incorporate an extra element, something that could only be done in a crossword puzzle. For example, I did a Roger Ebert tribute last year for the LA Times, with black squares looking like two thumbs up on the sides of the puzzle. That sort of thing.

Ouch, I hurt my shoulder from all that patting myself on the back, so onto the puzzle. I absolutely loved seeing the crazy sequence of consonants M L K J R in the dead center of the grid. How many times do you see an entry as cool-looking as that?

I wasn't as keen on seeing WE ARE FREE AT LAST as the final theme entry, because by itself, it doesn't carry Dr. King's message very well, seeming to imply that he's saying that we are (currently) free at last. Perhaps just FREE AT LAST would have been better, clued as something like "38-across's hope for all people in the future". It would also have been nice to see LET FREEDOM RING too, another important phrase repeated in Dr. King's speech.

Puzzle-wise, nice selection of long downs. As Will mentions, those are important slots in a puzzle, as the long downs are an important source of potential snazziness. And as always, Liz does a great job of plunking in two winners, even though her theme density is high. And she also manages to sneak in ATTACK AD, very impressive! The fill suffers a bit though, with such difficult (for a Monday) crossings as APIA/PIAGET as well as unslightly crosswordese (AROO, A IS, NES), and an arbitrary score (ONE ALL). Probably all fine for a later week puzzle, but it's not my personal preference for a Monday.

I hope everyone has the day off today. If you're doing some volunteer work today in remembrance, TWO THUMBS UP. Ow, my shoulder!

– puzzle by Elizabeth C. Gorski

Clever idea today. Word ladders, which connect a starting word to an ending word using intermediates where only letter has changed, have become a little overdone in crosswords, so these days it's important to have some new aspect. I really like Todd's idea here, calling it a WORD LOOP, and actually progressing from WORD to LOOP and back, forming a complete circle around the perimeter.

As with most perimeter puzzles, the fill suffers because of the high constraints. Each corner presents its own problem to the constructor, and even if he/she can get all the corners to fill decently, there's still the huge problem of getting everything to knit together. Typically puzzles can be filled from the center toward the edges, and usually each little edge area is not very constrained, so they're not that difficult to fill out. Trying to get four subsections to mesh together cleanly is a head-banging problem.

I do like how Todd's arranged his black squares to help ameliorate this issue. Notice how the grid is sort of cordoned off, forming a SW and a NE region, along with a barbell connecting the NW and SE? Very smart to do so, and it makes his middle region quite nice. Kudos for working in SVELTE and KEY WEST into a tough-to-fill area.

Due to the requirement of each perimeter answer being four letters, there are cheater squares in the NW and SE. That makes filling much easier. In the NW, Todd does a great job. Yes, there are a lot of RSTLNE letters, but I'm okay with that considering the cleanliness of the fill. The SE is a different story, with A DRIP and A TRIP crossing each other. Typically those are reasonable (if not pretty) answers, but crossing each other draws extra attention.

And the other corners, so carefully cordoned off as best as possible, also suffer. I love WE MADE IT! but not at the price of EBRO, OR ME, OBER all in that single corner. Same goes for NEWSWEEK and CARWAX, two great answers in the SW which bring the very heavy price of OGEE, ORNE, and AGRO all in one tiny space. It makes me wonder if Todd started in the middle and worked his way out, sort of painting himself into those corners.

All in all, an innovative puzzle, one I appreciated for its novelty, but also one so constrained that the trade-off felt a little shaky.

– puzzle by Todd Gross

Neat idea today; I really like puzzles visually representing something. The formation of a trail of bread crumbs from HOME to the WITCH was pretty neat to uncover. And the trail itself, depicted using circled letters, is beautifully laid out, giving a meandering path through the forest of letters. Very cool debut.

I stay away from rebus puzzles these days, as I personally think they're overdone unless there's a really good reason for them. Today, I really like having just those two rebus squares, making for a surprising a-ha moment when I struggled in the SW. And having the central answer refer to the places where the path started and ended was very helpful. Elegant way to do it.

For me, this felt like a case where "less is more" would have been appreciated. To start, I wasn't a fan of the lack of symmetry in the theme answers (ANDERSEN not having a symmetrically placed theme answer). Many will be okay with that, but I appreciate the beauty in the symmetry required out of crosswords. It's totally fine, even laudable to break crossword symmetry rules if there's a good reason (I did like the asymmetrical winding path, for instance), but I didn't think it was worth it here.

In addition, the huge amount of fill (four long theme answers plus BREAD CRUMBS strewn about) made for a difficult construction. Generally there's some great stuff (super impressive considering it's a debut puzzle) like NEAR BEER, CESAR RITZ, and HOME BREW, but EAT NO FAT felt to me like a strange partial (not really a lexical chunk on its own), plus tossing in the arbitrary TWO ROW, long partial A BEAR and the awkward RELOG felt pricey. And TNS, we shall not speak of ye. I imagine less theme density would have allowed for sparklier/cleaner entries.

Great idea, looking forward to more from Jared.

– puzzle by Jared Banta

Very cool idea from Michael today. Congrats on the debut! I always appreciate when constructors learn the rules well enough to break them with thoughtful deliberation, and I felt like Michael did just that. It's sort of a faux-rebus, one where supposed white squares are actually a physical block. I like it even better with his original concept (special squares being black, carrying a white number in them).

Very impressive was the way Michael interlocked his theme answers (BLOCKS OUT intersecting BLOCK PARTIES, etc.). It is true that there quite a few potential BLOCK-related answers to choose from, but to figure out a way to interlock at four different points is pretty neat. I don't think many solvers appreciate how difficult it is to do that.

Generally a good job on the fill too, considering the difficulty of filling around the interlocked theme answers… with the one main exception of TUBE PANS crossing SHEB. Oof, a very difficult crossing, one that I guessed on. Granted, I have roughly the skill of a one-legged dog in the kitchen, but I do know what a BUNDT PAN is. TUBE PAN does google strongly so perhaps that's on me, but yeesh, SHE? could have been anything, and TUNE PAN and TULE PAN felt like they were reasonable.

Final comment on fill: I would expect the NE and SW corners to be on the crunchy side, given 1.) how big they are and 2.) how constrained. But Michael dos a good job in the NE, with AS THE being the only blight. Even given my distaste for five-letter partials, it's not a bad one. The SW does suffer more, with IDAS and ENOW in close proximity. I wonder how many options Michael tried instead of BLOCKAGE, and if BLOCKADE or BLOCKERS or something would have cleaned that up.

Neat idea, perfect for a tricky Thursday.

– puzzle by Michael Hawkins

There are lots of people probably confused as to why they saw this puzzle last week. I don't know the exact details, but apparently there was a mixup at the Times, because Will wanted to run Kevin Der's puzzle just before the MIT Mystery Hunt, because it was actually integrated into the Hunt. Pretty cool, huh? Great idea, but the execution was lacking.

Onward and upward. Another fine offering from Ian, clean as a whistle. I have to think this follows more of his experiment placing the shorter stuff first, then seeing what he can get out of the longer answers. In general, I love the fact that there's hardly any ugly stuff (pretty amazing to have just AWS, HEPS, SSN, and only HEPS is really cringeworthy).

But the other day I had an interesting exchange with another constructor who's opposed to this approach, because it doesn't allow for as many new entries. Since you fix some constraints by placing shorties into the grid first, you limit your ability to insert truly snazzy stuff. I like the colloquial HOW ARE YA a lot and love MILE HIGH CLUB (if only this had been in another venue and we could have seen one of Will's clues), but other than that, nothing spectacular. And an entry like TRADE BOOKS is about as interesting to me as bond funds are to my nephew. (It's pretty fun to chase my nephew around, yelling about credit risk and durations. Ah, good times.)

For just about any constructor I'd stand up and clap upon encountering this puzzle. A 70-word puzzle is harder than a 72, and a 68 is verging on "only possible to do cleanly and sparkly for those named Berry." But Ian Livengood... well, he's Ian Livengood. I was expecting more sparkly goodness out of a 68-worder. It's kind of unfair, but once you achieve name-brand status a la BEQ, DQ, Nothnagel, Walden, et al, the ante is upped. Still, fun puzzle today; a good workout.

– puzzle by Ian Livengood

Now that's a great 1-across clue. (For all you non-physics nerds out there, Newton's first law of motion is often quoted as "a body at rest tends to stay at rest.") ZOMBIE in itself is a very fun answer, but the clue makes it golden.

I really appreciated the cluing today; a bevy of nice ones. Typically fun wordplay clues rely on a question mark, which gives away the fact that it's a fun clue. But "It's not normal" for ANOMALY, "Drink that often makes a person sick" for ONE TOO MANY, and "Get off the drive, say" for DELETE are really nice, all enhancing my solve. Beautiful stuff.

This puzzle is a great example of one having some fantastic, marquee entries at the price of a good amount of crosswordese. ZAGAT RATED, GLUTEN FREE, SURE ENOUGH, SO SUE ME, and NUMBSKULL crossing BOOB with the same clue, that's some great stuff. ULU, OFT, NAUT, ATO, ENL, SSW... that's not. At 72-words, it contains enough great answers to make me happy, but enough ugly stuff that people will complain.

What's going on? Let's take a look at the SE corner. Notice how the triple-stack of 10s turns the corner, forming an ell (where DR LAURA is)? That, combined with the fact that the stack must connect at the SOBER UP area, makes construction very difficult. Not only are there 10 across entries that must be filled in through that triple-stack, but so many of the slots are long. Not an easy task to accomplish.

And now look at the NE corner. See how small it looks in contrast to the SE? James does an admirable job in this corner, with some nice answers (WORD OF GOD!) and only SSW as the price to pay. Typically themeless constructors try to segment their grids into four roughly equally-sized subsections, each of moderate difficulty. But today's xw is a tale of four corners: two easy to fill, two hard.

Fun workout today, the clever cluing helping to make up for some of the uglier entries.

– puzzle by James Mulhern

★ The literal interpretation of OVER/UNDER (TWO under PAR = EAGLE, etc.) has been done many times in crosswords, so constructors have to up their game if they want to go this direction. Dan does just that, adding a new layer to his theme. It took me quite a while to figure out what was going on, but I had a nice a-ha moment when it finally dawned on me. BEWITCHED is literally under A SPELL... under a spell, get it? Clever stuff. It was fun to struggle with each of the six theme pairs.

Sometimes Sunday-size puzzles feel tiresome, like a regular 15x15 that's been stretched out just because it can be. I like the fact today that Dan has chosen enough long entries that makes a 21x21 puzzle the only option. TALK TO YOU LATER = (over) AND OUT, that's something you couldn't do in a 15x15. I really like when the puzzle *should* be a 21x21.

And Dan does a nice job on the fill. Like last week, this puzzle contains a high number of six and seven-letter answers, which have the potential to spice things up. WARTHOG, SKETCHY, BEHEADS, ANTONYM, THE MAGI = excellent! Loading up on six and seven-letter answers is a great way to add snazz to a puzzle. It's an audacious approach, which makes construction very difficult. His shorter stuff does suffer a bit, with ELOGE, NONNO, OSTEO, and the ASKA/BTEN/SANT trio, pulling things down. All in all though, an admirable fill with an above-average result.

Finally, a minor gripe I think others will share. It's perfectly acceptable to have an entry like SLIGO in a 21x, because it's just so darn hard to construct them. Only the Berrys of the world don't need to rely on these types of esoteric answers. But if you have them, I believe the crossings ought to be fair, with clues that aren't back-breakers. "Macros, e.g." for LENSES? Yikes. Maybe if you're a photography buff that's fair, but I thought TENSES or SENSES were as sensible as LENSES.

Very neat theme today, with a lot of great constructing.

– puzzle by Daniel A. Finan

★ Beautiful start to the week. Crosswords have evolved a lot over the past five years, so putting a new spin on this type of theme is important. And TIMESHARE is a perfect way to do that, each of the theme answers "sharing" TIME, i.e. AIR TRAVEL goes to AIR TIME and TIME TRAVEL. Neat concept, perfect for a Monday. Comprehensible for many beginners, interesting for many experienced solvers.

And check out the theme density! Typically constructors shy away from six long theme entries, because it causes all sorts of crossing problems. Four of James's themers are shorter (eight or nine letters) which makes the construction easier than having all 10+ letter entries, but James also throws in something unusual for a six-themer: two long downs. CLEOPATRA and FIBONACCI are fantastic entries. And he also tosses in MAGNETO and BLEND IN. Awesome!

The two long themers are fantastic, but they do come at a price. EMEER is one of those words that makes me cringe, especially on a Monday. EMIR is the more common spelling, never having taken the dreaded "variant" tag in a NYT crossword clue. Out of curiosity, I tried to rework that east area on my own, but with CLEOPATRA running through that area, I couldn't come up with anything better. It would take replacing the ????P???A pattern with ACAPPELLA or WIKIPEDIA or something, which is a pretty major change.

I was curious to see if one of these other long down alternatives could get rid of EMEER (and OLEIC too, hopefully)... but not THAT curious. Anyway, James's grid also incorporates I DUNNO and HOLD EM in that region, so it seems like a favorable trade-off overall. Still, EMEER...

If it hadn't been for EMEER, OLEIC, and the lesser ugly ECOL, this might have been one of my favorite Mondays of all time. And even with them, I found it mighty enjoyable. Well done.

– puzzle by James Tuttle