Agile Modeling, Prototyping, and Scrum Study Guide
Concepts of Prototyping and the Four Main Types
Definition of Prototyping: Prototyping is an information-gathering technique useful in seeking several critical system components:
User reactions to the system.
Suggestions for improvement.
Innovations that may not have been previously considered.
Revision plans for the development process.
The Patched-Up Prototype:
Defined as a system that works but is "patched up" or "patched together."
It is a working model that contains all intended features but operates inefficiently.
Users are able to interact with the system fully.
Data storage and retrieval processes may be inefficient compared to the final product.
Nonoperational Scale Models:
A nonworking scale model set up to test specific aspects of the design.
This is typically produced when the coding required for a full application is too expensive to prototype.
A useful conceptualization of the system is gained by prototyping the input and output only, omitting the internal processing logic.
First-Of-A-Series Prototype:
Involves creating a "pilot" system.
The prototype is completely operational.
This approach is useful when many installations of the same information system are planned.
A full-scale prototype is installed in one or two locations first. If it succeeds based on customer usage patterns and other key factors, exact duplicates are installed at all remaining locations.
Selected Features Prototype:
Building an operational model that includes some, but not all, of the features of the final system.
It focuses on essential features and is built in modules.
These modules ultimately become part of the actual final system.
Role of Users in Prototyping:
Honest Involvement: Users must be genuinely engaged in the testing process.
Clear Communication: The purpose of the prototype must be communicated clearly to avoid false expectations.
Feedback loops: System analysts must actively seek suggestions for changing or expanding the prototype and fostering innovation.
Fundamentals of Agile Modeling
Definition: Agile modeling is a collection of innovative, user-centered approaches to systems development.
Core Values of Agile Modeling:
Communication: Ensuring clear exchange between developers and business users.
Simplicity: Focused on building the leanest possible solution to the problem.
Feedback: Continuous input from users and the system itself to guide development.
Courage: The willingness to discard failing code or change directions entirely if required.
The Principles of Agile Modeling:
Satisfy the customer through the delivery of working software.
Embrace change, even if introduced late in the development cycle.
Deliver functioning software incrementally and frequently.
Encourage customers and analysts to work together daily.
Trust motivated individuals to complete the job.
Promote face-to-face conversation as the most effective communication method.
Concentrate on getting software to work as the primary measure of progress.
Encourage continuous, regular, and sustainable development.
Adopt agility with systematic attention to mindful design.
Support self-organizing teams.
Provide rapid feedback loops.
Encourage high quality in all deliverables.
Review and adjust behavior occasionally to improve effectiveness.
Adopt simplicity as a core design philosophy.
Basic Activities and Resource Variables in Agile
Four Basic Activities:
Coding: The only activity that is impossible to skip. Code is used to learn and to communicate ideas that would otherwise remain fuzzy or unshaped.
Testing: Automated testing is critical. Developers write tests for coding, functionality, performance, and conformance. Large libraries of tests exist for most languages and are updated throughout the project. Short-term testing builds confidence; long-term testing keeps the system maintainable.
Listening: This is done "in the extreme." Developers use active listening with both programming partners and customers. Because formal written communication is minimized, listening becomes a paramount skill.
Designing: A way to create structure to organize logic. Design is evolutionary; systems are always being designed rather than just once at the beginning. Effective design locates logic near the data it operates on and remains simple and flexible.
Four Resource Control Variables:
Time
Cost
Quality
Scope
Core Agile Practices:
Short Releases: Frequent software iterations.
-Hour Work Week: Limiting hours to improve worker effectiveness and prevent burnout.
Onsite Customer: Having a business representative available to the team at all times.
Pair Programming: Two developers working together at one workstation to enhance quality.
The Agile Development Process and User Stories
Process Steps:
Listen for user stories.
Draw a logical workflow model.
Create new user stories based on the logical model.
Develop display prototypes.
Create a physical data model using feedback from prototypes and logical workflow diagrams.
User Stories:
Explanation of a software feature written from the perspective of customers or end-users.
Serve as "conversation starters" between developers and users.
Primary focus: Finding out what needs to be done for best system performance.
Each story should be independent and stand on its own.
Goal: Prevent misunderstandings or misinterpretations of requirements.
Scrum as an Agile Methodology
General Philosophy:
Projects begin with a high-level plan that can be changed "on the fly."
The success of the project is paramount; individual success is secondary.
Project leaders have limited influence on specific details.
The team works within a strict time frame, typically weeks for development.
The Three Roles in Scrum:
Product Owner: Represents the interests of the business and maintains the product backlog.
Scrum Master: A facilitator who ensures Scrum practices are followed and removes obstacles.
Team Member: Self-organizes to complete work, generates estimates, and participates in all activities to help the project.
Key Scrum Artifacts and Tools:
Product Backlog: A list of features and deliverables (user stories) reorganized so the most important items appear at the top.
Example Registry: Number "W212" - "Make it easier to continue shopping once an item is placed in the shopping cart" for "Visitors to the website."
Sprint Backlog: The list of tasks the team commits to for a specific sprint.
Sprint Cycle: Usually a -week duration. At the end, the team determines if the product should be released.
Sprint Burndown Chart: A tool to track performance.
Horizontal Axis: Time elapsed.
Vertical Axis: Number of tasks remaining or hours remaining.
Visual Indicators: Red line shows hours of work remaining; yellow/gold bars show the number of tasks remaining.
Scrum Meetings and Mechanics:
Scrum Planning Meeting: Product owner presents the "wish list" of stories; resources are estimated.
Planning Poker: A collaborative estimation technique.
Moderator runs the meeting.
Product owner presents a story.
Members choose cards with numbers representing days, points, or difficulty.
Lower numbers indicate faster completion.
Cards are flipped simultaneously; highest and lowest estimators defend their ratings.
Daily Scrum Meetings: A "stand-up" meeting lasting only a few minutes. Members state what they did since the last meeting, what they expect to do today, and identify obstacles.
Sprint Review: A meeting at the end of the sprint to document completed user stories and note unfinished tasks.
Kanban and DevOps
Kanban Elements:
Visualize the workflow (often using a board with columns: Backlog, Sprint, To Do, Doing, Verify, Done).
Keep work-in-process (WIP) as small as possible.
Reevaluate the workflow and reassign priorities as needed.
Strive for continual improvement.
DevOps (Development Operations):
A cultural shift where development and operations work in parallel streams.
Goal: Decrease deployment time for new applications and maximize profit by rapidly addressing market opportunities.
No-Code Software Development
Shift in Development: Software creation has shifted from professional programmers to nonprogrammers using graphical interfaces.
Mechanism: Drag-and-drop content onto web pages or apps without knowing computer code.
Examples: Wix, Bubble, WordPress.
Advantages:
Cost-effective.
Easier and quicker development.
One person can perform the work of a full team.
Disadvantages:
Apps/websites may not be developed strategically.
Often lacks documentation.
Work Management Systems and Jira
Jira Software: An example of a system that manages and automates "Epics" (collections of user stories).
Features: Includes templates for Kanban, Scrum, and bug tracking.
Epic Documentation involves:
Reporting responsibilities.
Revealing underlying user stories.
Sizing the epic.
Deciding on the time period.
Advantages of Agile Work Software:
Keeps work on track through structured approaches.
Automates processes.
Summarizes and reports progress efficiently.
Comparative Analysis: Agile vs. Structured Methods
Efficiency Strategy | Implementation via Structured Methodologies | Implementation via Agile Methodologies |
|---|---|---|
Reduce interface time/errors | Adopting organization standards (coding, naming, forms) | Adopting pair programming |
Reduce process learning time | Managing update release schedules | Ad hoc prototyping and rapid development |
Reduce time to structure tasks | Using CASE tools and diagrams; reusing code | Encouraging short releases |
Reduce nonproductive work expansion | Managing project/establishing deadlines | Limiting scope in each release |
Reduce data search time/costs | Structured data gathering (interviews, observation) | Allowing for an on-site customer |
Reduce communication costs | Separating tasks; establishing barriers | Timeboxing |
Reduce information overload | Applying filtering techniques to shield analysts | Sticking to a -hour workweek |
Risks and Lessons Learned
Six Vital Lessons:
Short releases allow system evolution.
Pair programming enhances quality.
Onsite customers benefit both business and development.
A -hour work week improves effectiveness.
Balanced resources/activities support goals.
Agile values are crucial to success.
Risks in Adopting Agile:
Fit of development team culture.
Determining the best time to innovate.
Training costs for analysts and programmers.
Client reaction to the new methodology.
Impact of agile on individual rights of programmers and analysts.
Disadvantages of Scrum specific methods:
Improperly documenting features.
Releasing products with errors or too soon for the user.
Completing the backlog under pressure.
Difficulty working with geographically dispersed teams or teams requiring specialized skills.
Difficulty replacing team members who depart.