Collaboration
Almost every program you use was built by a team.
The app that tells you when the next bus arrives, the game on your phone and the system your school uses for grades were each designed, written, tested and fixed by groups of people who had to share ideas, split up the work and settle disagreements.
This section is about why that teamwork matters: how working with others, and especially with people who think differently from you, makes a better product, and which tools and habits make teamwork run smoothly.
On the exam, collaboration is tested in the multiple-choice section, as part of Big Idea 1 (Creative Development, 10-13% of the multiple-choice questions), and it can appear in the five-question reading passage about a computing innovation.
No written-response prompt asks about it directly, but it shapes your Create performance task: you may write the program with a partner, while your video and your Personalized Project Reference must be your own work.
- Each multiple-choice question is worth 1 point. A "Select two answers." question earns its point only if you choose both correct options.
- The written part of the exam has four prompts about your own Create task code (Written Response 1, 2(a), 2(b) and 2(c)), each worth 1 point. Your video and your program meeting the requirements earn 1 point each, so Section II is worth 6 points.
- A prompt's point needs everything the prompt asks for. If it asks you to identify something, describe it and explain why, a perfect explanation with no identification earns nothing, so answer every part.
- Answers must match your code exactly: the first statement the prompt names, specific values where values are asked for, the procedure or list you name.
- Anything inaccurate or inconsistent with your program costs the point, even if the rest is right. Several answers to one prompt must all be right.
- When a prompt says "if it is not possible, explain why", that route earns the point only with a reason taken from your code.
- Nothing is subtracted for a wrong answer, so never leave a multiple-choice question blank.
What counts as a computing innovation
A computing innovation is any new product, tool or idea that depends on a program to do what it does.
You need to know that "A computing innovation includes a program as an integral part of its function."
The word integral matters: take the program away and the innovation no longer works.
A self-checkout kiosk without its software is a box with a screen; a mechanical pencil has no program at all, so it is not a computing innovation, however clever it is.
Computing innovations come in three kinds.
You need to know that "A computing innovation can be physical (e.g., self-driving car), nonphysical computing software (e.g., picture editing software), or a nonphysical computing concept (e.g., e-commerce)."
A physical innovation is a device you can touch whose behavior is run by a program, such as a fitness band that counts steps.
Nonphysical software is a program itself, such as a spreadsheet app.
A nonphysical concept is a way of doing things that only exists because programs make it possible, such as online ride-hailing or video streaming: no single device or app is the concept, but every version of it runs on programs.
Why sort them?
Because the exam's reading passages describe innovations of every kind, and the questions ask about the data each one uses and the effects it has.
Seeing that a ride-hailing service is a concept carried out by many programs (a rider app, a driver app, a matching server) helps you reason about where its data come from.
Why diverse teams build better innovations
A team is more than extra hands.
People bring different talents (one person is good at layout, another at debugging, another at explaining things to users) and different perspectives (different ages, languages, abilities, neighborhoods and experiences of technology).
You need to know that "Effective collaboration produces a computing innovation that reflects the diversity of talents and perspectives of those who designed it."
Perspectives matter most because of bias.
Every designer assumes things about users without noticing.
A team of students who all ride the bus to school might design a school-transport app that never shows car-pool or walking routes.
A team whose members all read English easily might never think to offer Spanish.
A team where no one uses a screen reader might put important information only in pictures.
None of these are deliberate; they come from who was in the room.
You need to know that "Collaboration that includes diverse perspectives helps avoid bias in the development of computing innovations."
A teammate or tester with a different experience spots the assumption early, when it is cheap to fix.
Bias in computing comes back in Big Idea 5, where you study how biased data and design can harm groups of users.
Talking to users
The people who will use a program are part of the team too, even if they never write a line of code.
You need to know that "Consultation and communication with users are important aspects of the development of computing innovations."
Developers consult users through surveys, interviews, watching people use an early version, and feedback forms in an app.
Users tell you what the program is actually for from their point of view, which is often not what the developer guessed.
You need to know that "Information gathered from potential users can be used to understand the purpose of a program from diverse perspectives and to develop a program that fully incorporates these perspectives."
For a school bus-arrival app, students might say they want a push alert five minutes before the bus comes; parents might want to know when the bus has left; a student who uses a wheelchair might need to know whether the arriving bus has a lift.
Each conversation adds a purpose the team had not seen, and the final program can serve all of them.
Tools and models for working together
You need to know that "Online tools support collaboration by allowing programmers to share and provide feedback on ideas and documents."
Examples you will meet:
- Shared documents and slides that several people edit at once, with comment threads for questions and suggestions.
- Shared code repositories (version control), where each change is saved with who made it and why, so the team can review a change before it is accepted and can go back to an earlier version if something breaks.
- Code review, where a teammate reads your code and leaves comments before it joins the main program.
- Issue trackers and task boards, which list bugs and features and who is working on each.
- Messaging and video calls for quick questions and meetings when people are not in the same room.
Teams also follow agreed ways of working.
You need to know that "Common models such as pair programming exist to facilitate collaboration."
In pair programming, two programmers share one computer.
The driver types the code.
The navigator reads every line as it is written, thinks about the overall plan, and points out errors and better ideas.
They swap roles often.
It feels slower at first, but the navigator catches mistakes as they happen, both people understand all of the code, and ideas from two heads are combined.
Interpersonal skills
Tools do not make a team work; people do.
You need to know that "Effective collaborative teams practice interpersonal skills, including but not limited to: communication; consensus building; conflict resolution; negotiation"
Skill
What it looks like in a programming team
Communication
saying clearly what you are working on, asking when you are stuck, listening before replying
Consensus building
collecting everyone's ideas and finding a choice the whole team can support, such as voting on three possible app names after each person explains their favorite
Conflict resolution
settling a disagreement by focusing on the problem, not the person: testing both versions of a feature with users instead of arguing about whose is better
Negotiation
trading off what each person wants: one teammate gets the color scheme they prefer, the other gets the extra settings page they asked for
These are the skills of Practice 6 (Responsible Computing), which the Create task asks you to use when you work with a partner.
They are not assessed in the multiple-choice section as a practice, but questions on collaboration test whether you can recognize them in a scenario.
Collaboration in your Create task
You may develop your Create task program with one or more partners, and your Personalized Project Reference may show code you wrote together.
Your video and your Personalized Project Reference must be made by you alone, and the written responses on exam day are yours alone.
Any code you did not write yourself (a partner's code from another program, starter code from your teacher, code from a library or from a generative AI tool) must be acknowledged in your program code.
How to do that is in the section on documentation; the legal and ethical side of using other people's work is in Big Idea 5.
- Computing innovation: A computing innovation includes a program as an integral part of its function.
- Kinds of computing innovation: A computing innovation can be physical (e.g., self-driving car), nonphysical computing software (e.g., picture editing software), or a nonphysical computing concept (e.g., e-commerce).
- Pair programming: a common collaboration model in which two programmers share one computer, one (the driver) writing the code while the other (the navigator) reviews each line and plans ahead, swapping roles often.
- Interpersonal skills: Effective collaborative teams practice interpersonal skills, including but not limited to: communication; consensus building; conflict resolution; negotiation
- Bias: an unintended unfairness in a program that favors some users or leaves others out, often caused by the assumptions of the people who designed it.
For each item, decide whether it is a computing innovation and, if it is, which kind.
Give the reason.
Item
Computing innovation?
Kind
Reason
A robotic lawn mower that maps a yard and avoids flower beds
Yes
Physical
A device you can touch; its program decides the route and when to turn.
A music-notation app that plays back what you write
Yes
Nonphysical software
The app is the program.
Online ride-hailing
Yes
Nonphysical concept
A way of getting a ride that exists only because rider apps, driver apps and matching servers run programs.
A ten-speed bicycle with mechanical gears
No
(none)
No program is part of how it works.
A digital thermostat that learns when a family is home
Yes
Physical
A device whose heating schedule is set by its program from the data it collects.
Answer: four are computing innovations: the mower and thermostat are physical, the notation app is nonphysical software, and ride-hailing is a nonphysical concept.
The bicycle is not, because no program is integral to its function.
The test each time is the same: would it still do its job with the program removed?
A team of four juniors is building an app that tells students at their high school when their bus will arrive.
All four ride the same bus route.
Explain how adding people with different perspectives to the design process could improve the app.
Step 1: spot the shared assumption. All four ride one route, so they will design for a student who rides a bus, reads English and can see the screen.
Step 2: name who is missing and what they would add. A student on a rural route with weak cell signal would ask for the arrival time to be sent as a text message.
A parent who speaks Spanish at home would ask for a Spanish version.
A student who uses a screen reader would ask that the arrival time be written as text, not only shown on a map.
Step 3: link it to bias. Without those people, the app works well only for students like the designers.
Consulting them removes that bias before the app is released.
Answer: The team all ride one route, so the app would reflect only their experience.
Including students from rural routes, families who speak other languages and students who use assistive technology adds needs the team would not have noticed (text alerts, a Spanish option, screen-reader text).
Collaboration that includes diverse perspectives helps avoid this bias, so the app serves every student who rides a bus.
Ana and Marcus are pair programming a quiz game.
Ana is the driver.
Marcus keeps telling her to rewrite the scoring code his way, and Ana has started ignoring him.
Describe how each of the four interpersonal skills could get them back on track.
- Communication: Marcus explains why he prefers his version (it handles a tie), and Ana says what she was trying to do, so both understand the two ideas.
- Conflict resolution: they stop arguing about whose code is better and test both versions on the same three cases, including a tie, and keep the one that works.
- Consensus building: they agree on a rule for next time: the navigator raises an idea once, and if the driver disagrees they test it together.
- Negotiation: Ana keeps her layout for the question screen; Marcus's tie-handling goes into the scoring code. They swap roles every 20 minutes so both drive.
Answer: good collaboration here means listening to reasons, settling the disagreement by testing against the problem and agreeing on a way of working, with each partner getting part of what they want.
The game ends up with both partners' best ideas, which is the point of pair programming.
A team is developing a voice-controlled app for checking out library books.
All of the testing so far has been done by the team's five members.
Which of the following actions is most likely to reduce bias in the app?
(A) Testing the app more times with the same five team members (B) Asking users of different ages and with different accents and speech patterns to test the voice commands (C) Adding more books to the library catalog the app searches (D) Making the app's buttons larger
Reasoning. Bias here means the voice commands may work well only for voices like the team's.
(A) repeats the same narrow test.
(C) adds data, but not the missing perspectives.
(D) may help some users, but it does not address the voice commands, which is where the untested assumption sits.
Answer: (B).
Testers with different accents and speech patterns bring the perspectives the team lacks.
Three students are building a club website from their own homes.
Which of the following best explains how an online shared code repository supports their collaboration?
(A) It makes the website load faster for visitors.
(B) It lets each student share changes, see who made each change, and comment on changes before they are combined.
(C) It prevents the students from ever editing the same file.
(D) It removes the need for the students to test their code.
Answer: (B).
Online tools support collaboration by letting programmers share work and give feedback on it.
(A) is about the website's performance, not teamwork.
(C) is false: a repository lets people work on the same file and helps them combine their changes.
(D) is false: shared code still needs testing.
Collaboration questions often offer one option that sounds hard-working but adds no new perspective ("test it again with the same people", "have one expert do it all").
The answer that reduces bias or improves the innovation brings in people or feedback the team did not have before: users from different groups, a teammate with a different skill, a review by someone who did not write the code.
- Thinking any device is a computing innovation. A mechanical tool with no program is not one. It is tempting because "innovation" sounds like "anything new". Ask whether a program is integral to how it works.
- Thinking collaboration just means splitting the work. Splitting tasks saves time, but the gains you need to know are the diversity of talents and perspectives and the bias it helps avoid. An answer about speed alone misses the point.
- Confusing the pair-programming roles. The driver types; the navigator reviews and plans. It is tempting to think the navigator is idle, but the navigator's checking is what catches errors early.
- Assuming users are only testers at the end. Consultation with users belongs at every stage, starting before any code is written, so the team understands the purpose from the users' point of view.
- Treating a disagreement as one person winning. Conflict resolution and negotiation look for the best result for the program, often by testing both ideas, not by giving in.
Check your understanding
1. Which of the following is the best example of a computing innovation that is a nonphysical computing concept? (A) A smart watch that measures heart rate (B) A photo-filter app installed on a phone (C) Online grocery delivery, where orders placed in an app are sent to shoppers and drivers (D) A desk lamp with a manual on and off switch Total [1]
2. A company is designing a navigation app for city bus riders. Which of the following best explains why the company should gather information from a wide range of potential users before writing the program? (A) Users can write parts of the program code for the company. (B) Users with different needs reveal purposes and problems the developers did not consider, so the app can be designed to serve all of them. (C) Gathering information from users guarantees that the app will contain no errors. (D) The company can avoid testing the app after it is written. Total [1]
3. Which of the following best describes pair programming? (A) Two programmers each write a different program and then compare the results. (B) One programmer writes the code while a second programmer reviews each line and plans ahead, and the two switch roles regularly. (C) Two programmers split a program in half and never look at each other's code. (D) One programmer writes the program and a second programmer writes only the comments. Total [1]
4. A team building a school events app cannot agree on whether the home screen should show a calendar or a list. The team leader asks each member to explain their reasons, then the team picks the design that best meets the needs everyone raised. Which interpersonal skill is the team leader mainly using? (A) Consensus building (B) Negotiation, because one member gives up something in exchange for something else (C) Pair programming (D) Debugging Total [1]
5. Which of the following are ways that online tools support collaboration among programmers? Select two answers. (A) A shared document lets team members leave comments on a proposed design. (B) A shared code repository records each change and lets teammates review it before it is combined with the main program. (C) A faster processor lets a program run in less time. (D) A larger monitor lets one programmer see more lines of code at once. Total [1]
6. A team of software developers, all of whom are right-handed, designs a drawing app in which every important button is placed on the right edge of the screen. Users who are left-handed find the app awkward to use. Which of the following would have been most likely to prevent this problem? (A) Writing the program in a different programming language (B) Including left-handed users in the team or in early user testing (C) Adding more drawing tools to the app (D) Releasing the app sooner so that more people can download it Total [1]
7. Which of the following are examples of consultation and communication with users during the development of a computing innovation? Select two answers. (A) Interviewing music teachers about how they track student practice before designing a practice-log app (B) Choosing the color of the app's icon from the developers' favorite colors (C) Watching a group of students use an early version of a homework app and noting where they get stuck (D) Rewriting a procedure so that it runs faster Total [1]
8. A student and her partner are building a program for the Create performance task. Which of the following is LEAST likely to be a benefit of developing the program collaboratively? (A) The partners can catch each other's errors as the code is written. (B) The partners bring different ideas about what the program should do. (C) Each partner can submit the other's video as their own. (D) The program can reflect the different talents of both partners. Total [1]
Answers
1. (C). Online grocery delivery is a way of shopping made possible by programs (ordering app, shopper app, routing), not one device or one app. (A) is a physical innovation; (B) is nonphysical software; (D) has no program, so it is not a computing innovation at all.
2. (B). Information from potential users with diverse perspectives shows the purpose of the program from their points of view, so the program can incorporate those perspectives. (A) is not what consultation is for. (C) and (D) are false: no amount of consultation removes the need for testing or guarantees error-free code.
3. (B). This is the driver and navigator model. (A) and (C) are two people working separately, not together; (D) wastes the second programmer, who should be reviewing the code itself.
4. (A). Collecting everyone's reasons and choosing an option the whole team can support is consensus building. (B) would need a trade between members; nobody trades anything here. (C) is a collaboration model, not an interpersonal skill, and (D) is about finding errors.
5. (A) and (B). Both let programmers share work and give feedback on it. (C) and (D) are hardware that helps one person, not collaboration tools.
6. (B). A team that included left-handed people, or tested with them, would have spotted the assumption that every user is right-handed. This is collaboration with diverse perspectives avoiding bias. (A), (C) and (D) do nothing about the missing perspective.
7. (A) and (C). Interviews and direct observation are ways of consulting users. (B) uses only the developers' own preferences, and (D) is a programming change with no contact with users.
8. (C). The video must be made individually, so using a partner's video is not allowed and is not a benefit of collaboration. (A), (B) and (D) are all real benefits: errors caught early, more ideas and a mix of talents.