Friday, August 4, 2017

Should NYT Constructors Demand More Money?

[This is the grid I submitted.  It's the same as the final version, save the first letter in the puzzle.  During edit this was changed to BIG A/BSS.  I think my entries are better, but I understand (I think) the logic behind the change.  RIGA/RSS could be a straight-up Natick at the R, whereas BIG A/BSS is inferable, even if less elegant.]

The puzzle in today's New York Times is a themeless one by me.  I like it fine -- standard Friday fare, I think.  Actually, I'm going to give myself more credit: I think it's a little better than the average Friday puzzle.  I don't have too much to say about it other than that, but I do want to talk about something else crossword puzzle related.

A few days ago, noted crossword puzzle blogger Rex Parker sent out an intriguing tweet calling for constructors to stop "submit[ting] xwords to the NYT until they pay fairly".

I was going to send a reply tweet with some thoughts, but I didn't, because over the years I've developed a strong aversion to responding to things on social media.  I found one of three things happens: (1) nobody cares or responds to my post; (2) somebody cares too much, and I end up in a "comments battle," in which nobody can express themselves fully, both sides just dig in deeper and deeper, and nothing gets accomplished, other than I waste a bunch of time to get stressed out and irritated over nothing; (3) I have a satisfying and informative exchange.  I will let you guess which of these three happens to me the least frequently, by far.  I do post comments on-line, but it's typically about uncontroversial topics in fora in which I already feel comfortable.  (Twitter is not one of these; I think I have three followers and have sent fewer than ten tweets total.)

But this is a topic I'm interested in, so I want to write about it.  The first question:

What is fair pay?
Currently the NYT pays constructors $300 for daily puzzles and $1,000 for Sunday puzzles.  After some number of puzzles (20, I think), you get a 20% increase, so I now get $360 for a daily and $1,200 for a Sunday (theoretically, I haven't submitted a Sunday-sized puzzle in years).  If you do the arithmetic, this means the NYT likely pays its constructors less than $175,000 total per year.

On first blush, this might not seem that low.  I would gladly write the New York Times puzzle everyday for that annual salary.  But, of course, the quality of the puzzle would suffer drastically, because I would very quickly run out of good ideas, and I would be time-pressed on the weak ideas I did have.  If I'm getting over half my puzzles rejected now, when I can cultivated them over days or even weeks, how would things look if I had to pump one out everyday?  I don't think any constructor could keep that pace (although Zhouqin Burnikel appears to be trying!).

That's the thing about the NYT puzzle.  It's the cream of the crop (as judged by Will Shortz, of course), culled from the oeuvres of many good constructors.  It's like having an inhuman super constructor on staff.  And what is fair pay for an inhuman super constructor?  Well, it depends, I suppose, on how much revenue they generate for the paper.  And on this topic we can only guess.  I don't know of any publicly available analysis speaking to this, and it's possible the paper doesn't even really know itself, as it's a very difficult thing to quantify.

But here's a number I'll throw out there: $8 million a year.  It comes from this guy, whose source is a BuzzFeed article, whose source is somebody "familiar with" the paper.  This only includes money from the digital crossword puzzle subscription, so it is very clearly an underestimate of the total amount the puzzle generates for the paper, as it excludes the puzzle's probably still rather substantial dead-tree value.  But I'd rather underestimate than overestimate, so let's go with $8 million.

There is no hard-and-fast rule dictating the percentage of a business' sales that should go to payroll, but from what I can gather reading online, it's usually between 15% and 30%.  So let's go 20%.  On $8 million, that's $1.6 million.  Let's say half of this goes to NYT crossword puzzle staff -- editors, testers, bloggers, etc. -- and half of it goes to the constructors.  (I have no idea what the real percentage is; I'm just trying to make reasonable, conservative guesses, here.)  That's $800,000 a year for constructors.  That's roughly five times more than we get now.

If I'm not wildly overestimating things (and I suspect, if anything, I'm way off in the other direction) then the New York Times could easily afford to pay constructors around $1,800 for a daily puzzle and $4,500 for a Sunday puzzle.  That's a significant bump.

So then what gives, why don't they pay more?
The obvious answer: They don't have to.

I suspect that this is due, in large part, to inertia.  This is the pay model that was in place decades ago, so it's the pay model that's in place now.  I don't think businesses alter their operations to increase costs unless they are absolutely forced to.  And the NYT puzzle seems to be selling fine.  From their perspective -- why change?

After all, even if you think there are better puzzles out there now (say, Fireball or various indies), the NYT has the reputation and they have Will Shortz, who's not just a crossword celebrity, but an actual celebrity.  These two reasons are enough for people to buy the puzzle (for now, anyway).  In the short to medium term, the solvers are mostly expendable.  The "puzzle master" Will Shortz is the star, and the paper is clearly staking a claim on his reputation.  You can see this in the print version of the puzzle: Will's name appears larger and in a more prominent location than the constructor's.

And, by the way, this isn't a criticism of Will, with whom I've only had positive (if limited) correspondences.  He earned his reputation through decades of good work.  Throughout his tenure, he's significantly transformed the NYT puzzle in a positive way and deserves a lot of credit for that.  Also, I think I heard that he's the main reason why the constructor has a byline at all (previously we were anonymous).  And he claims to be an advocate for higher constructor pay.  To what extent and what that means exactly -- I have no idea.

Now, if all the top constructors heeded Rex Parker's call and stopped submitting to the NYT, then I do think the quality of the puzzle would degrade enough that solvers would eventually catch on and get their puzzle fix elsewhere.  (Rex would tell you -- or rather, he does tell you, everyday on his blog -- that the quality erosion is already in full effect and has been for years.  I'm not nearly as down on the NYT puzzle as he is, but, if I'm being honest, I don't think he's totally wrong.)

But a "constructors' strike" is not something that's likely to happen.  For one thing, there is a collective action problem.  For another thing, there is no "constructors' union".  Somebody would have to organize all this from scratch, and I don't see anybody agitating to step up and do so.  And even if somebody did, I don't know if a critical mass of constructors would go along with it.  I suspect a decent percentage of us: (a) don't think we're underpaid (they take the libertarian view -- our work is worth only what we can sell it for on the open market); (b) don't care; (c) do care, but are still willing to submit.

I definitely fall (fell?) into camp (c).  In the same Twitter thread mentioned above, Rex attributes this general perspective to ego and the thrill of seeing one's name in the paper.  He's not totally wrong, but I think he's overly cynical.  It is exciting to see your byline appear above something in the New York Times.  But the thrill wears off pretty quickly.  What it's more about for me is distribution infrastructure.  When you do something you think is cool, you want to share it with as many people as possible.  It's not ego; it's a different compulsion.  And the New York Times is still the best for this.  They have the biggest audience.  I once tried to get my own indie site going, and it sucked and I hated it.  I'd much rather send my puzzles somewhere and be done with them.

There's also the fact that constructors love constructing and do it as a hobby for fun all the time.  If you have a hobby, and you can make a couple grand a year on it, it feels a lot like "found" money.  It's hard to get riled up about worker exploitation, when you are periodically getting $360 checks, for spending your free time the way you want to spend it.  I'm well aware that I can only do this because I don't depend on crossword puzzles as part of my income and that this mindset is precisely what allows places like the New York Times to underpay everybody.  But on the other hand, it doesn't feel right to tell somebody they shouldn't sell their work for whatever they are willing to sell it for.

So what do you do about it?
Well, I'm probably not submitting to the New York Times for a while, but that's more because I have "constructor's block" than anything else.  To make matters worse, after a tremendous streak of acceptances, I've had my last few puzzles rejected, and it's perpetuated something of a crossword funk.  I don't feel like I have any good ideas, and when I do get one, it completely falls apart when I attempt to translate it to the grid.  Right now, it feels like I'm done submitting to the NYT, because it feels like I'm done submitting everywhere.  But that's probably not the case.  I'll probably come out this, and when I do, I'm not exactly sure how I will proceed.

Well, I think I've spilled enough ink on this topic for now.  Please comment if you have any criticisms or thoughts on this analysis.  Until next time...

Friday, June 30, 2017

Friday's Puzzle and an Indie 500 Recap

Friday's Puzzle
A puzzle of mine is running in the New York Times today.  It's one I made a while ago, which is unsurprising, since the queue for NYT themeless puzzles is l-l-l-long.  I've actually stopped submitting themeless puzzles for the time being, in part because the field is so crowded.  The NYT themeless supply apparently far outweighs its demand, so it's getting harder to get themeless puzzles accepted and the wait until they run is just as long as always.  Also, I'm currently going through a bit of a "blah" period with respect to puzzle making in general.  My crossword-constructing battery is low, and I have other projects I want to work on in my (very limited) spare time, so I will probably take a little respite from constructing.  However, I still have a bunch of puzzles in the NYT pipeline (both submitted and already accepted), and these breaks never last too long, so my fans, of which I can only assume I have myriad, should hardly even notice my absence.

Anyway, I don't have much to say about today's puzzle,* so instead I will give a recap of my favorite annual crossword puzzle tournament -- and also the only one I've ever attended -- The Indie 500.

*One thing I will say: I wish the clue for FRANZ FERDINAND wasn't so straightforward.  The clue I submitted was something like "Royal Prince who famously died in an automobile along with his wife in 1914."  Now that has some teeth to it.

Indie 500 Recap
This is my third year at the tournament (it's also the tournament's third year at the tournament), and each year I seem to arrive on the latish end of the *time* spectrum despite living just a few miles from the tournament's location -- or maybe it's because of this.  Many solvers come in from out of town, so this is the only reason they are in D.C. to begin with.  But I go through my usual morning routine (clothing and feeding two small children) before the tournament starts, and my usual morning routine involves me rushing out the door like a madman.

There's only scattered seating available in the tournament ballroom when I arrive, so I grab two available seats in the corner.  I'm supposed to meet a friend there, but he no-shows -- even worse, he doesn't text me to say he can't make it, so I'm there holding his seat like a schmo while the last few entrants are looking for a place to sit.  When people ask if the seat is taken, I have to do that apologetic, shoulder-shruggy "yeah, sorry."  Man, I hate that.  (After the tournament ended, he texted to say something came up and his phone battery was dead all day.  That sounds pretty excuse-y to me, but I'll give him the benefit of the doubt, lest something serious happened, and I'm the asshole for even broaching the subject.  Also, he never expressly asked me to save him a seat, so it's kinda on me anyway.)

I had two goals going into this competition: (1) Finish all the puzzles in the allotted time without error; (2) Don't finish in the top 25% of the field.  The latter is so that I will still qualify for the "Outside Track" (the JV Division) next year.  The former is my own goal since the actual goal of the tournament -- winning -- is not achievable given my current skill level.  I keep telling myself that one of these years I'm going to give competitive solving a serious go and train and try to be a force in my division, but I never do, and I doubt I ever will.  The truth is I don't love doing crossword puzzles.  I very much like doing crossword puzzles, but I'm not fanatical about it the way the top solvers are.  (I've always been way more into constructing than solving.)  I usually "only" do one a day, often in the background while I'm at work, which isn't nearly sufficient to get me near the top of even the lower division.  So I set my own little goals and enjoy the puzzles and the company.

Speaking of the puzzles...

Puzzle 1: "Before and After" by Angela Olson Halsted

A straight-over-the-plate, easy-breezy, Monday-level puzzle to kick things off.  The theme answers consisted of two words, one of which can precede the word TIME (the tournament theme, if you didn't already know) and one of which can follow the word TIME, paired together in zany ways.  So one theme answer was HAMMMER BANDITS (Hammer Time/Time Bandits), and the clue was "Construction site thieves?"  Then everything was tied together using Cyndi Lauper's timeless hit "___ After ___".

I thought this puzzle was pretty good.  It was enjoyable and didn't have any big flaws, but I'm not crazy about the randomness of the TIME phrases.  There are countless combinations one can come up with (Gratis use of the flute on Super Mario Bros. 3?: FREE WARP), so the ones that are chosen really have to hit the mark -- be very funny or clever -- or tie together in some other way.  The ones in this puzzle are fine; they're cute, but none of them stuck with me after I finished.

And I finished quickly, for me -- 5:03 with no errors.  This was actually near the top of the Outside Track.  I always thought I was more of a long-distance solver than a sprinter, but it seems to be the exact opposite.  What I really need to do is work on my endurance, as the bigger puzzles gave me much more trouble than this one.

Puzzle 2: "Jam Session" by Paolo Pasco

A nice offering by constructing wunderkind Paolo Pasco.  (He apparently couldn't be at the tournament because he had to take his SATs; I don't think this was a joke.)  This one had four CRUNCH TIME theme answers with a unit of time "crunched" two-letters per box.  The clues were also written under a faux time crunch, so they got sloppier and sloppier as the puzzle went on (e.g., "Suffres s form" for HAS).  I liked that added dimension, but I heard several other solvers grumbling about it between rounds.  I could see how, in the context of solving under an actual time crunch, one might find it annoying, but I thought it was fun.

I picked up on the the theme fairly quickly, but spent a lot of time trying to parse out some of the proper nouns.  It wasn't the spate of millennial pop culture that got me, as one familiar with Mr. Pasco's work might think (although I had never heard of this ANSEL Elgort guy), but rather it was some old stuff -- an Ottoman leader named ALI PASHA and a Tom Wolfe story from 1976 called "THE ME DECADE".  I had to Google both of them after I finished to make sure they were correct.  They were, so I was two-for-two with clean grids.

Puzzle 3: "This Mashup's for the Byrds" by Tracy Bennett

Every year there is a puzzle that I do not understand at all while I'm solving it and barely understand after I'm finished and looking back over it carefully.  This year it was Puzzle 3.  The basic idea, which I didn't fully appreciate while solving, is that the theme clues are lyrics from the Byrds song "Turn, Turn, Turn" with one letter added and then the answer is something cute that fits that new lyric.  So the first one is "A time to gather stoners together":  BURNING MAN.  The actual lyric from the song is "A time to gather stones together", but, man, you have to really know the song well to pick up on something that subtle.  And why Burning Man?  It's just an answer that fits?  And then the extra letters apparently spell out RENT, and one of the long down answers is a song from the play Rent.  What the what?  There must be a connection here that I'm missing?  Was Rent written by David Crosby or something?  I totally don't get it.  (On the plus side: cool looking grid.)

In addition to being, in my opinion, too obtuse a concept, the execution seemed off in a few places.  For example, one of the clues was "A time you many embrace", which I cannot read in a manner that makes grammatical sense.  I was able, however, to piece together the answer, GROUP HUG.  In fact, I was able to piece together all the answers.  My times aren't great, but I'm perfectly accurate at the lunch break.

Puzzle 4: "Non-Linear Narratives" by Erik Agard ft. Allegra Kuney

This puzzle was weird -- really weird.  It was also my favorite one of the bunch.  I don't think I can succinctly describe the theme, but it involved writing in animals in the wrong direction, and it had three revealers -- one for a pair of theme answers in the grid, one for another pair of theme answers in the grid, and one for all four of them.  While I was solving it, it seemed like madness, and I honestly wasn't too keen on it, but after finishing I really appreciated it.  It's like when you see a movie and you're not sure if you like it or not, but it sticks with you for the rest of the day, so you conclude that it must have been pretty good -- that's how this puzzle was for me.

It was also, time-wise, probably my worst puzzle of the bunch.  I got stuck in a corner and was getting so irritated, I contemplated filling in half-random guesses just to be done with it, but I'm glad I didn't.  Again, I was not in contention for any sort of real prize, so there was no reason to not stick it out until the bitter end.  And as it turns out I didn't need until the bitter end anyway.  After about ten minutes of staring at the empty boxes and rereading the same clues a hundred times, something clicked (as it often does), and I finished the puzzle without error.  One more to go!

Puzzle 5: "In Search of Lost Time" by Neville Fogarty

A perfectly cromulent add-some-letters/subtract-some-letters puzzle to close out the day.  Other than Puzzle 1, this one was the most straightforward of the set, and coming on the heels of Erik's and Allegra's avant-garde behemoth, it was very welcome for this solver (I'm pointing to myself).  The theme involved adding and subtracting the letters ERA from common phrases to create new zany phrases, e.g., "Secure movie rights for a haboob?" was OPTION DESERT STORM and "Smudge on a Scantron form?" was ERASURE THING.  (I originally read the latter clue as "Smudge on a Scranton form," so I thought it had something to do with The Office.)

This, of course, is a very common crossword conceit, but that's not necessarily a bad thing.  A puzzle with a well-worn theme can still be really fun, if it's really well done, and this one was really well done.  Transforming SHOCK TREATMENT into SHOE RACK TREATMENT was especially clever.

And best of all (for me) I solved it without an error.  Yay!  I did it.  Five-for-five!  No mistakes!

I finished 20th of 89 in the Outside Track, and I finished 56th of 128 in the combine field -- a little better than average, which is where I think I fit in in life in general.  Pretty much everybody at the tournament is a really good crossword puzzle solver compared to the general public, and for the things I'm really good at, when I'm tossed in with a group of people who are also really good at them, I feel I'm usually somewhere between the 55th and 60th percentile.  At grad school, I definitely wasn't the best student, but I think I was solidly above average.  In high school wrestling, I would get soundly beaten by state champion contenders, but I was varsity for three years and had a winning record overall.  When I used to play competitive Scrabble, I was usually near the top of the middle division.  Always the same place.

Maybe that's a good new nickname for me "The 57th Percentile".  It doesn't sound too impressive, but one must keep in mind that it's among people who are already quite good at doing the things they are doing.  With that qualifier in mind, it's not too shabby at all.

Anyway, getting back to the tournament, I actually solved the final puzzle faster than any of the finalists in the Outside Track.  Almost certainly this is in very large part because I didn't have the stress of solving in front of everybody on a big board.  Also, I chanced into the trick pretty early on -- I randomly started solving at the bottom first, which I don't do normally (and probably wouldn't have done had I actually been in the finals) -- which allowed me to use the Across clues pretty early on in the process -- a huge boon, obviously.  Congrats to Eric Cockayne, the Outside Track winner, and to Katie Hammil, the Inside Track winner.

I split pretty quickly after the finals (I wanted to take advantage of my kid-free day and stop by a friend's birthday party), so I missed the annual pie-in-the-face tradition -- each year a solver, selected randomly, gets to throw a pie in the face of a constructor of his or her choosing.  I heard Erik Agard was the pie-ee this year.  It's just as well I missed it, as it makes me feel uncomfortable, to be honest.  I don't get what's enjoyable about watching somebody get hit with a pie in a completely contrived setting.  It seems like a big mess for not much payoff.  It also seems kinda mean to me (even though I know the recipient is very much in on the joke).  But others seem to like it, and it's all in fun, so whatever -- no harm, no foul.

Well, that was my experience at The Indie 500 2017.  It was great.  My final grade for the entire opERAtion: A.  I hope to see you there next year.  Until then...

Friday, May 26, 2017

Keep Your Shade to Yourself... Or Don't (It's Cool Either Way)

If I'm counting correctly, including this puzzle I have nine themeless puzzles in the NYT queue for publication and another two submitted.  Out of these 11 puzzles, this one is my least favorite.  This isn't to say I don't like it.  If I don't like a puzzle I don't submit it.  It's just to say that I like the other ones better.

The flaws with this puzzle gnaw at me more than usual for some reason.  The "Texas" region is particularly grating.  I hate using a plural of an uncommon name, and ETTAS is right on the borderline for me.  It's better than GISELES, but not as good as ANNS.  (Although it does make me feel better that they were contemporaries in the same field.  It's easier to imagine an organic usage of ETTAS in this setting: The two Ettas of mid-twentieth-century jazz -- James and Jones -- had 18 Grammy nominations between them.)  And I really, really hate using partials, so much so that I tried to clue I ATE as a complete sentence.  My clue, changed during edit, was "Already had dinner."  But it was an act of desperation on my part, as I can't really think of a non-contrived scenario in which somebody would say "I ate" as a standalone sentence.  One might say "I ate dinner already" or even "I ate already," but just "I ate?"  Eh, probably not.  Then, to make things even less elegant, ETTAS and I ATE cross the abbreviation SEPT.  Normally I would have no problem with SEPT in my puzzle -- it's a perfectly cromulent abbreviation -- but when it's crossing the two worst answers in the puzzle, it becomes a weak entry as well.  Guilt by association.

The other big thing I'm not so keen on is that I didn't totally nail the long answers.  ANNOTATE and ANTENATAL are both pretty boring, and DIETETICS -- I dunno.  I think it's an interesting word, but your mileage may vary.  I like COLPORTEUR too, because it's fun to say, and I enjoyed learning its definition, but I would understand if others thought it was too obscure for such prime real estate.  Similarly, I like the clue for FLORIDA TECH (classic Saturday-level misdirection), but it's appearance in this puzzle is pretty random.  It's not like Cal Tech or even Georgia Tech when it comes to prestige and notoriety.  I imagine many solvers inferring it from crosses and thinking, "Florida Tech?  Okay.  Sounds like a real place to me."

But, of course, there are many things I like about this puzzle as well.  PALE BLUE DOT is a cool debut entry, especially for Neil de Grasse Tyson fanboys like myself; FLEXITARIAN is pretty good, even if my original clue didn't make the cut ("Person you might occasionally have beef with?"); HAVISHAM is a nice literary reference; and THROW SHADE makes its first (to my knowledge) appearance in a crossword puzzle.  Then there is DON'T THAT BEAT ALL, which is my favorite answer in the grid, because it's a good colloquialism, and because it's very difficult to get any grid-spanning answer through the center of a triple stack, let alone one with some sparkle.

In general, I hope it isn't lost on the solver that this grid shape is a bit more challenging, from a construction standpoint, than that of a typical themeless puzzle.  It's only 66-words, six less than the themeless max, and it has a wide open middle which was a bear to fill.  So I think I did pretty well, given the constraints, but "given the constraints" is not always to the benefit of the solver.  There is is not necessarily a positive correlation between difficulty of construction and enjoyment of solve.

Another area in which I think I did nicely with this puzzle is the mid-range fill.  FACADE, EGO SURF, TOP SHELF, NO FRILLS, NOOGIE, VAPE, BOSOMY, and PREFAB are all solid to good, in my opinion.  

The big question: Overall, does the good outweigh the bad?  I think so, but, just for fun, let's do a full accounting of this puzzle using Jeff Chen's "asset/liability" score.  That is, you simply add up all the "assets" of the puzzle and deduct all the "liabilities."  If this difference is around 10 you have a decent themeless puzzle; if it's much above 10 you have a great puzzle; and if it's much below 10 you have not-so-great puzzle.  Now, obviously this is a highly subjective, overly simplistic metric, but here goes nonetheless...


8.  VAPE  

2.  I ATE

2.  SEPT

I decided to parse the scoring a bit more finely by using half assets and half liabilities.  And under this categorization, my score for this puzzle is 8 + 4 - 2 - 1 = 9, which seems... spot on to me actually.  It completely coheres with my gut feeling on this puzzle.  I guess the system works!

Anyway, I hope you liked this puzzle.  If you didn't, feel free to throw some shade in my direction.  I never object to honest feedback.  Plus, there is a decent chance I won't read it anyway, at least not right away.  We are taking the kids to Sesame Place in Langhorne, PA for the holiday weekend, so I'll be busy with that.  The only "cross words" I'll be dealing with are the ones I'll be muttering under my breath, as I'm pushing through a throng of people so that my sons can shake hands with some poor schmo in an oversize Elmo costume.  Instead of sitting at home refreshing the comment sections of the crossword blogosphere, I'll be spending quality time with the family -- waiting in line so that we can get our picture taken with a giant cardboard cutout of The Count.  It should be fun.

Until next time...

Thursday, April 20, 2017

Tales of a Grammar Nazi in Go-Go Boots

[Image lifted from XWordInfo where I got the weekly "POW!"  Nice!]

Writing clues is my least favorite part of making crossword puzzles.  If I ever become a professor of cruciverbalism, I'm definitely delegating clue writing to my graduate students.  I might deign to make a suggestion here or there (How about "Work with intelligence?" for SPY STORY -- get it?), but I certainly wouldn't be doing the grunt work.

When I first started constructing puzzles, clue writing was tortuous, and the torture was completely self-imposed.  I would slave over each individual entry trying to think up the cleverest, most personally stylized clue possible.  It took me nearly as long to write the clues as it did to build the grid -- and it took me a long time to build the grid, as this was back in the day when I used graph paper and pencil, and the only "software" I had was an electronic pocket crossword puzzle helper.  (By the way, I feel no romance at all for those days; I much prefer it now, doing everything on Crossword Puzzle Compiler and utilizing massive word lists.)  It was not effort well spent.  What I found is that I was very good at writing clues that were almost certain to be changed during edit.  That is, if my puzzle even got accepted anywhere, which, at that point in my "career," it probably wouldn't.

Although I would like to blame all those stuffy editors for failing to the see the genius in my clue writing, the truth is, I was very likely going about it wrong.  Being clever is good -- most fun puzzles have clever clues -- but not for all 70-some entries.  As a solver, I've come to realize there is often a fine line between clever and annoyingly contrived, and trying to force the former is a great way to achieve the latter.  I think most solvers rely on and appreciate a certain consistency and repetition in cluing.  Editors understand this, which is why clues for a given entry are very frequently the same.  In fact, what I've noticed about the New York Times is that it often goes on a "mini-run" for a particular entry in which clues are the exact same or nearly the exact same for several consecutive puzzles.

For example, here are three screen shots from Cruciverb's database showing the clues for the entry SLAY.  This seems to happen too frequently to be a coincidence.

Eventually I caught on and I stopped putting so much effort into writing original clues.  What I do instead is, for the vast majority of entries, I look it up on Cruciverb, filtering by the publication to which I'm submitting, and then I pick a clue I like that's day appropriate and use it or something very close to it.  I figure, if it's going to be changed to a standard clue during edit anyway, then why not make things easier on everybody?  But then after that I pick out a handful of entries, usually the "marquee" entries, and try to come up with new, clever, and fun clues for them.

Thus far, this process has worked out pretty well.  Most my clues, even my original ones, survive edit (though often slightly reworded), and I don't spend hours and hours writing a bunch of clues that will never see the light of day.  I now feel significantly less like a character in the Beatles' song "Eleanor Rigby."  (Damon Gulczynski writing the clues for a crossword that no one will see.  How can that be?  All the lonely people, where *do* they all come from...)

In this puzzle, I came up with a few clues that I really like.  In fact, one of them is probably my favorite clue I've ever written:

"Type for who this clue will be annoying?" GRAMMAR NAZI

I'm quite proud of that one.  Another one I like:

"A batter receives four for a grand slam" TOTAL BASES

Hopefully fellow baseball nerds will appreciate the redirect from the answer that immediate comes to mind but doesn't fit, RUNS BATTED IN.

Then there is this one:

"Iconic part of Sinatra's attire?" GO-GO BOOTS

I was quite disappointed to see that this one did not survive edit, and that the name "Nancy" was added before Sinatra.  It totally ruins my intent, which was to make the solver first think it was referring to Frank Sinatra (why doesn't FEDORA fit?).  That's a clever misdirect, no?

I'm not sure why it was changed.  The only thing I can think of is that Will thought the solver wouldn't really get it.  Maybe Nancy Sinatra isn't well-known enough or closely enough associated with go-go boots without her first name?  Perhaps.  But she did have a no. 1 hit back in the day explicitly about her boots.  So... I dunno.

Anyway, this puzzle, like the one published a few weeks ago, took me three tries.  The first didn't turn out well at all.  It was one of the earliest themeless puzzles I had ever attempted, and I should have chucked it when I was done with it and called it a learning experience.  I actually can't find a copy of it to post, which I'm not particularly sad about.

Below is my second attempt.

I actually really like this grid, my reasons why are the same ones why it was rejected.  The upper half is a procession of proper nouns -- DISCO STU,  DODIE (Stevens), OTTO HAHN, SAM ELLIOTT, (Justin) THEROUX, DON QUIXOTE, RAQUEL WELCH.  So much good, fun stuff!  But, alas, as I've heard before, not everybody likes what I like, and one thing people seem to like are puzzles that don't feel like a pop culture trivia contest.  So it goes... And if I want to keep publishing puzzles (which I do), so I have to go too.  But that's okay.  I think this one turned out pretty well as it is.  We shall see if people agree.

Until next time...

Friday Morning Update: Judging by what I'm reading on the blogosphere, people do think it turned out pretty well.  More than a few solvers, however, are put off by seeing NAZI in the puzzle, even as part of the tongue-in-cheek phrase GRAMMAR NAZI.  This is completely understandable.  I probably would not use this entry if I made this puzzle today, given the political climate.  I still love the clue though.

Also, I see that in the newspaper version of the puzzle, the clue for EMU included a symbol.  Cool.  That's a good way to liven up a common answer.

Wednesday, April 5, 2017

It's about the X's and O's (Not the Jimmys and Joes)

[Published Grid]

It took me a couple tries to get this one right.  The grid you see above, the final grid, was my third submission.  My first attempt, seen below, was very different.

[Version 1]

I actually don't even remember making this grid.  I knew that I made three of the them, but then I couldn't find this one on my computer right away, so I started doubting myself, thinking I must have only made two.  And now that I have found it and am looking it, it only vaguely rings a bell.  I often construct at night in a sleep-deprived state (I have two young boys, so it's that or nothing), and I must have constructed this mostly on auto-pilot.  And if that's the case, it turned out pretty well considering.  But not quite well enough.  It got rejected -- encouragingly rejected, but rejected nonetheless.  The grid is too segmented and contains too many three-letter words (22).  Also, Will apparently doesn't care for BARF in his puzzles -- noted.

I put it aside for a while, and when I came back to it, I had a very difficult time opening up the grid and finding good theme entries that fit.  One thing that made this grid tricky is that the X and the O are not in symmetric squares, so if you want to make the theme answers symmetrical, which I decided I did (to keep things tight and to "reward" those solvers who figured out the trick early on), then you have to put two pairs of theme answers right next to each other.  Like I said, I struggled.  I talked myself into the grid below, but I never really felt good about it.

[Version 2]

This grid has some highly problematic entries -- for example: MAN OS.  A "man hug" is most definitely a thing in my world (I've given and received quite a few), so it might be fine as a non-rebus entry, but with the HUG part replaced by an O, solvers might not have any clue what's going on.  The same goes for THE SQUARED O.  Outside of professional wrestling fans, people might not have ever heard this term, completely negating the opportunity for that elusive "eureka" moment.

So I had my suspicions that this one would also be rejected, and a few weeks after I submitted it, those suspicions were confirmed.  But I tried again and finally got something that works.  My breakthrough was using those Utah-shaped blocks of black squares in the upper-right and lower-left of the final grid.  It's something I'm loath to do, because it's not very visually appealing, but in this case it made all the difference in the world.  It allowed me to put in the much more crossworthy BEAR OS and replace X CENT with X SPEED, which I think is an upgrade, and I didn't have to sacrifice the answers I really like (THREE X A LADY, X MY GRITS, and O SUM GAME).

The final piece to the puzzle was finding a new 16-letter "circle" entry at 26-Down.  My word list didn't have any entries, but it did have some that were 15 letters (unsurprisingly, given most girds are 15 x 15), and one of them was BRING FULL CIRCLE.  So I could have gone BRINGS FULL CIRCLE, but that didn't sound quite right to my ear.  The phrase, as I know it, is "come full circle," but that's too short, and it would require a non-rebus O, so I had to make do with TURNED FULL CIRCLE, which probably isn't something I would ever say, but it's in the dictionary, and it sounds better to me than the "brings" version, so it's a decent compromise.

Overall I'm pretty happy with how this one came out.  Let's hit some bullets and call it a post.

  • Hopefully the solver noticed that I didn't use any X's or O's outside of the theme entries.  This was a serious pain in the ass.  It turns out filling in a crossword grid without the letter O is quite challenging.  And I don't know how many times I thought I got something to work and then realized "Oh no! It's an O!"  You just look right past them, even when you are explicitly trying to catch them.  Once I even did this with an X, which just felt wrong.  It's like, c'mon, avoiding X shouldn't be what trips me up!
  • Interesting tidbit about the letter X: There has never been a player in any of the big four American professional sports leagues whose last name starts with X.  There was a baseball player named Joe Xavier who made it as far as Triple-A (with my childhood local minor league team, the Tacoma Tigers) but he never made the show.
  • Not so interesting tidbit about the letter O: There have been many players in each of the big four leagues whose last name starts with O.  My personal favorite is former Cubs great Orval Overall.  But the one most crossword puzzle solvers know is surely Mel Ott (see the final across entry of my first grid).
  • One review is in as I write this.  Jeff Chen at XWordInfo liked my puzzle and even gave it his weekly "POW!"  Sweet!
  • Wait... another one just came in.  Andy at Crossword Fiend like it as well.  Two for two.  I probably will have to wait until the morning to find out what Rex Parker has to say about it, but, I mean, c'mon that's a lay in, right?  That guy never has a bad word to say about the New York Times crossword puzzle. [Update: Rex said the he "enjoyed this one," which is basically a rave review from him, so I'll happily take it.]
I might check back in later.  But if I don't, until next time...

Monday, March 13, 2017

The Salad Days of a Man's Man

This puzzle is meant to have two layers of theme.  One layer obviously is that the theme entries are actors whose names are types of salad.  But also I wanted to chose actors who would actually evoke nostalgia for solvers' SALAD DAYS.  I wanted to chose big name movie stars who spanned a large portion of film history.

Unfortunately, crossword puzzles require entries be within a certain range of word length and have an appropriate symmetry, and there are only so many last names that are also salads and only so many people who have ascended to the height of movie star, so I had to do make do with what was available.  I like LEE J. COBB (great actor too -- 12 Angry Men is a classic) and SID CAESAR.  I'm fine with ORSON BEAN.  But I would have liked an upgrade over TOM GREEN.  He's the least movie-starry of the four, and, personally, I never thought he was particularly funny.  My SALAD DAYS included very little Tom Green.  In retrospect, I could have easily gone with EVA GREEN, but I didn't realize this until the puzzle had already been accepted.

Eva Green also would have been better because it would have gotten a woman into the puzzle.  As it is, we got four dudes.  If you keep up with the crossword blogosphere, you are certainly aware there is a push among many puzzle enthusiasts to make crossword puzzles more inclusive.  I think this is a good thing.  Constructing and competitive solving are dominated by men. In the former, there are a few prolific females (particularly C.C. Burnikel who I swear has a puzzle published somewhere every day) and then that's about it.  The big question: Why is this the case?

It's something I've thought about quite a bit.  In a way, it's something I've been thinking about most my life, as I discovered at a relatively early age that most my major interests -- sports analytics, math, Scrabble, crossword puzzles, etc. -- are predominantly male pursuits.  After many hours thinking and rethinking about this, reading opinion pieces, scientific studies, and the occasional controversial "academic exploration of hypotheses," here's what I came up with as the reason: I have no fucking idea.  I haven't come across a single theory yet that I've found particularly convincing.

The root cause I think is clearly sexism.  Back in the day, women were discouraged from doing these activities, and so we're seeing the residual effects of that today.  This makes sense, but the missing piece -- the thing I really don't understand -- is why aren't we seeing this male-female gap to the same extent in other fields.  Back in the day, women were discouraged from doing pretty much anything other than raising a family, right?  So why is it that today woman are much better represented in fields like law and medicine (though still not close to 50-50) than they are in fields like math and crossword puzzles?

Like I said, I have no answers.  I do think however that if somehow we could get more women involved in underrepresented fields that it would self-perpetuate.  I remember hearing about a study, the author of which I can't remember on a podcast I can't remember (get sourcing, huh?), demonstrating that representation really matters a lot when people are in their formative years.  If a child sees somebody with whom they can identify doing something then they are much more likely to pursue that thing than they would be otherwise.  This is why diversity matters.  Even if you don't believe it's inherently good, without it you effectively block people, particularly young people, from pursuing things they might otherwise want to pursue (and might be very good at), and that's not the way things are supposed to work in the land of the free and the home of brave.

So it seems to me the only solution I have for getting more women involved in something like crossword puzzle constructing is to get more women involved in crossword constructing.  And that's not actually a solution at all.  It's a tautology.

(And by the way, if there are any aspiring female constructors reading this right now, and you need some guidance, I'm available!  Actually, I'm available even if you're male or anywhere else on the gender spectrum.  I'll just be excited somebody is asking me for advice about something.)


Getting back to my puzzle, I'm completely satisfied with it.  I'm putting this up before I read any commentary about it, so I hope others feel the same way.  The theme ended up being relatively simple, but I'm fine with a simple theme.  If you have one though, I think it's imperative to make the non-theme entries sparkle and keep the grid clean.  I feel like I did a good job with both of these.  MAN'S MAN, MIC DROP, AW GEEZ, COIN-OPS, PAYPAL, TANK TOPS and I GUESS are all pretty solid entries, and the worse entry, in my opinion, is ESAS, which isn't even that bad.  I'm guessing some will balk at RVER and AAHS, but I stand by both of those.  RVer is a perfectly cromulent term for a person who travels by RV, and "oohs and aahs" are what people say at fireworks shows.

Another thing I anticipate some solvers not liking is all the proper names in the puzzle (NAMES, ironically, not being one of them).  In addition to the theme entries, you've got ANI DiFranco, NGAIO Marsh, JADA Pinkett Smith, Nellie BLY, Judd APATOW, and Susan DEY.  I concede that's a lot, but... that's kinda my thing.  I've found I like a much heavier dose of proper nouns and pop culture trivia in my puzzles than most people, and since I'm me, I often design my puzzles the way I like them.  I mean, crossword puzzles are supposed to have personality, right?

Alright, that's all I got for now.  I might pop back on after I read the reviews.  It depends on what they say.  And it depends on how much time I have.  It's supposed to be a snow day tomorrow (I'm writing this Monday night in D.C.), and when you have two children under the age of five, a snow day is not a day off.  It is very much a day on.

Friday, January 27, 2017

Don't Hate the Elver, Hate the Eeler

[Published version]

One of the hardest things for me as a crossword puzzle constructor is avoiding the "great answer trap."  That is, when you have a "great" answer in your puzzle (or a "great" section), and you are unwilling to let it go, no matter how difficult it makes completing the rest of the puzzle, no many how many undesirable compromises it requires.  I've gotten a lot better about this as I've gained more experience (I now try to embrace the YAHOO philosophy -- You Always Have Other Options* -- and I try to remind myself that my great entries are never as great as I think), but it's still something with which I constantly have to grapple.

In a bold attempt to combat the great answer trap, I came up with the idea of making two puzzles in parallel with very similar grid shapes.  My thinking behind this was that if I came up with a great answer that didn't fit in puzzle A, I could use it in puzzle B, which would not have the exact same constraints as puzzle A, and therefore would be more amenable to my great answer.  In this way, I would be able to keep my great answers and get rid of them (and move on) at the same time.  Win-win.  But of course it didn't work at all.  It's probably pretty obvious why.  Once you get a few entries into each puzzle they diverge to the point that you can't just move an entry from one puzzle to the other.

What ended up happening is that I just made two themeless puzzles with the same basic shape but very different fill.  You can see the grids below.

[Two rejected crossword puzzles]

I submitted both puzzles... and both were rejected.  It was the lower left corner with each of them.  In the first one, Will didn't like IT'S DEJA VU, because it's not really a standalone phrase.  (The saying is "it's deja vu all over again.")  In the second one, he just felt like the Southwest was way too drab -- which it is.  I fell into a different trap there, the "surely, everyone likes what I like" trap.  I went to University of Maryland for grad school, so I overrated how good of an entry TERRAPINS is.

Of the two I actually think the first one is more promising, but I got the impression that Will and Co. liked the second one better, so I reworked the lower left corner (which turned into a rework of the entire lower half) and submitted a new version.  As you probably already pieced together, it was accepted.

I'm curious to see people's responses to it.  Joel Fagliano, in the acceptance email, said, "Filled with lively vocabulary, and practically no compromises. One of your better constructions to date. Big yes on this, for a Saturday," which I was pleasantly surprised by.  I actually wouldn't put this one among my best puzzles.  I like it, of course.  I think it's good.  But if I was putting together a portfolio of my ten best puzzles or so, I'm not sure this one would make the cut.  But, hey, if other people really like it that's totally cool by me.  It's certainly better than the alternative.  We shall see.

Alright. let's do a few bullets and call it a post.

  • If DOSCH were a thing, I would have made BAD JOKE into DAD JOKE.  It's only a matter of time before this neologism finds its way into a crossword puzzle grid.
  • My clue for BON JOVI referenced "Lay Your Hands on Me" (changed during editing to "Blaze of Glory"), which is my favorite Bon Jovi song by far.  I'm not a huge fan by any means, but I do have a soft spot for "Lay Your Hands on Me," because it's the first song on the album New Jersey, which my friend got on cassette in fifth grade, and we would listen to the first track, rewind it to the beginning, and repeat ad nauseam.  That's a funny thing to think about now -- rewinding.  I remember buying cassettes based almost entirely on what the first song was on each side, because those were the only ones you could easily find.  Rewinding and fast-forwarding to the exact start of a song is something kids today will never have to deal with.  Actually, come to think of it, it's something I almost never had to deal with.  By the time I really got into music, CD players were already pretty mainstream.  I still remember my first CD: Flesh and Blood by Poison.  Don't worry, I redeemed myself with my second CD: Shake Your Money Maker by The Black Crowes.
  • The RED ROBIN with which I'm most familiar is the hamburger joint.  I loved this place as a kid.  As an adult, eh... not so much.
  • The HOT SPUR with which I'm most familiar is Tottenham Hotspur F.C. of the English Premier League.  I don't follow soccer, but what I've found is that I've consumed so much sports throughout my life that now I somehow just absorb it through osmosis.  Even when I'm not actively paying attention I still know basically what's going on -- like I haven't been watching the Australian Open and I don't really follow tennis, but I know both top seeds have already been eliminated.  (Update: I saw Serena beat Venus in an all-Williams.  Hell yeah!  I love the Williams sisters.  Serena might be my very favorite athlete ever.)
  • I used to know a song about JOHN HENRY as a kid.  The only part I remember now is the lyric "John Henry was a man who could beat a machine."  I wonder if I can find it online.  Ah, yes, easy.  It's a pretty well-known song.  Although it doesn't actually have the lyric I remember.  This isn't surprising at all.  Things are almost never verbatim how you remember them.  One of the interesting things about keeping a blog is going back and rereading it years later.  It's like auto-correct for your memories.

  • To all the ELVER haters out there: Elvers are real things in real life.  If you don't believe me, click here or here or here.  Eelers on the other hand are total bullshit.
A few additional thoughts, added after reading some of the reviews of my puzzle.
  • Jeff Chen at XWordInfo seemed to mostly like the puzzle, but he found it frustrating because he had a rough time in some sections (and not, apparently, in a satisfying, "a-ha moment" type of way).
  • Deb Amlen at Word Play also liked it (but her column is never very critical of the puzzle) but also found it hard. Difficulty, of course, is highly subjective, as evidenced by the fact that...
  • Amy Reynaldo at Crossword Fiend  definitely liked it, but found it too easy...
  • And Rex Parker didn't enjoy it very much because he found it way too easy.
  • Rex also implied that kir is Crosswordese, which, in my opinion, it most definitely is not.  It's a word I've seen many times outside of crossword puzzles (like on drink menus at French restaurants).  In fact, here's an article written last year about how kirs are becoming trendy.  Whether or not this is true or not, I do think it strongly indicates kirs are real things even to non-crossword solvers.
  • He also posted a tweet of a scuba instructor and equipment technician who had never heard of AIR PIPE.  Fair enough, it doesn't appear to be super common, but you can find it "in the wild," so it's not totally made up.  Here's a quote from the website of Gensis Diving Institute of Florida (bold is mine): 
Minimum Age 10 years, Jr. Open Water Certificate, or equivalent, open water equipment with compass, 50 lb lift bag, 10 ft rope and 2 marker buoys. It is suggested to have a pony bottle of 30 cu ft with an air pipe attached to a low-pressure inflator hose of the regulator. 
*I got this acronym from the notes of another constructor, but I can't remember whom.  If anybody knows who coined this, please post it in the comments, so that I can give credit where credit is due.