Practice · Software Process Models

Q1

Which statement about software processes is correct?

A Every team has a process, even if it is informal; choosing it well is an engineering decision. B Agile teams do not have a process. C Only large companies have a software process. D A process is the same as a programming language.

Q2

What did Winston Royce's 1970 paper — often cited as the origin of the waterfall — actually say about a single-pass sequential process?

A Testing should be skipped to save time. B It is the only correct way to build software. C It is risky and invites failure; feedback between phases and a pilot version are needed. D It should be replaced by Scrum.

Q3

In the V-model, which testing level is paired with the user requirements?

A Unit testing B Acceptance testing C System testing D Integration testing

Q4

For which project is a plan-driven (V-model) process the most appropriate?

A A startup testing whether students want a new social app B A weekend hackathon game C Firmware for a certified insulin pump with stable, regulated requirements D A research prototype whose goal is still unclear

Q5

What is the main weakness of a pure waterfall process for a new student app like StudyBuddy?

A It produces too little documentation. B It requires daily meetings. C Users see working software only at the end, so wrong requirements are discovered late, when they are expensive to fix. D It cannot be used with Java.

Q6

The central idea of Boehm's spiral model is to…

A avoid planning altogether B identify and resolve the biggest risks early in every loop, for example with prototypes C release to users every day D write all documentation first

Q7

The StudyBuddy team is unsure whether an AI model can answer course questions accurately enough using only the instructor's files. Following the spiral model, what should they do first?

A Write the user manual B Build a quick prototype of Buddy with real course files and measure answer quality C Design the login page and the database schema D Buy more servers

Q8

Which is one of the four values of the Agile Manifesto?

A Processes and tools over individuals and interactions B Comprehensive documentation over working software C Contract negotiation over customer collaboration D Responding to change over following a plan

Q9

A student says "Agile means we write no documentation and make no plans." What is the best correction?

A That is correct. B Agile only applies to documentation, not to code. C Agile forbids plans but allows documentation. D Agile values documentation and plans, but values working software and responding to change more; the items on the right still have value.

Q10

In Scrum (2020 Guide), who is accountable for ordering the Product Backlog to maximize the product's value?

A The most senior programmer B The Developers C The Scrum Master D The Product Owner

Q11

What is the purpose of the Sprint Retrospective?

A To plan the work of the next Sprint B To show the Increment to stakeholders C To assign blame for bugs D To inspect how the team worked and decide improvements to its process

Q12

Mei says a feature is "done" because it works on her phone. Daniel says it is not done because there are no tests. Which Scrum element prevents this argument?

A The Product Goal B The Daily Scrum C The Definition of Done D The burndown chart

Q13

A Kanban team finishes on average 3 cards per day and has on average 12 cards in progress. By Little's law, what is the average cycle time?

A 36 days B 4 days C 0.25 days D 15 days

Q14

Throughput stays at 2 cards per day. The team reduces average WIP from 10 cards to 4 cards. What happens to the average cycle time?

A It grows from 5 days to 8 days. B It falls from 5 days to 2 days. C It stays the same. D It becomes 20 days.

Q15

The "In review" column has a WIP limit of 2 and already holds 2 cards. Omar has just finished coding a new card. According to Kanban, what should he do?

A Start coding another new card. B Raise the WIP limit to 10. C Move his card into "In review" anyway. D Help review one of the two waiting cards so that the column has space.

Q16

What is the difference between continuous delivery and continuous deployment?

A In continuous delivery every passing change is ready to release at the push of a button; in continuous deployment it is released automatically. B Continuous deployment has no tests. C Continuous delivery is only for mobile apps. D There is no difference.

Q17

Which of these is not one of the four DORA key metrics?

A Time to restore service B Change failure rate C Number of lines of code written per day D Deployment frequency

Q18

Out of 40 deployments last month, 6 caused an incident that needed a fix or rollback. What is the change failure rate?

A 15% B 6% C 34% D 85%

Q19

DORA research found that high-performing teams deploy often and have low failure rates. What best explains this?

A Small, frequent changes are easier to review, test and roll back, so each deployment carries less risk. B It is a coincidence. C They have more expensive servers. D They skip testing to deploy faster.

Q20

Using Boehm and Turner's factors, which situation points most strongly toward a plan-driven process?

A 300 developers across 5 subcontractors, stable requirements, failures could cost lives B Three students building a prototype for a competition C A team that enjoys experimenting with new ideas every week D Small team, frequently changing requirements, experienced developers

Q21

Why does the StudyBuddy team keep its privacy and security requirements fixed early, even though it uses Scrum?

A Because the Product Owner does not like security. B Because these requirements come from law and university policy, carry high risk, and must not drift sprint by sprint — a small plan-driven part inside an agile process. C Because security cannot be tested. D Because Scrum forbids changing requirements.

Q22

An AI agent can now produce a 600-line pull request in two minutes. According to this chapter, what becomes the main limit on the team's safe speed?

A How fast humans can understand and verify changes (the review bottleneck) B The number of programming languages the team knows C Typing speed D The size of the office

Q23

Which item is the best addition to StudyBuddy's Definition of Done in the age of AI?

A "AI-generated code is read and understood by the committer, covered by tests a human wrote or checked, and no personal data was pasted into external tools." B "All code must be written by AI." C "The AI said the code is correct." D "No code may ever be written with AI."

Q24

The team wants to know whether their new AI coding assistant really makes them faster. What is the best approach?

A Count how many lines of code the AI writes. B Ask each developer whether they feel faster. C Trust the vendor's advertisement. D Compare measured cycle time, deployment frequency and change failure rate before and after adoption.

Q25

A team uses Scrum for new features and a separate Kanban board with a WIP limit for urgent production bugs. This is…

A continuous deployment B the waterfall model C a reasonable hybrid: Sprints for planned work, flow for unpredictable urgent work D forbidden by the Scrum Guide

Q26 Short answer

Explain Little's law in your own words and use it to argue why a StudyBuddy team of five should not have ten features "in progress" at the same time. Include a numerical example.

Q27 Short answer

Choose a process for each project and justify it with at least two of Boehm and Turner's factors: (a) software for a railway signalling system; (b) StudyBuddy; (c) a one-week hackathon prototype.

Q28 Short answer

Here is a deployment log for one week: 5 deployments; lead times from commit to production of 2, 4, 1, 3 and 5 hours; deployment #4 caused an incident that took 90 minutes to fix. Compute the four DORA metrics.

Q29 Short answer

Write a short "AI in our process" policy (5–7 bullet points) for the StudyBuddy team, covering at least: which activities may use AI, the size of AI-generated changes, review, tests, and data protection.

Q30 Short answer

Draw (in words or ASCII) the V-model and explain why it is described as "testing planned early, executed late".