All products featured on WIRED are independently selected by our editors. However, we may receive compensation from retailers and/or from purchases of products through these links.
These days, many newspapers and magazines are offering online versions. Many puzzle solvers might wonder what will become of the crosswords and other challenges featured in many of these publications. Will the solving experience translate to an online format as well as reading does? The people at Puzzazz, including founder Roy Leban, certainly think so, and since 2008 have been working hard on cultivating quality puzzles and developing TouchWrite, a groundbreaking new technology for doing puzzles on the screen.
Wired: How did Puzzazz start? What gave you the idea to start the company?
Leban: I’ve been creating both puzzles and software since I was a kid. As a puzzle guy, I’ve been published all over, including a bunch of crosswords in the New York Times. And, as a software guy, I’ve founded or been a key employee at seven previous startup companies.
It seemed to me that there was an opening for a company selling great puzzles — rather than games or puzzle games — but I didn’t know what it was going to look like. Since the best way to learn is to start something, I decided to start Puzzazz in 2008. I started off small, and I ran experiments. We’ve had a puzzle of the day, puzzles running 24 hours a day on Twitter, and super-sized puzzles on Sundays. We’ve had ads on the site and we’ve recommended puzzle and game related products you could buy on Amazon. And I did a bunch of market research to learn as much as I could.
We learned that people are willing to pay for high quality puzzles, so that’s what we’re doing, in our puzzle store that’s currently available on iPad and iPhone. We also sell some puzzle ebooks for Kindle. And we applied what we learned about other things. Our web site has no ads, and we’re still running the Puzzle of the Day. We have a few users who have solved it every single day for over four years. That’s loyalty!
Wired: How do you cultivate your content?
Leban: Our goal is world-class puzzles and to get that we need world-class authors. I’m fortunate to have a lot of contacts in the puzzle author community, so our initial authors [including prolific constructor Brendan Emmett Quigley, cryptic compilers Joshua Kosman and Henri Picciotto, and Pre-Shortzian Puzzle Project founder David Steinberg] are a mix of people we contacted and people who reached out to us when they heard about what we’re doing. Most of our content comes from these outside authors.
One of our core principles is that the entire community — both authors and solvers — will benefit from a better puzzle ecosystem, so we also spend time helping authors who aren’t quite there yet.
Wired: Tell us about your technology called TouchWrite. How did that come about? How does it work?
Leban: TouchWrite came about when we were looking at our first prototypes for iPhone and Android phones. The screens are just so small, and it’s hard to fit everything in, especially with a puzzle the size of a crossword. We looked into a lot of other approaches, like showing the grid and only one clue, or showing the clues and not the grid. There are some other companies that have taken those approaches, but we found that that they detracted from the experience of solving. We know that solvers jump around in the grid and they go back and forth between Across and Down clues, and we want solvers to think about solving, not our interface. With TouchWrite enabled, you can see an entire crossword, plus a bunch of clues, even on an old iPhone.
We collected handwriting samples from a wide variety of people — men, women, young, old, right-handers, left-handers, people with long fingernails, those who write with a stylus, and even some samples from people for whom English isn’t their first language. It’s an arduous process as we gather thousands of individual samples from each person. We feed all the samples into an analysis system which uses some clever proprietary algorithms to create a compact and fast recognizer. It’s extremely fast because all the hard work has been done in advance. While it’s very accurate on a wide range of handwriting styles, we prioritized speed over accuracy. A slow recognizer would be unusable no matter how accurate it was, while people can easily learn to write a few letters differently (usually just neater) if they’re having problems. As we collect additional samples, the recognition quality will continue to improve over time.
It was a big investment, but it’s worth it when we see how well it works for users.
Wired: What happens at Puzzazz day-to-day? How much do you work on creating and polishing content, refining the technology, etc.?
Leban: I wish I could say that I spend all day thinking about puzzles! A very important realization for me was that we’re a technology company that really understands puzzles at our core, not just a technology company or just a content company. We’re building the best way to sell, deliver, and solve puzzles. So, overall, we spend more time on technology than content. Most of the time we do spend on content is working with outside authors, editing and converting puzzles, etc.
I personally juggle my time between improving and expanding our technology, working with incoming content, marketing and business development, plus the grunt work somebody has to do at a startup, like stocking the office fridge with drinks. I do that so everybody else can stay focused.
Unlike many startups, we do not have a foosball table. But we do have lots of puzzles and games in the office.
Wired: What’s next for Puzzazz, with regard to both content and technology?
Leban: We are amazingly busy. We’re launching new content all the time. In December, we launched two new titles — Out of Left Field, some collected cryptic crosswords from The Nation, and Off the Beaten Path with Virgilius, some British-style variety crosswords. We’re improving our engine on iOS, so we can support additional puzzle types that we don’t support yet, and we’re also working on Android and Windows 8.
Right now, we’re in the midst of a fun campaign on Kickstarter, the crowdfunding site, for a year of puzzles — Unique Puzzles for a Yankee Echo Alfa Romeo. The campaign features something that’s never been done before on Kickstarter — the campaign itself is a free mini puzzlehunt, with many explicit puzzles and a few hidden puzzles in the story and updates.
