This is a portion of the eBook at doi:10.7551/mitpress/14266.001.0001  Downloaded from http://direct.mit.edu/books/oa-monograph/chapter-pdf/2478089/c005300_9780262380553.pdf by Intv Prime 

Becoming a Videogame Programmer at Mattel 

 In the previous chapter we discussed Intellivision's platform culture, within which programmers learned what Intellivision could do. This included exploring and shaping the platform and even creating tools for game development.­ These practices can be understood as what Aphra Kerr has termed a "production culture," invisible to the consumer plugging a cartridge into their console.1 All platform cultures include production cultures. Investigating production cultures is crucial to platform studies since a "critical reflection of video game production can uncover the economic, cultural, and ­political structures that influence the final form of games."2 Moreover, closely examining production cultures and the "design history of video games," as Raiford Guins called it, helps account for "the full spectrum of design practices."3 Moving from programming to programmers, this chapter extends our analy­sis of Intellivision production culture by examining how programmers became videogame designers in "Applications Software," the group that arose within Mattel Electronics to make new titles for Intellivision. We explore how designers were hired and learned to make games, given that very few had prior experience since videogames were simply too novel. Making Intellivision games involved not just new work but a new workplace that was the center of development: in this mainframe era t­here were no desktops or laptops, and programming took place almost exclusively in the office. Contrary to the myth of the lone programmer, from the outset, this work involved emerging norms and dynamics of collaboration. In the Introduction, we discussed how Intellivision was s­ haped not only by prior videogame systems but also by Mattel's history of designing and manufacturing toys. In this chapter, we expand this analy­sis by tracing how this history ­shaped the social relations and everyday practices of Applications Software programmers. As a legendary "funny farm" of creativity, Mattel's Preliminary Design department for incubating new toy ideas was a key pre­ce­dent for Applications Software (chapter 1). Preliminary Design provided a template to foster brainstorming, an ethos of work as play, and procedures for selective engagement with man­ag­ers and marketers. When Mattel Electronics began separating from Mattel's Toy division in 1978, it emerged from Preliminary Design and included a new department called Design and Development (D&D). Richard Chang, previously a Director in Preliminary Design, became the head of this new "department of 30 engineers, programmers, industrial designers, and writers" responsible for handheld games and Intellivision software.4 A few years ­later, Applications Software would emerge from D&D. 
 
 Eggs in Mattel's Basket 

APh Technological Consulting (APh) played a crucial role in Intellivision's development, a point we stress throughout this book. While we focus on the production culture of Applications Software in this chapter, we frequently compare it with APh, since APh's existence and success was the primary force driving Mattel Electronics to create its own team of programmers. As Intellivision was developed and brought to market, APh was the center of game design for the console. All of the original twenty cartridges listed in Intellivision's 1980 brochure were programmed at APh, and the consulting firm continued to produce many games ­under the Mattel Electronics brand. However, this dependence made Mattel Electronics executives increasingly uncomfortable. Heather Edison (later Brendel), who eventually became Chief Executive Officer at APh, emphasized that Mattel Electronics had begun "to see that we were getting smarter, that we were getting very close to saying, ‘From now on it's ­going to be not $12,000 [per game] and ­we're done. It's g­ oing to be $100,000 and royalties." When Mattel Electronics sought to internalize videogame development, Edison recalled one APh meeting with Mattel man­ag­ers where "they started by saying, ‘We just want you to train our programmers.' They ­didn't say, ‘We want to cut our relationship with you' so much as they ­were saying, ‘We're too dependent on you. We'd like you to share your technology.'" The earliest Mattel programmers spent considerable time in APh's offices since t­ here were not yet tools for videogame design in D&D at Mattel. Brian Dougherty, hired at Mattel before Intellivision was released, recalled, "­There was prob­ably a four-­or five-­month period that my job was driving [168] to APh." Glenn High­tower, ­founder of APh, allowed Mattel's visiting staff to use their equipment, but for the most part t­here was no direct collaboration. For example, Mattel Electronics programmers Mike Minkoff and Rick Levine worked on PBA Bowling at APh, soon joined by John Sohl. Sohl recalled that they "sat in a room by ourselves at the far back of the building and ­people would be polite, perhaps. Not r­ eally effusive, and they wouldn't be wandering in and out talking shop. So we were sort of in Siberia ­there in APh." Despite such separation, circulation of ideas could occur. For example, APh staff Larry Zwick was programming an Asteroids clone a few desks over from Sohl, who eventually designed Astrosmash (based partly on Asteroids). At this point Gabriel Baum was working in London, developing language education programs for the Atari 400/800 home computer. Seeking additional customers for his business, Baum met with Josh Denham, President of Mattel Electronics, to discuss Intellivision. Baum recalled, "I went on a rant about the prob­lems of managing the development process for creative ­people, how dif­fer­ent it is from managing the ­process from an engineering perspective." Baum added that he must have been persuasive ­because Denham recruited Baum to join Mattel Electronics in January 1981 as head of the "Intellivision Programming Group." By the end of the year, this group was known as "Video Applications Software Programming," commonly shortened to "Applications Software."5 Baum recalled that "the issue for Intellivision management was that all their eggs were in one basket, and that was APh. They were very anxious to form an internal development group." Baum initially reported to Chang within D&D, yet since Chang directed many operations and Baum had specific experience with managing software development, Baum's hire made sense. As Intellivision programming grew at Mattel Electronics, Baum pushed for autonomy. APh programmer Tom Loughry noted that around 1982 Baum replaced Chang representing Mattel Electronics at meetings with APh, and in April 1982 Applications Software officially broke away from D&D to become an i­ ndependent group.6 Chang had emerged from Preliminary Design managing the creative development of toys and was a key force in extending this institutional heritage to Intellivision. Baum's cultivation of Applications Software built on this tradition, exporting a paradigm of playful, protected, freewheeling innovation to videogames. Baum emphasized that Denham was "very receptive to my approach to creating ‘a com­pany within a com­pany' as I sometimes put it. I suspect that cultivating creativity in a Preliminary Design-­type group was in the DNA of the com­pany, and I was able to build on that approach."7 However, Applications Software's autonomy signaled growing ­organizational conflict as electronics became further separated from Mattel's institutional history with toys.8 6 

  
Even while Mattel Electronics internalized game development, work with APh continued. In May 1982, while Intellivision was booming, ­there ­were twenty-­five titles in pro­gress, and twelve of t­hese—­almost exactly half—­were ­under development at APh.9 The production culture of Intellivision game design was thus dually located, extending to APh and its Caltech roots (chapter 1), and also within Mattel Electronics. Hiring the Creative Team 

Bringing videogame design within Mattel Electronics meant hiring and managing ­people for a job category that in large part did not yet exist. This ­process was deeply tied to Mattel's location in Southern California, with its nexus of engineering and entertainment. "Southern California is a land of unimaginable variety, beauty, and excitement," read Mattel's promotional recruitment materials, an enticing location associated with sunshine, entertainment, and fun.10 The marketing researcher Craig Spitzer noted that "Los Angeles sounded like heaven on earth" when he was recruited from Chicago, a point echoed by other marketers coming from the Midwest and East Coast. However, for programmers and engineers, a common pattern extending back to Mattel's early years was for the playful world of toys and games to fuel a brain drain from aerospace, particularly military aerospace contracting. Many programmers and engineers e­ ither came from that domain or joined Mattel rather than pursue a c­ areer in aerospace engineering. Most prospective Intellivision programmers already lived in the area. For instance, Stephen Roney was raised in Santa Monica and Eddie Dombrower in Long Beach (Dombrower recalled driving by "the old Mattel sign on the 405 freeway. It was an icon"). A majority of Applications Software programmers received undergraduate degrees from Southern California institutions. A few majored in computer science, but since the field was new, many received degrees in electrical engineering. Moreover, many programmers ­were in their early twenties when hired, often as their first full-­ time job. In the world of con­temporary videogame design, programmers are also often in their early twenties when hired. However, during the second generation, the very idea of programming games was in its "youth." One reason many programmers and engineers were from Southern California is that Mattel Electronics primarily recruited for t­ hese positions via local newspaper and radio advertisements. Many advertisements did not even mention experience in game design or programming, offering instead the chance for engineers to use "creativity" as well as "a lot of enthusiasm, imagination, and energy."11 For example, in 1981 Roney was working at Hughes Aircraft, and Mattel Electronics was headquartered on the other [170] side of the freeway. A sign in front of the building stated they ­were seeking Intellivision programmers, and Roney de­cided to apply. Concerned that six years at Hughes Aircraft would imply the need to pay him a high salary, Roney said, "I took my resume and crossed out my top-­secret clearance and wrote that I had a hundred board games."12 However, the initial screening process focused on technical ability, not game design. Programmer Bob Newstadt recalled how "Mattel ­couldn't hire ­people who had already done home videogames. It was too new." Mark Urbaniec, another Mattel programmer, confirmed that "we didn't hire game programmers. We ­didn't hire game designers. Those ­didn't exist. You are hiring somebody who worked on a guidance system for a missile or was programming a part of a payroll program." Using what became known as the "Minkoff ­Measure" (named a­ fter the programmer and director Mike Minkoff), Roney recalled that "they gave you a block of assembly language code and asked you what it did. It was inclusive OR."13 This operation was not in the instruction set of Intellivision's CP1610 micropro­cessor, so programmers would need to understand the code to describe its function. Programming ability, however, was not the only consideration. Creativity was a valued asset. Programmer Steve Ettinger explained that he was a "total liberal arts guy" in college and failed the Minkoff ­Measure, but through per­sis­tence was able to be hired. Baum recalled interviewing other candidates who passed the Minkoff ­Measure, and "if they ­didn't come up with an arts subject they were passionate about, I lost interest." Prospective programmers needed a touch of the playful, evidence of passion in other aesthetic and cultural pursuits. Indeed, the technical structure of Intellivision's platform eased ­these hiring practices: as we described in the previous chapter, Intellivision's hardware, operating system, tools, and training made programming more accessible. Thus, Baum sought to hire individuals with backgrounds and experiences in addition to engineering, from musicians and visual artists to Shakespeare experts and Renaissance role-­ players. Dennis Marchand, Director of Recruiting, recalled Baum emphasizing that "I need to hire people who are creative and they don't sit in that engineering structure of just A + B = C." Indeed, Mattel Electronics' 1982 recruitment materials, titled "The Creative Idea," consistently emphasized the importance of creativity entwined with technical innovation. The brochure explained that R&D programmers and engineers "should be innovative and creative, with a mind that thinks ahead—to explore the f­ uture, and the unknown."14 ­These hiring practices reveal a prehistory of what is now often termed "playbor"—­the blurring of work and play—­and link to Mattel's tradition of valorizing creativity, one source of the "creationist capitalism" pervading the con­temporary digital technology industry.15 6 


"The Creative Idea" brochure also linked creativity to a diverse workforce, with the cover showing three ­women (two of color) and five men (three of color). Mattel had been cofounded by Ruth Handler. Handler was no longer with the com­pany during the time of Intellivision, but ­there ­were some female executives in Mattel Toys. ­There ­were also female staff members in Systems Software and other divisions outside and within Applications Software, including the Educational Products group headed by Mona Theiss. Moreover, ­there were a significant number of ­women working on Intellivision marketing, including women of color like Audrey Leeks and Karen Terry. As an industrial engineer working in the Operations division, Linda Bruce found that Mattel was "forward-­thinking everywhere," but "in Operations, ­women ­were not as easily advanced." Sharon Hartley, who moved from Mattel Toys to Mattel Electronics to direct marketing for handhelds and eventually Intellivision, noted that in leadership positions "­there ­were some women, but t­here was a much, much better balance in Toys. Electronics was still very much a man's world." The Product Man­ag­er Kathy Austin Lockton concurred with this assessment. Applications Software hired a considerable number of women to work on Intellivision. Some ­were programmers, for instance, Connie Goldman (Thunder ­Castle), Julie Hoshizaki (Thin Ice), Minhchau Tran (Pinball), Ji-­Wen Tsao (Shark! Shark!), and Jane Terjung ("Advanced Dungeons & Dragons: Tower of Mystery" for Atari). ­Others worked in design or coauthorship roles—­labor that also required programming knowledge—­including Peggi Decarli, Monique Lujan-­Bakerink, Michelle Mock, Karen Nugent, and Karen Tanouye (later McConathy), among ­others. Applications Software staff noted greater gender inclusion at Mattel Electronics than at positions they held afterward. For instance, Dombrower recalled, "­There ­were more ­women in that group then t­here were the following years l­ater as I moved to work with Electronic Arts and Activision. And I mean women in the production part of it." Despite this relative gender inclusion, like other videogame companies at the time, a masculine ethos predominated. This can be seen in the informal humor that circulated among at least some programmers. For example, one practical joke contained signed pictures of scantily clad women wrestlers from the adult entertainment industry, written to Mattel Electronics executives.16 An accompanying note satirized "the cause of women's liberation" and joked that the file of images would "accurately and insightfully depict the true essence of the female psyche" and "encourage and support the attainment by ­women of heretofore unforeseen levels of intellectual achievement."17 Such materials ­were couched in joking "play," reminding us that playfulness can be channeled in ways that demean, diminish, and discriminate.18 [172] 

  

   

Mattel staff members were overwhelmingly white, but ­there ­were staff and ­consultants of color, including UCLA professor Gordon L. Berry (chapter 10), Black executives like Dennis High­tower (Vice President of Corporate Strategy, who for a time was United States Deputy Secretary of Commerce in the Obama administration) and Bill Gillis. About a month ­after Gillis was hired into Mattel Electronics as Vice President of Marketing in 1981, he was contacted by Jerry Lawson. A pioneering figure in videogames and one of the earliest Black engineers in this domain, Lawson had been the leader of the development team for Fairchild's Channel F videogame system. Lawson had left Fairchild and founded Video Soft, a third-­party developer of Atari VCS games. Lawson hoped to make games for Intellivision and was seeking a marketing partner for the venture. Gillis declined but remembered staying in touch "to see if ­there might be ­future opportunities, both professionally and personally, to support each other." Gillis added they "agreed that Black executives lived a fishbowl existence—­ our ­every move would be watched closely. And should we become too rambunctious and flip out of the water, we s­ houldn't expect to be rescued and 19 given a second chance." This assessment was based in part on Gillis's own employment history. Prior to Mattel, Gillis had been at RCA and recalled a white colleague informing him that another engineer was distributing Ku Klux Klan pamphlets (a confrontation ensued). Later at AT&T, he was admonished by his superior for traveling to a meeting with a white female subordinate. However, Gillis emphasized that "with regard to Mattel, I found them to be merit based and with more people of color executives than most." Emphasizing that "Mattel was indeed proactive in terms of hiring ­people of color,"20 Gillis "made it known to ­Human Resources that I intended to hire minority candidates" and brought in staff like Terry and Leeks.21 Other executives of color included Richard Chang and Ramon Fuentevilla, Vice President of Regional Field Sales. There is no rec­ord of openly LGBTQ staff, but other forms of diversity ­were represented. Gillis noted at least two staff used wheelchairs, emphasizing that "if you saw the entire staff, you could immediately recognize that ‘wow,' this is dif­fer­ent even for that time."22 Within a context of male and white normativity, t­here were forms of o­ rganizational diversity that pointed t­ oward a broader vision of the creative ­human working within the world of consumer technology. Half Stress, Half Party 

In 1988 programmer Don Daglow recalled the Mattel Electronics workspace from half a decade ­earlier, writing, "Picture a building converted from an old 6 



   

Hot Wheels factory with a second floor added to the interior to hold the game designers... A seemingly endless sea of cubicles beneath a corrugated metal roof and exposed air conditioning conduits."23 Within this space, programmers were learning to code Intellivision games and man­ag­ers ­were learning how to facilitate their work. When discussing programmers in an open-­plan office in the mid-2010s, the anthropologist Nick Seaver emphasized how "social architectures, software architectures, and physical architectures echo each other. Walking into a tech com­pany's offices ­today, one can conjure the feeling of walking into the software itself."24 The origin of ­these dynamics can be traced decades ­earlier, to environments like Applications Software. The Applications Software offices reflected the games produced—an environment of fun and play but also one of learning, work, stress, and tension. By late 1981, Applications Software had grown to seventeen programmers and a management structure was taking shape.25 Baum promoted Daglow and Minkoff to Application Software Man­ag­ers, dividing Applications Software into two groups.26 The bipartite o­ rganization was fundamental, including office layout. Even Applications Software's two softball teams ­were named the "Daglow Decles" and "Minkoff ­Measures." When ­these two groups became too large to manage effectively—­Daglow's team peaked at fifty-­two—­"Group Leaders" ­were added, each managing two to six staff members ­under ­either Daglow or Minkoff. Group Leader duties included screening new hires, training programmers, and proj­ect supervision.27 As a group leader, Russ Haft recalled "managing the game titles that you had to deal with, and the timing and reporting up the chain, helping ­people when they had issues, whether it was coding issues, debugging issues, so every­ body was sharing resources."28 ­Because Group Leaders were always recruited from existing programmers, they could help translate between designers, man­ag­ers, and marketers. After being promoted to a Group Leader, Ettinger emphasized a key contribution was that he "could talk to the other tech people and figure out how to bridge t­ hose communication gaps." Mentorship in Applications Software was often informal. This was also the case at APh. Loughry remembered "wandering around and looking at what ­people were ­doing and sitting down. ­People ­didn't know how to tweak a game, and I was helping people out." Ken Smith, one of the Caltech undergraduates hired during the summer to code games including Sea ­Battle and NHL Hockey, recalled that "from time to time Glenn High­tower (or somebody ­else) would sit down and we'd play a game and he'd have comments about what was fun, what ­wasn't fun. And then you go back and eventually you ran out of the 4K. It was good enough." At Mattel Electronics, Urbaniec recalled helping new hires who ­were typically told, "‘Now y­ ou're ­going to be designing and developing a game' and literally start with a blank slate. ­Those people [174] would look at you and go, ‘Do what?' There was no screenplay for the game, no flowchart for what the gameflow would be." Programmers were given the Killer Tomatoes training program and told to play around with it to begin coding their first game. Urbaniec recalled easing new programmers into the daunting task, "­Don't think of it as ‘Oh my God, how am I ­going to do this ­whole game?' Let's eat the elephant one bite at a time. Let's think about how the game opens. What does the screen look like? Let's make that screen." At Mattel Electronics, such informal mentorship was facilitated by a playful, open-­office environment. Hoshizaki recalled how "we could yell over the cubicle wall. Ray Kaestner was down on the other end of that row that we shared. And so he would often ask me to test his game with him. We were able to share ideas or help each other." Daglow emphasized that as a man­ag­ er, "I would literally wander down the aisle, pick a cube, recognize somebody who had had a breakthrough. I'd stop in at that station and say, ‘What you did was so cool. That was just ­great.' And because it's all cubicles, ­people could overhear." In addition to playing games in development, Urbaniec described how a favorite pastime was the arcade-­like Biplanes videogame included on the Intellivision ­Triple Action cartridge, where the programmers would "have just such a good time beating each other and learning when to use a stall to your benefit, how to use the edge of the screen to sneak back from this side to that side." Daglow added, "What was g­ reat about it is you can play a game ­really quickly, so you weren't taking too much time off. Somebody would walk by—­‘y­ ou're about to play Biplanes? I'll play you.'" Biplanes became so ­popular that a 1982 April Fool's memo warned, "Programmers are spending too much time in their cubicles playing games [and] Bi-­Planes ­will be erased from all system disks immediately."29 Programmers would also play arcade games near the cubicles (including Bump ‘N' Jump, BurgerTime, Lock ‘n' Chase, Rootin' Tootin', Tempest, and Tron) and sometimes the tabletop game Dungeons & Dragons.30 Within Applications Software, Baum linked play to profit, articulating a production culture of creative ­labor. Play and work, ­labor and leisure, blurred. It was what Urbaniec called "the magic of being in a place where in the cubicle next door, you would be hearing some game piece over and over. Gabriel Baum never walked by a group of people sitting ­there playing and said, ‘What's ­going on?' He'd go, ‘Keep up the good work.'" When an executive asserted that programmers were wasting time, Urbaniec recalled Baum responding, "I despise your Puritan work ethics. What we need is creativity. We need this flow of quality ideas that will make us products." This creativity fed a ludic culture of pranks. Programmer Dave Warhol recalled how at one point, several programmers found an enormous stuffed snake, which they proceeded to wind through the r­ afters of the offices, moving it e­very day a ­little closer to Baum's office. On another occasion Ettinger and Keith Robinson discovered a box of yellow plastic ducks at Mattel's com­pany store, bought them all, and proceeded to play golf with them among the cubicles. "We had a lot of playtime back then," Haft recalled. When asked about the relationship between work and play at Mattel, Haft replied, "It was all the same ­thing," adding that it was "looking into the ­future in terms of some of the ­things the tech companies would do."31 Robinson wrote that when working at Mattel Electronics, "Life seemed to be half stress, half party."32 Demanding work was coupled with child-­ like freedom—­what Hoshizaki termed a "happy and relaxed" atmosphere. Indeed, the most consistent reflection from our interviews was that designing Intellivision videogames was fun—­often described as the most fun someone enjoyed in their entire c­areer. Warhol emphasized that it was "easily the dream job of a lifetime. I moved to be five minutes away. I c­ ouldn't wait to get in. I was in by 8:00 in the morning. I'd start to see people head out and I would be ­there ­until 7:00 p.m." This represents an early example of the "aspirational ­labor" that has become a hallmark of digital work, with its conflation of doing what one "loves" with creatively producing for an employer.33 Ultimately, Baum had to balance programmers' freedom to play with meeting deadlines. Robinson noted that "the pressure of deadlines was constant," and office days tended to be long.34 Dombrower recalled that "it ­wouldn't be unusual for us to go to dinner at 10 ­o'clock at night at The Kettle in Manhattan Beach, and some of us would go back to work." Eric Del Sesto's September 1983 timesheets reveal only one eight-­hour workday that month: twelve-­or even sixteen-­hour workdays ­were typical, and one week Del Sesto clocked seventy-­six hours.35 The programmer Karen Tanouye (later McConathy) added that "it was such an in­ter­est­ing place to work. You didn't just go ­there 9 to 5; you lived ­there." This was also a hallmark of Preliminary Design, where Derek Gable recalled that toy designers "loved that com­pany so much, they would stay to midnight to get their jobs done if they had to." Baum even sought permission to provide programmers with Intellivisions to take home as "an incentive for the programmer to invest time and effort over and above that normally expected from them."36 Play and leisure outside the office was ­imagined as the continuation of ­labor. In an era where working from home in the technology industry was nearly unheard of, Intellivision's status as a domestic product could allow programmers to expand their workday into the home. Documentation played an impor­tant role in t­ hese emergent practices of learning, informal mentorship, and playful ­labor. Originally documentation was solely official, like General Instrument's CP1600/1610 manual and STIC [176] specifications, and above all "Your Friend, The EXEC," which outlined Intellivsion's operating system. When working on NBA Basketball, APh programmer Ken Smith recalled, "The first ­thing was they handed me ‘Your Friend, The EXEC,' which set the tone of a playful place to work" given its informal, joking style. When programmers encountered novel issues, the open-­ office environment at Mattel Electronics itself was a resource. David Stifel recalled how "very often, Sneakernet was the way you got information. ‘Hey, go talk to Bill Goodrich, go talk to William Fisher, go talk to so-­and-so.'" Over time, such informal "Sneakernet" tips ­were codified into notes that ­were saved and circulated. As Stifel emphasized, "We ­were making this up as we went along and getting information shared was very impor­tant. So a lot of ­those memos ­were just our way of communicating, ‘Here's a l­ittle quirk that ­hasn't been documented.'"37 Programmers thus began producing documents themselves. By March 1982, the Applications Software documentation library included internally produced materials like "How to Keep Moving Objects from Crashing Into Background ­Unless You Want Them To" and an explanation of the hand controller ports (figure 6.1).38 While informal mentorship remained the norm, beginning in January 1982 programmers also began o­rganizing seminars. Urbaniec emphasized that t­ hese provided an opportunity "to actually stand up and make ­presentations—­get ­people to get up out of their cubicle and learn something new." Not counting general orientation or question-­and-­answer sessions, at least 117 distinct seminars were held, falling roughly into three categories. First, t­here were occasional topical seminars (for instance, "CES: the Untold Story," where programmers who attended the Consumer Electronics Show (CES) described their experiences).39 Second were "game evaluation seminars" for work in pro­gress (for instance, "World Series Baseball Software Support Group" and "Pinball Software Support Group").40 Third and most frequently, ­there ­were "programming seminars" providing technical knowledge. Some ­were "training seminars" or "orientation seminars" for new hires, including one coordinated curriculum of seventeen seminars. Many seminars had humorous titles reflecting the playful environment of Applications Software, including "Assembly Instructions Made Human," "Fancy-­ Smancy Title Screens," and "Return to the Planet of the Moving Objects."41 Inside the office, the ludic atmosphere was coupled with more formal management systems—­reflecting how videogame software encodes fun but requires structure. For example, Group Leaders provided regular updates regarding videogames in their groups.42 One "proj­ect pro­gress" memorandum from Haft to Baum detailed work on nine videogames, listing the programmer, size, "confidence level" on the game's pro­gress, and deadline for expected completion.43 More formally, Daglow recalled that Baum's 6 



Hand-­drawn explanation of the hand controller inputs and ports (by Bob Newstadt). 

"management style was to have regular meetings. We had the Goldenrod meeting, which was a standard." ­These weekly meetings ­were ­organized around yellow charts called "Goldenrods" that showed a videogame's start date and size and tracked three impor­tant timelines: "Applications Software" (the programming schedule), "Graphics" (referring to the packaging and instruction manual), and "Manufacturing." This reveals a tripartite framework for Intellivision game production. At Goldenrod meetings, some videogames would be presented as "story­boards," and games further in development would be demoed. Through this oversight, man­ag­ers ensured videogames were completed on schedule, and sometimes de­cided if work on a game should cease. 
   

While managerial structure was needed, Baum emphasized that "I was ­there to be protection for the teams under­neath me, so they were not besieged by management." This protection allowed playful creativity to thrive. However, Baum also had to arrange more meddlesome product review meetings.44 ­These meetings involved the President and the ­Senior Vice-­Presidents of Production and Marketing moving from cubicle to cubicle, with programmers allotted ten minutes to pre­sent their videogames.45 Such meetings were usually viewed as intrusive, as captured by Creative Media staff Brad Gea­gley's satirical "Dante's Infernal Guide to Applications Software," consisting of short statements accompanied by images from Dante's Inferno (figure 6.2). As executives examined games in pro­gress, Daglow recalled that "some were pleasant and made reasonable comments. Others ­were petty and dictatorial (‘That background ­ought to be blue. Change it.'). I spent a ­great deal of time talking my best designers out of quitting a­ fter being the targets of such visits."46 Brainstorming Videogames 

Brainstorming was another playful aspect of working at both Mattel Electronics and APh. Mattel was no stranger to creative brainstorming. Its success as a toy com­pany was predicated on intensive innovation, with entire departments like Preliminary Design dedicated to generating new ideas. Mattel Electronics' brainstorming sessions took place in offices, parks, and often a private room at Barnabey's Hotel and Restaurant in Manhattan Beach. Urbaniec recalled "eating and having a beer or a glass of wine or ­whatever. We'd ­really start to loosen up and new ideas would come out." In an announcement for a "Blue Sky Session" in July 1981, Chang set forth several ground rules, including, "Do not be too quick to reject a half-­ baked idea," "Break from the past . . . ​sort out the technical prob­lems ­later," and "If you already have a good idea t­ oday, do have it properly documented now."47 While APh did not have an institutional history with Mattel, creativity was part of the job. Initially, APh was asked to create sports titles or traditional games like blackjack, poker, and horse racing—­videogame ideas where the source material was ready for simulation. Other early titles, like Space ­Battle, evolved from design documents originating at Mattel. However, APh did originate ideas, like Star Strike (which drew from the Death Star attack scene in Star Wars). Loughry recalled early visits to Mattel Electronics with High­tower, where Chang "was the person in charge, and we'd pre­sent the dif­fer­ent games. And then they would tell 


6.2 Image from the humorous memorandum "Dante's Infernal Guide to Applications 

Software" (Brad Gea­gley, Blue Sky Rangers archive). Gea­gley chose Gustave Doré's 1857 illustration of Caiaphas from Dante's Inferno to accompany the text. 

us which ones they wanted us to actually do." David Rolfe recalled Loughry maintaining a notebook with videogame ideas. Loughry also created one-­ paragraph descriptions of cartridges in development. One such summary from June 1981 listed eleven videogames, including Sub Hunt, Star Strike, and Frog Bog (figure 6.3).48 Once Mattel Electronics hired its own team of Intellivision programmers, they generated ideas individually, through brainstorming sessions, and via competitions. Some sessions ­were general while ­others ­were specialized, 

[180] 

  

   

6.3 Videogame summaries (Tom Loughry, "Preliminary Cartridge Descriptions," June 30, 1981, Tom Loughry archive). 


   

including one for educational cartridges (with ideas like "how weather works" and "building vari­ous inventions" like a clock, plane, and nuclear power plant).49 Sometimes programmers conducted on-­the-­spot ranking of brainstormed ideas. For one such session, the top-­ranked proposal was Connie Goldman's "Heaven Sent." This game was to revolve around Gabrielle, "the littlest angel," whose "mission is to climb to the gates of heaven" while avoiding ­devils throwing pitchforks and brimstone.50 More formal competitions ­were held as well. In addition to a competition for Kool-­Aid Man (chapter 7), a competition was held for a magic-­themed title. Ideas also emerged through a slower, iterative ­process. For example, Warhol wrote a memo proposing what became Mind Strike: "I would like to start spending my time developing a strategy game. It's based on an idea I had many years ago and can be best described as a cross between checkers, chess, and Stratego."51 On occasion even executives would propose videogames, for example, about the Rocky Horror Picture Show or the 1985 return of Halley's Comet.52 Such suggestions ­were not pursued, but marketing staff played a more significant role in brainstorming new games. Frank O'Connell, ­Senior Vice President for Marketing, referred to this as a "front-­end strategy" that would begin by "defining how many games we want in what category." This is another way in which videogame development inherited frameworks from Preliminary Design, where for toys, "marketing represent[ed] both the beginning and the end of the new product cycle."53 The marketer Mike Doepke concurred, "the research was being done as to what was selling in the marketplace. We would feed that back to the programming group." Indeed, marketers sometimes participated in brainstorming sessions with programmers. When announcing one such session, Chang's first ground rule was, "Every­one speaks as equals, including marketing people."54 Collaboration in Videogame Design 

Many early Intellivision games had one primary author who, as Loughry put it, "wrote the manual for goodness sake, as well as ­doing all the graphics and the sounds and the coding and the design." Hoshizaki emphasized how as a result, "it wasn't just putting a tiny bit of something into the proj­ect. It was our proj­ect and we ­were responsible." Stifel noted a mentality of "one programmer, one game. One programmer did it all. Graphics, sound, design, programming, every­thing. ­You're ­going to program the game, ­you're ­going to design the game, y­ ou're ­going to invent this game." While sole authorship was an impor­tant ele­ment of early game design, Intellivision game design frequently involved collaboration, complicating narratives of the "one-­person-­one-­game system" commonly described in [182] 

  

   

videogames history.55 Even early Intellivision titles involved more collaboration than is often recognized: Dave James's "Battlestar Galactica" concept, worked on by Mattel employee Bob Del Principe, was developed into Space ­Battle by APh programmer Hal Finney with assistance from Glenn High­ tower (chapter 2). Working with someone e­ lse's concept or graphics was common. For instance, when Smith was programming NBA Basketball at APh, James provided the screen layout and basketball court art. Smith was surprised that the scoring area occupied the upper half of the screen: "Only like 30% of the screen was being used for the gameplay. It was the vision they had of what Intellivision was ­going to look like." APh programmers could also request graphics. Loughry recalled looking through the graphics needed for a videogame, and for "the t­hings that I could not do, I had somebody that I could contact at Mattel." In Applications Software, a game might be passed to other programmers because the original programmer was experiencing difficulties, was needed for another title, or had left the com­pany. B-17 Bomber, mentioned in the Introduction, was first developed by Sohl, with input from Del Principe. Kai Tran and Peggi Decarli helped Sohl develop the graphics. Due to difficulties completing the videogame, Fisher and Roney ­were brought on to help.56 Sohl worked in the ­evenings, with Fisher and Roney working on dif­fer­ent parts of the code during the day. Even in this early period, it was becoming apparent that such collaboration unlocked creative possibilities. In one brainstorming session ­after B-17 Bomber's completion, concern was expressed that "colossal games" with multiple authors were difficult to manage, but Fisher responded that such videogames could "take advantage of team, versus singular, programming."57 Over time, specialization emerged in Applications Software through everyday collaboration, aided by the emergence of focused design tools (chapter 5). Specialists usually continued to program their own titles while contributing to other videogames, but by 1982, the programmer Eric Wels noted that about four additional staff members had been hired solely as graphic designers. In December 1982 Andy Sells accepted "the responsibility of coordinating music/sound effects" and "lining up a sound specialist for your game as well as supervising and troubleshooting the aural effort."58 By May 1983, a draft list of cartridges assigned to "graphic designers, sound specialists and educational product specialists" showed, for instance, that among other assignments Joe Ferreira (later King) was given graphics duties for Motocross and Mind Strike, Goldman was assigned Vectron and Scooby Doo, and Sells was tasked with sound design for Mission X and Maze-­A-­Tron.59 Haft, the designer for Maze-­A-­Tron, recalled how in addition to Sells's sound work, Wels did the graphics: "We had fun. It was an opportunity to work 6 



   

with other ­people ­because it can get ­really boring just being in a cubicle by yourself." The emergence of collaboration fostered ­labor specialization but also extended playful modes of interaction and game design. Less explicit forms of collaboration occurred as well, like when marketers helped brainstorm videogame ideas. During the development process, marketers could be hands-­off. Doepke emphasized that "game development was pretty much determined within the programming group and marketing just signed off on it." However, other marketers suggested that marketing could play a more collaborative role. Lockton emphasized that marketers ­were familiar with videogames and frequently contributed ideas to product design and development. Likewise, Gillis emphasized the "very impor­tant interaction between marketing and the a­ ctual software developers in terms of how the games themselves were executed," recalling that when Mattel Electronics needed more space videogames, marketing worked closely "with the programmers to even determine how explosions should be on screen." We call ­these suggestions "marketing mechanics." Game mechanics are broadly understood as the rules, procedures, and actions that define play in a par­tic­u­lar game. We speak of marketing mechanics when gameplay dynamics are ­shaped by marketer requests or, even more fundamentally, market share. For example, in chapter 9 we discuss how marketing mandated that titles for the Intellivoice extension be unplayable without its voice feature, leading to instances where players had to remember spoken numbers, then enter them to pro­gress. For videogames designed for the Entertainment Computer System, discussed in chapter 10, programmers had to add gameplay features that would allow players to change or manipulate a game—­like designing their own levels. Programmers sometimes found suggestions and contributions from marketing staff intrusive, recalling e­ arlier efforts by Preliminary Design toy designers to keep marketers at arm's length. Gea­gley's satirical memo, mentioned previously, included the quip "You will learn that Marketing has the final say on what products you will assist in developing."60 Creating manuals also involved collaboration. Scholarship on Atari games has concluded that they "­were truly a ‘pick up and play' experience. ­Little documentation was required for a user to understand and enjoy simplistic games such as Asteroids or Space Invaders."61 In contrast, many Intellivision titles ­were nearly unplayable without their manuals, reflecting the influence of board games and the aesthetics of intelligence and sophistication distinguishing Intellivision from its rival. This included not just the crucial early sports titles like Major League Baseball and NFL Football but strategy games like Utopia. Utopia's overlays show icons for the keypad numbers 1 through 0, including a red cross for 5 and a black rook for 1 (see figure 2.9). Only from the manual could one learn that the red cross is a hospital costing [184] 

  

   

75 points, which "greatly increases factory production," and the black rook is a fort costing 50 points, which "guards surrounding land area against rebels." Programmers usually drafted their own manuals, sharing them among colleagues before sending them for editing and production to Visual Design, a longstanding Mattel division. For instance, Tom Loughry circulated an eight-­page handwritten "gameplay and instructions" for Boxing, followed three months ­later by a memo detailing changes to the game.62 William Fisher noted in the Space Hawk code, "10/12/81 . . . ​Prepared an early version of the instructions for the writers to see."63 Final versions of a manual sometimes retained programmers' original prose: the phrase "Strategy should be a mixture of offense and defense and common sense" in Loughry's draft for Boxing appears in the final manual as "Strategy should be a mix of offense and defense, plus common sense." ­These emergent forms of collaboration s­haped ­labor practices and videogame design, and increasingly became an explicit topic of discussion at Applications Software. Asserting that "our current design methods are haphazard, unprofessional, and unreliable," in July 1983 four programmers circulated a memo proposing "a new game design ­process."64 Bob Newstadt, one of the authors, recalled "being r­ eally surprised when I went to Mattel. I was very excited to learn how to do videogames. So I went around, ‘Oh, how do you make it?' And every­one had a dif­fer­ent story." Newstadt added that by mid-1983 the programmers had "talked about the commonalities of it. Our main goal is we thought the games would be better if it wasn't just the heroic work of one person." Examining Mattel Electronics' production culture involves "turning the attention to game industry professionals' everyday practices," as Olli Sotamaa has emphasized.65 Sotamaa located this approach within "studio studies, a subset of game production studies that puts focus on individual studios," phrasing that reflects cinematic metaphors structuring con­temporary vid66 eogame design. Intellivision, however, illuminates a prehistory of studio studies. The language of game "studios" was unknown during the second generation. The idea of publicly crediting individuals was contested, and design was analogized to computer programming, not moviemaking. The practices through which staff at APh and Applications Software staff made Intellivision games provide an archaeology of studio studies, including APh's status as a proto-­independent studio and Applications Software's status as a proto-­studio within a larger com­pany. Through collaborative practices, programmers, man­ag­ers, and marketers developed videogame code and tools, texts and reports, office layouts and physical devices, and social norms and conventions. ­These constituted Intellivision's production culture, the legacies of which inform the world of videogame production t­ oday. 6 


 This is a section of doi:10.7551/mitpress/14266.001.0001 

Intellivision How a Videogame System Battled Atari and Almost Bankrupted Barbie ® By: Tom Boellstorff, Braxton Soderman Citation: Intellivision: How a Videogame System Battled Atari and Almost Bankrupted Barbie ® By: Tom Boellstorff, Braxton Soderman DOI: 10.7551/mitpress/14266.001.0001 ISBN (electronic): 9780262380553 Publisher: The MIT Press Published: 2024 

The open access edition of this book was made possible by generous funding and support from MIT Press Direct to Open