Soccer enthusiasts nationwide are eagerly anticipating the upcoming college bowl season, where they will not only witness on-field excitement but also be dazzled by elaborate “card stunts” performed by spectators in the stadium. These meticulously coordinated group efforts can craft intricate images akin to the distorted visuals on computer screens.
A parallel example illustrating how our surrounding systems operate and why we need not fear a future dominated by machines can be found in Michael Littman’s latest book, Code to Joy: Why People Truly Understand a Bit of Programming. Geared towards novice programmers of all ages, Code to Joy serves as a user-friendly gateway to the basics of coding, covering everything from sequencing commands to managing variables.
GIMME A BLUE, please!
Card stunts resemble flash mobs in that participants require no specialized skills and can even rehearse beforehand. By holding up colored signs, they create a massive, transient billboard in front of the stadium crowd. Participants simply need to follow brief control sequences provided to them. These instructions direct the stadium audience to raise the appropriate poster-sized colored cards when prompted by the stunt leader. A standard set of guidelines for card stunts typically includes:
- Diligently follow the instructions.
- Maintain the card at eye level (but not above the head).
- Orient the desired color towards the designated area (away from oneself).
- After the stunt, pass the cards along the aisle (do not discard them).
While these guidelines may appear straightforward, deviating from them can lead to chaotic outcomes. Yet, there is always a curious individual, like Felix, who might later inquire, “Pardon me, what was that first instruction again?” Undoubtedly, that would be my role.
The climax unfolds when a specific command sequence is issued, which for one individual in the group might entail:
- Blue
- Blue
- Blue
But hold on, take a moment to contemplate the bigger picture. The essence of card stunts lies in the arrangement of spectators in stadium seating, transforming them into a colossal computer display screen by hoisting rectangular colored signboards. Each participant effectively acts as a pixel, contributing to the overall image—creating a visual spectacle larger than life.
The genesis of card stunts can be traced back to the 1920s when audience engagement in school sports spurred their development. While initially, individuality was encouraged, it became apparent that collective efforts enhanced their appeal. By the 1950s, there was a growing aspiration to craft more elaborate displays. Cheer squads painstakingly designed and scripted instructions for each seat, investing considerable effort to showcase their team spirit. Subsequently, in the 1960s, several schools pioneered programs to convert hand-drawn images into personalized instructions for each participant, leveraging computer technology to streamline the process. This instance underscores the shift towards computers generating command sequences for people to execute, contrasting with the earlier scenario where people dictated commands to machines. Automation facilitated by computers enabled the creation of more intricate displays, exemplified by the detailed command series:
- Light up 001, away
- Blue 003
- Illuminate 005
- Red 006
- White 008
- Orange 013
- Gleam 015
- Conclude 021
- Brighten 022, away
- Descend 035
- Cease White 036
- Blue 043
- Conclude 044
- Halt White 045
- Dark Silver 057
- Culminate 070
Reading the instructions proves more engaging than witnessing the final product; in this case, a segment of an animated Stanford “S.” An announcer in the stadium calls out the step number (“Forty-one”), synchronizing these commands. Each participant then follows their designated actions, such as “I’m still holding the white card I raised on the 36th step, but by step 43, I’ll be preparing to swap it for a blue card.”
Participating in a card stunt is relatively straightforward, as mentioned earlier, yet it exemplifies how command sequences can be formulated and executed with computers guiding us on what to do, reversing the conventional flow of instructions. However, even in simplicity, mishaps can occur. During the 2016 Democratic National Convention, supporters of Hillary Clinton organized a stadium-wide card stunt intended to symbolize national unity. Despite its patriotic intent, some attendees hesitated to participate, resulting in a bewildering display that awkwardly spelled out “Stronger Along.”
Today, computers streamline the translation of images into instructions for displaying colors, essentially converting digitized images into directives for showcasing specific color combinations at each location. Given that most images comprise millions of colored pixels (megapixels), while a stadium’s card stunt section may consist of 1,000 seats, translating an image into card instructions poses a unique challenge. Rather than tasking each individual with holding up numerous cards, it is more practical to determine the average color in a given image segment and select the closest matching color from a standard palette (e.g., sixty-four Crayola options).
The process of computing color averages may initially seem perplexing. How does one instruct a machine to blend green and yellow to achieve a spring-like hue? By analyzing the composition of red, green, and blue in each color dot of an image, computers can effectively determine the average color. This method resonates with how our eyes perceive colors—by interpreting various wavelengths of light as distinct combinations of red, green, and blue. By averaging the proportions of these primary colors across a group of pixels, computers can accurately represent the average color of an image segment. While subsequent mathematical manipulations further refine this process, the fundamental principle remains consistent: aggregating color data to generate a representative shade.
After computing the average color, the system must identify the closest available color match from the designated palette. Employing the Euclidean distance formula—a standard method for comparing color similarities based on their red, green, and blue values—the system determines the most akin color option to the calculated average. This distance calculation process involves:
- Calculating the square root of the difference in brightness between two colors.
- Squaring the disparity in blue content between the colors.
- Combining these squared values to ascertain the divergence in green content.
- Taking the square root of the sum.
Thus, the system selects the color from the available options that best approximates the average color of a given image segment, assigning corresponding colored cards to individuals seated in that section.
The resemblance between color averaging and distance calculation is intriguing, albeit coincidental. Sometimes, a simple core concept remains just that—fundamental and uncomplicated.
By leveraging these operations—color averaging and color matching—the computer aids in formulating the card stunt command sequence. Inputting the specific image, seating layout, and available color cards, the computer generates a chart detailing which colored cards should be held up in each seat to replicate the image accurately. In this scenario, the computer primarily handles logistical tasks, such as selecting the nearest color match, while leaving the creative decisions to the users. Consequently, the process of crafting command sequences becomes more efficient and less labor-intensive, with the computer assuming a supportive role in executing user-generated images.
This paradigm shift empowers the system to play a more proactive role in generating command sequences, transitioning from explicit instructions to implicit guidance. By utilizing a 2×2 grid from chapter 1 as an analogy, this shift enables users to transition from dictating commands to machines (explicit instructions) to providing incentives for machines to interpret and execute tasks (implicit instructions). This iterative process enhances user-machine interaction and streamlines the creation of complex visual displays, akin to the challenge posed by sudoku puzzles rendered from facial images. In this context, the computer generates command sequences to render images, incorporating user-defined parameters to achieve the desired visual outcome, akin to the dance-step analogy.
