The logical history of the 'Zoombinis’

An anthropomorphic cliffside in dire need of some Zyrtec. Three slick-talking tree stumps with a virtually indomitable appetite for pizza. A technicolor wall of mud; a stone lion with a magic paw; an impatient Cajun ferryboat captain. And sixteen azure beans with sunglasses, roller skates, springs for legs, and nearly blue-black hair.

If any of these descriptors jog your memory, you might have played Brøderbund's bestselling educational-veiled-as-pure-adventure computer program The Logical Journey of the Zoombinis as a child. And unlike those exhaustively annoying memes, it kind of is something only '90s kids will remember.

The game begins on Zoombini Isle, where 625 Zoombinis live in peace producing "small useful things that were prized the world over," such as paper clips and those strange hard bindings at the tips of shoelaces (ed’s note: they’re called “aglets”). But the Zoombinis' harmonious existence is promptly disrupted, and they flee their native soil in a storyline I've since learned has purposefully heavy Socialist undertones. The Bloats — another word for "suits," if you will — overrun the isle after promising to advance the Zoombinis' lives through modernization of their manufacturing processes.

Instead, the Bloats cancel holidays, steal profits, and increase the Zoombinis' workload, compelling them to take matters into their own hands (or lack thereof) and get the hell out of Dodge. At this stage, it's your duty as the player to shepherd the miniature blue refugees to freedom — a charming place called Zoombiniville where the Zoombinis are free to rebuild their formerly utopian commune.

There are three stipulations to the original game: First, you can only bring 16 Zoombinis at a time, tweaking them with five available traits and four attributes, before you set sail (it's possible to literally create 625 completely individual Zoombinis). Second, you must steer your flock through nine of 12 puzzles and through four increasingly seedy areas with names like “The Mountains of Despair,” and “Deep, Dark Forest.” Third, there are no directions.

With its four levels of difficulty, the game is undeniably pretty damn complicated (co-creators Scot Osterweil and Chris Hancock set the age range for play from ages 8 to 100). Each time you reach Zoombiniville with a new herd, the game constructs a building they require to prosper on their unfamiliar territory — a library, a general store, a bowling alley, or a paperclip museum, to name a few. The more you play, the more complex the puzzles become, and for an excellent reason: to usher all 625 Zoombinis to Zoombiniville, one must run through 40 perfect games. That's a lot of pizza making.

Zoombinis is a game so enchanting, unique, and bizarrely ahead of its time that no game has since come close to replicating it. Simply put, If there were a Criterion Collection for games, Zoombinis would be in it — and likely pretty high on the list. After all, what other cultural '90s computer game was newly resurrected and re-released with the help of $101,716 in Kickstarter donations that more than doubled the initial goal? But something I've often mulled over with friends who have similarly joyful memories of the game is Where did the Zoombinis even come from? The sheer eccentricity of a program loaded with wholly genderless, anti-racist, pro-immigration, Marxist overtones raises the question: How did any of this happen at all, let alone in 1996?

Tabletop Jr.

In the early '90s, software developer and Zoombinis co-creator Hancock created the data visualization tool Tabletop at TERC (Technical Education Research Centers), a nonprofit in Cambridge, Mass., Hancock designed the software for grades four through 12 and allowed students to play with data as small icons arranged into histograms, box plots, cross-tabulations, Venn diagrams, and scatter plots. The point was to encourage children to become comfortable with and visualize data, roughly a decade before "data visualization" was a commonplace expression.

Zoombinis' co-creator Osterweil was working a few floors below Hancock at the TERC office. The two weren't friends prior, but TERC had brought Osterweil into the business to add art to their computer programs. "Only a few years before, PCs were just green or amber text on screen," Osterweil says, "and they started needing graphic interfaces, so I was hired."

“Only a few years before, PCs were just green or amber text on screen.”

Hopeful over Tabletop's success within the company, Hancock then developed Tabletop Jr., software marketed to youths in grades kindergarten through sixth as a "playful and powerful math tool for representing, exploring, and graphing data.'' It was in Tabletop Jr. that the Zoombinis were born — Osterweil's creation — albeit a crude, flattened version he'd initially christened "Snoids," in homage to legendary cartoonist Robert Crumb. But Osterweil's Snoids weren't defined yet; they represented numerical and categorical data to be arranged and cataloged by children the way the Zoombinis would subsequently be sorted — by hair, eyes, and means of mobility.

The problem was, neither development was an actual product yet. To Hancock, they were mere "research prototypes" he needed to get to market. All he needed was a good publisher.

Brøderbund

Hancock went to California to meet with Davidson & Associates, a developer of "edutainment" software responsible for the educational Math Blaster series and later publisher of Math for the Original World. He likewise met with Brøderbund — possibly the most reputable gaming company of the '90s, with hits like Carmen Sandiego, Prince of Persia, and Myst.

He initially favored Davidson — whose president was so passionate about Tabletop he traveled to Cambridge after their introductory meeting — claiming they "probably would have done a better job." But it was Brøderbund executive producer Laurie Strand who ended up scoring a deal with Hancock and Osterweil.

"Her teenage daughter was in the office during a family event and started playing with the Snoids and all the combinations you could make with them," Hancock recalls. After witnessing her daughter's reaction to the tiny blue creatures, Strand quickly returned to Hancock and inquired if he and the TERC team could generate a game centered on the "very appealing" Snoid characters. If Hancock agreed, Brøderbund would publish Tabletop and design and distribute the game focused on Tabletop Jr.'s Snoids.

Hancock returned to Cambridge to move Osterweil on the idea. It was an easy sell. Brøderbund paid $1.5 million for the game's development, allocating $200,000 to TERC and paying Osterweil and Chris their modest salaries in the mid-$40k range.

Zoombinis!

Hancock and Osterweil spent months generating ideas throughout the incubation period, each bringing something the other lacked to the table. It was a melding of the minds: Hancock brought his left-brain, analytical, puzzle-solving intellect to every meeting. Osterweil, whose artistic background was mainly in theater and television, would arrive prepared with maps, visuals, and storylines floating around his head.

Osterweil drew the map and forged the names for each destination the Snoids would need to pass through to reach Zoombiniville, offering tribute to J.R.R. Tolkien along the way. "Tolkien is the granddaddy of world-building and heroic quest journeys," he says adding that the creatures "are kind of hobbit-like in that they're smaller than everybody else, but they're smart — and so resourceful." Hancock and Osterweil knew they wanted four journey stages, through four distinctive regions, all with logically cohesive themes like logical constraints, dimensionality, adjacency puzzles, and algebra. The two would meet closely and fling around puzzle ideas until one stuck.

They piloted each puzzle they designed to elementary school students at Fletcher Elementary School in Cambridge, where a teacher had already been kind enough to assign his pupils a project to draw what they thought a Snoid was and how one acted.

"We demonstrated the puzzle with kids first," Hancock says, adding, "you do paper prototyping with these types of things when you can." Hancock and Osterweil would produce images of the characters, and students would create their own personalized versions, determining from the Snoids' individual attributes.

Before Brøderbund insisted on the Allergic Cliffs in the first Zoombinis puzzle, the co-creators had developed the same game with trolls at the bridge. "I'd present them with their pictures of Snoids and then the trolls, and I'd act it out with these grumpy voices," Hancock recalls. The students would approach Hancock with a guess, just like in the game, and he’d either send them back or allow them to pass through. Hancock and Osterweil tested everything on paper — and in person with the kids — before handing them over to TERC for development.

In the infancy of the games' progress with Brøderbund, Osterweil and Hancock realized they couldn't use Snoids due to Crumb’s copyright. According to Osterweil, the co-creators had been "fixated on a punchy, one-syllable name." After tossing around the excessively obvious Snoids' subsidiary 'Snoods' for a period, a naming contest commenced within the company. Product director Dennis Leahy's suggestion "Zoombinis" stuck.

"Zoombinis made sense," Osterweil says. "They're these little beanies, and they're zooming — and they're all heads — they're brainy, they're inventive, they're clever." (By pure coincidence, a game was distributed in 1996 — the same year as Zoombinis — called Snood, which featured colorful critters that were also just heads.)

Twelve of 15 puzzles ended up making it into the game, one of which — Mudball Wall — Brøderbund initially developed into a "schematic prototype," which was a hit with TERC faculty. This was a tell-tale sign of success; the program was challenging and playable even without the story, characters, and journey — the "sweetener," as Hancock calls it. Then the meat was added. While Brøderbund commissioned 12 actual paintings for the backgrounds of the scenes, their programmers helped Hancock build algorithms for each puzzle, supporting infinite replayability at four-difficulty settings.

“It's this persisting in the face of these arbitrary rules adults always impose that kids don't understand.”

Osterweil, whose politics were (and remain) leftist, wished to portray an underdog story with Marxist overtones — the Zoombinis were intelligent creatures living communally when their native shore was overtaken by Bloats. Following their flight, they repeatedly find themselves in desperate straits. They must submit to the world's harsh and unfair rules until they arrive at a utopian paradise where they can work cooperatively to reconstruct their commune. It's crucial to mention that the Zoombinis are kids, "knee-high to everyone they meet," as Osterweil says, and both the Bloats and mostly every other creature — good or bad — they encounter on their journey are adults.

Suppose you play Zoombinis now, as an adult — you'll immediately notice the bigotry of the Allergic Cliffs and the Stone Cold Caves, demanding the Zoombinis sort themselves by their physical attributes. Hancock claims this was purposeful. “The Zoombinis were always kids, and always underdogs," he says. "And they're indomitable; they keep trying. The rules aren't their rules — they're not interested in sorting themselves — it's the adults in the world who keep sorting them. It's this persisting in the face of these arbitrary rules adults always impose that kids don't understand."

Marketing woes

Hancock claims the central drama in the development process came from Brøderbund's marketing team insisting on two things: that the Zoombinis be "Disneyfied" into cutesy, rosy-cheeked creatures in line with popular '90s animation, and that the product be packaged as solely educational rather than entertainment.

As they developed it, the game doesn't allow you to name your 16 Zoombinis before setting sail. But when Hancock and Osterweil were testing the Zoombinis as paper prototypes at Fletcher Elementary, the naming process was a big part of the "emotional interest," as Hancock calls it. "They would say stuff like 'These are my babies' or 'This is my family,' and give them all names."

Osterweil originally developed a name database that pushed back against gender stereotypes; "We wanted kids to get over thinking the Zoombinis were either boys or girls," he says. "The database would make it so you could generate a name for a Zoombini with a crew cut that would end up with the name 'Betty,' or one with a ponytail would be called 'Mack.'"

The issue was one of Brøderbund's programmers had created his own program to generate names at random — the nonsensical monikers assigned to the Zoombinis in the game — and the co-creators felt too guilty to fight against it. "He was so proud of himself," Osterweil recalls of the programmer. Hancock adds, "It wasn't the most important battle to fight."

But they argued against Brøderbund's Disneyfication (and Westernization) of the Zoombinis. Both co-creators were adamant that the Zoombinis remain enigmatic. For one, less regard to detail would "help you cognitively to pay attention to everything going on in the game and maintain interest in what's going on," and also because "kids have their own imagination," Hancock says. The more minor the detail, the more likely a child would project themselves into the game. "Kids would create their Zoombinis and be like, 'This is me!'"

The first draft of the Zoombinis that would appear in the game was, according to Hancock, so cartoonish he "almost quit the project.” He and Osterweil created versions of the creatures out of Fimo — a popular brand of polymer clay — and Hancock penned an essay to Brøderbund explaining why their interpretation of Zoombinis wouldn't work. "I used Buster Keaton and Bugs Bunny as two examples: I compared how much more successful Bugs is to Roger Rabbit. Roger is so aggressively cutesy, whereas Bugs Bunny is kind of a man of mystery. He's always cool. And Buster Keaton, he never cracked a smile, but you were always projecting onto him." Brøderbund agreed and met the co-creators halfway. But then they started introducing problematic hairstyles.

“If you give them red and blonde hair, you're saying they're Caucasian, you know, like they're European. I did not want to suggest these were white kids."

"They came back with all these different hairstyles, and they were red-headed or blonde," Osterweil groans. He says he'd purposefully decided their hair would be dark, and it wasn't up for debate. "Most of the world has black hair,” he insists. “If you give them red and blonde hair, you're saying they're Caucasian, you know, like they're European. I did not want to suggest these were white kids." Brøderbund agreed to meet Hancock and Osterweil halfway on this one too, which is why the Zoombinis' manes aren't entirely black, and instead more of a dark blue.

Hancock and Osterweil's last fight was over the marketing. With Zoombinis, Hancock says they "caught a wave at the right moment, when there was tremendous hope over what was possible on CD-Roms," and before Brøderbund was publicly traded. Brøderbund had given the creators ample resources, where a few years prior, the game may have never been greenlit, and a few years later, the world wide web would become a universe of endless possibilities and unlimited online games. CD-Roms would soon join their predecessor — the floppy disk — in technology archives.

"When we pitched our game, Brøderbund was already being beaten down by market changes," Hancock says, alleging the marketing department refused to release it as anything other than educational. Their priority was that it would fly off the shelves at Walmart, where computer games had to be labeled precisely for what they were — academic or entertainment. The team's first imposition on the co-creators was the name: The Logical Journey of the Zoombinis.

"We hated it. We absolutely hated that title." Hancock says. “We didn't think of it as education. We just thought of it as entertaining."

"It was the death of interesting computer games," Hancock says of Brøderbund's insistence to prioritize educational marketing for Walmart's acceptance. Osterweil adds, "Then they took Zoombinis and a couple of other educational games and crammed them into this Active Mind Series."

Internet explosion

After nearly three years of development, the game was finally released in March 1996 as part of the Active Mind Series with the aforementioned title The Logical Journey of the Zoombinis and sold in software stores like Egghead and, of course, Walmart. And it sold reasonably well — “if somebody was looking for interesting software, they would go into a store and pick up a Brøderbund game; that name meant something at the time,” Hancock recalls — but neither co-creator was happy with the name or the marketing, believing the word “logical” was too academic sounding to draw children in. “We wanted it to be an adventure game,” Osterweil sighs. However, the educational packaging increased Zoombinis’ reign in elementary and middle schools, where kids would play it, come home, and immediately ask: Where can we get this game?

But by the time students wanted to bring Zoombinis home, Brøderbund was embarking on their own journey, with no quaint Zoombiniville ending in sight. After attempting to acquire The Learning Company (TLC) in 1995, TLC turned around and enveloped Brøderbund in 1998, consequentially firing over 40 percent of its staff just two years after the release of Zoombinis. A year later, TLC and Brøderbund were swallowed up by Mattel in a $3.6 billion deal that's gone down as one of the worst business agreements in history. But Osterweil and Hancock's reign at Brøderbund prevailed as Disney tried — and failed — to poach both Osterweil and Hancock. Despite Brøderbund’s inevitable demise, The pair developed two sequels: Zoombinis: Mountain Rescue in 2001 and Zoombinis: Island Odyssey in 2002.

In 2001, however, Brøderbund was sold again, this time to Irish publisher Riverdeep Interactive Learning. In 2006, precisely one decade after releasing the original The Logical Journey of the Zoombinis, Riverdeep bought the educational publishing house Houghton Mifflin Harcourt (HMH). And as Zoombinis’ code deteriorated, CD-ROMs fell out of style as people favored internet-based games, and its compatibility with new computer software suffered, HMH would hold onto the game for nearly a decade. And until the mid-2010s, the game’s software was orphaned, much like the Zoombinis themselves. But unbeknownst to HMH, whose marketing team allegedly never checked until the last minute, the original game and its two sequels had sold over one million copies.

Kickstarter

While Osterweil and Hancock had both moved on from TERC and into high-level positions at MIT (Osterweil as the creative director of the Education Arcade and Hancock, who earned a PhD at MIT Media Lab), a few folks at TERC who had either persisted through the changing of the guards or joined sometime around the co-creators' departures were interested in relaunching Zoombinis.

By 2012, TERC's CTO David Libby, with the company for over 20 years, and Jodi Asbell-Clarke, an astrophysicist in charge of TERC's Educational Gaming Environments group who was at TERC when Zoombinis was first published, knew the game would do phenomenally well if re-released. They had a steady stream of fan mail coming in from people saying things like, "I played Zoombinis as a kid, and now that I have kids, I want them to play it!" Libby and Asbell-Clarke were already well aware of the millennial cult fandom that Zoombinis had accumulated over the years, with groups of thousands dedicated to the game on Facebook and Reddit. The only problem was the pair didn't have access to any of the assets. They also had to convince TERC to fund the relaunch, which would take nearly two years.

With the help of an eagle-eyed TERC lawyer, they wrote HMH on a failure to publish clause Hancock had negotiated more than a decade prior. In early 2014, just around the time TERC had agreed to fund, Libby got a package in the mail. “There was a notice saying, ‘Here's Zoombinis, and now you have the legal rights back,’” he says. With the legal rights, HMH included the original assets — 20 gigabytes’ worth. TERC now had all three Zoombinis games, along with the graphics, audio files, and game codes — although, as Libby says, Brøderbund's in-house tools used for the game "weren't around anymore."

The Boston educational media studio Fablevision beat out four other developers in a bidding war for Zoombinis’ relaunch. TERC hadn’t immediately invited Osterweil and Hancock back (which Osterweil admitted was upsetting), but their return was baked into the FableVision contract as necessary. TERC also gained the aid of Learning Games Network in the bid, a company devoted to “identifying and partnering with organizations that are working to create, research and disseminate game-based learning tools,” according to their website.

TERC launched A Kickstarter in February of 2015 to gauge interest and seek funds to push the project forward. With rewards like Zoombini plushies, poster-sized maps, collectors pins, hoodies, and more, the company doubled its original goal of $50,000 — 3,424 backers donated just over $100,000.

Libby tells me the $101,716 did not pay for the redevelopment. "The redevelopment was well over half a million dollars,” Libby says, “and we footed the bill." He adds that the primary funding came from within TERC, not FableVision. The Kickstarter money was used to improve artwork and ensure the relaunch was compatible with iPad, Android, and Amazon tablets in anticipation of the tablet boom.

$2,127,636 study

With funding, Libby and Asbell-Clarke set up a fan advisory board made up of diehard fans (yes, there are Zoombinis super fans who advised Libby against changing characters names because they already had them tattooed on their skin). They would send anything new to the advisory board to test before okaying it. "We wanted things to stay very close to the original, but we also wanted it to look like a game that people would play in 2015," says Libby.

The game was nowhere near publishable, but small strides were being made. They brought artist Chris Cyr on to recreate the original paintings. FableVision slowly picked the game apart — with Hancock's help creating new algorithms and Osterweil's help undoing some things he disliked most about the original — and put it back together. One of the first things Osterweil did was reshape the Zoombinis back to their original Fimo-based prototype: more elongated bodies with an option for the black dot eyes Osterweil and Hancock fought for in the early '90s. He also added more inclusive hairstyles and removed the ability to create twin Zoombinis in the customization stage.

Osterweil also removed instructions Brøderbund had later added to the game following its release, arguing that "they made things too easy." Asbell-Clarke, whose work was on the education research side, later went back in and added minor instructions she thought would make the game more approachable to a new, young audience unaccustomed to the "hunt and click" computer games that defined the '90s. The team also reduced the number of Zoombinis needed to finish the game, from 625 to 400.

A year before the game officially relaunched, TERC's board asked Libby and Asbell-Clarke what proof they had that Zoombinis actually teaches anything, focusing on computational thinking, or CT (this term wasn't thrown around in the '90s as computers were more of a hobby than a learning force). They didn't have an answer, so they wrote a proposal to the National Science Foundation to research how the game helps kids perform CT. TERC was awarded $2.1 million to lead a team in four years of educational data mining, with Asbell-Clarke at the helm.

"We first looked at 70 to 100 kids, screen-recorded their Zoombinis gameplays, and had researchers label or code things for systematic testing, trial and error, implementing a partial solution, or implementing a full solution in all these puzzles," she tells me, adding that "you can definitely tell when a kid is doing computational thinking or is not being systematic." After a while, the team tested kids who had been playing Zoombinis more with external CT tests. "Lo and behold,” she says, “the more kids played Zoombinis, the more they grew on their measured pre- and post- of a CT test."

Asbell-Clarke has also studied the link between CT and neurodiversity. Her studies pointed toward kids with autism, ADHD, and dyslexia scoring higher at Zoombinis than other games — and better than their neurotypical classmates. "Teachers would come up to us repeatedly and say, 'It's the kids who never do well in class' or 'It's the kid who hasn't talked all year,' where those students are becoming the class leaders because they can solve problems in the game the other kids can't,” says Asbell-Clarke. “And suddenly their whole identity changes and the way their classmates look at them changes."

“You can definitely tell when a kid is doing computational thinking or is not being systematic.”

One of the critical elements of why the game is so life-changing among neurodiverse kids is that unlike standardized tests or games with time-limit or "game over" as a setback, Zoombinis has neither. There is no time limit, and if you lose a Zoombini, you find them again at your last resting place, which was very purposeful messaging from Hancock and Osterweil — there is no death, and with each setback, you still make progress. TERC continues to push elementary and middle schools to implement the game in classrooms as an alternative approach to neurodiverse students' studies.

So what makes The Logical Journey of the Zoombinis work so well at capturing the hearts and minds of children? It's a combination of multiple elements: complexity, story, intention, and encouragement. The game is complicated enough to stump even the cleverest of adults, so young players never feel talked down to. The story and the world built around it is fascinating enough to hold the attention of any adventure-seeking kid. The intentionality behind creating the Zoombinis as underdogs discriminated against by authority figures allows kids to empathize with them as they guide them along. And the way the game is constructed with no time limits, no deaths, no game-overs, and the ability to rescue any Zoombini left behind motivates players to shepherd them to their final destination over and over again.

There's no failure in Zoombinis, no starting over. Mistakes advance you rather than make you feel incapable or lacking in ability. The game was designed to allow players to exist in a state of partial understanding while still growing. "Sometimes you have to make mistakes to learn," Libby says. "And that's the message of Zoombinis. Mistakes contribute to your knowledge; they're not necessarily a bad thing."