Nieman Foundation at Harvard
“The way we raise the money at The Guardian is different than any place I’ve ever been”
ABOUT                    SUBSCRIBE
Jan. 11, 2022, 3:50 p.m.

The New York Times debuts a fellowship for crossword constructors

NYT Games editorial director Everdeen Mason on building a more diverse set of puzzle constructors and an “ecosystem” for solvers. “I don’t want people to just come in and play a game and leave.”

The New York Times announced a new fellowship for crossword constructors on Monday aimed at increasing the number of puzzles created by underrepresented groups, including women, people of color, and those in the LGBTQ community.

The New York Times Diverse Crossword Constructor Fellowship, which will be open to applications for a month starting Feb. 7, is one answer to criticism that the Times has been slow to diversify its crossword clues, answers, editors, and (yup) stable of constructors.

“You’ll get a rejection from the Times saying ‘This is not something that the average solver will know,’ which carries with it this connotation that an average solver is a white man in his 50s,” one crossword constructor told me last year. “There’s an expectation that the person solving your puzzle looks like Will Shortz.” (Until editors Wyna Liu and Tracy Bennett were hired in 2020, the Times puzzle team consisted of three white men.)

The chosen fellows will receive three months of mentorship from one of five puzzle editors — that’s Liu, Bennett, Joel Fagliano, Sam Ezersky, and the legendary Shortz — as they work to construct a puzzle for general submission. Only those who have not yet had a puzzle published by The New York Times will be considered.

The new constructor fellowship is the brainchild of Everdeen Mason, who joined the Times as editorial director for Games almost exactly a year ago. Mason said creating a fellowship has been on her mind from the very beginning.

The New York Times, even while interviewing Mason for the job, had acknowledged it need to change. Part of the diversity problem was self-perpetuating, Mason saw. Constructors seemed to have a fixed idea of the kind of puzzles that The New York Times chose to publish.

“It was becoming clear that people were sending us the kinds of puzzles they thought would get in rather than really pushing the limits and trying to show us new things,” Mason said.

The fellowship, she hopes, can help kick off a new pattern. “I think that in mentoring these constructors, we’re going to learn stuff, too,” Mason said. “And we’re going to be able to model certain things. If we see a really cool puzzle with new kinds of clues and fills and publish it, hopefully that encourages people to give us more like that.”

The learning curve for potential constructors can be steep, and the digital tools most popular with professional constructors can be costly and user unfriendly. Mason noted the barrier to entry and ensured the fellowship application does not require a fully-constructed puzzle. Instead, applicants can submit “a theme set with theme clues,” a partially filled 15×15 grid, and a grid as small as 7×7.

Under Mason, the Times has also launched a new weekly column to give insight into puzzle answers new, old, and evolving. (The first edition gave the story behind the answer to “Italian cheese city.”) Mason has also debuted a testing panel that gives feedback on puzzles and has described the panel as “a vibe check” that — aside from looking for typos and fact checking, done separately — is particularly interested in whether the puzzle is fun for a diverse group of people to play.

The crossword editors also now meet for editorial workshops where the entire team gets to “argue and philosophize” about puzzles, Mason said. “It’s a really good chance to equalize the playing field. The editors have an opportunity to talk about their point of view when editing puzzles, so it wasn’t just trying to replicate what Will would do.”

It’s in these editorial workshops that the NYT Games team asks itself, when does a word or bit of slang or cultural reference become puzzle-worthy? Those creating puzzles tend to consider whether the term has longevity (or will go the way of, say, “tubular”) and what portion of their audience can reach the right answer from reading a clue.

Editors at the Times have long checked newspaper and magazine archives to test a word’s popularity and now take online search results and social trending topics into account, too. Clues for perennially popular answers can evolve as well. Take “ogle,” for example. The New York Times recently shared that the term has been used 438 times in its crossword, but that “descriptions of the word have gone from ‘flirt,’ in 1942, or ‘gaze amorously,’ in 1994, to ‘It’s not a good look’ or ‘eye lewdly,’ in 2021.”

Mason considers herself a word game lover but doesn’t go in for streaks or solving for speed. She’ll often use AutoCheck, she says, and gets more delight from clever clues and personality-filled themes than competing against her previous times.

“Frankly, I’m Black and Puerto Rican and queer and I’m in my early 30s,” Mason said. “Any time I get to a solve a clue, and it’s something that I know, something that feels really relevant and fresh to me — whether it’s pop culture that I’m familiar with, or a food — I get so excited.”

She recalled seeing “mofongo” — a plantain mash — in an unpublished puzzle, and said she’d love to see more submissions with specific points of view.

“I’m really on the lookout for removing fill that’s not relevant but we just decided everyone needed to know,” she said. If clues can be solved with “a ventriloquist from the 70s,” why can’t they also be solved with answers taken, say, from early aughts rave culture?

Sharing more of this behind-the-scenes work is just one part of a larger plan to, as Mason says, “beef up” the “context and storytelling side” of NYT Games as it seeks more daily solvers and more subscribers. (Games, along with Cooking, passed 1 million subscriptions in 2021.)

“A lot of what I have been working on is building what I call an ecosystem. I don’t want people to just come in and play a game and leave,” Mason said. “Our games provide not just an outlet for people to self soothe and improve their mental health, but also to connect with other people. You want to provide a place where people can get a full experience.”

“It’ll be a lot of throwing spaghetti at the wall in the beginning,” Mason added. “But I hope that it helps us connect with new and different audiences who are maybe interested in nerdy things like etymology or the culture behind different words and slang — and brings them in in this other way. That’s something I’ve been really mapping out and working towards.”

I had to ask Mason about Wordle, that dead-simple-but-seriously-addictive game that has been clogging up your Twitter feed with gray, yellow, and green squares. Mason said that she hadn’t played the game yet herself but that everyone else on the Games team was giving it a go.

“I think if anything, it has sort of lit a fire under my ass because it’s clear people are hungry for more games, more word games, and more unique games,” Mason said. “I’m excited to hopefully work on new games in the future so it’s really inspiring.”

Sarah Scire is deputy editor of Nieman Lab. You can reach her via email (, Twitter DM (@SarahScire), or Signal (+1 617-299-1821).
POSTED     Jan. 11, 2022, 3:50 p.m.
Show tags
Join the 60,000 who get the freshest future-of-journalism news in our daily email.
“The way we raise the money at The Guardian is different than any place I’ve ever been”
“This is truly a jointly owned responsibility among the business side and editorial.”
What’s with the rise of “fact-based journalism”?
“To describe one form of journalism as ‘fact-based’ is to tacitly acknowledge that there is also such a thing as ‘non-fact-based journalism.’ And there isn’t.”
Britney Spears and the generational shift in celebrity coverage
“There was just this nastiness that emerged in the way celebrities were covered in the 2000s.”