Last updated: Dec 08, 22 20:13 UTC | Permalink

Project Plan Due Wednesday October 12, 11:00am ET

All projects will involve frontend and backend development of a new feature for Covey.Town. Once teams have been formed, you and your team will decide what kind of new feature you would like to build. Your feature should be something that can be implemented within the timeframe allotted (5 weeks, plus 2 weeks of planning), and will be implemented in a fork of the main Covey.Town codebase. You can play with a demo deployment of the app at app.covey.town, and in the coming weeks, we will provide tutorials and instructions for you to run the entire application in a local development environment, and also to deploy it to the cloud. Given that you will be up-to-speed on the Covey.Town codebase (and have been introduced to TypeScript, React, NodeJS, and testing frameworks), and that you will have a team of four, we expect that the feature that you propose will be at least as complex as the feature implemented in the individual homeworks.

Feel free to look at existing systems like Gather.Town, Sococo, Reslash, Screen, and Gatherly.IO for inspiration on new features to build for Covey.Town. Also see the recent NYTimes Magazine article The Race to Fix Fix Virtual Meetings (click here to access through NEU libraries). Examples of features that students might propose include:

Create some new form of “Interactable” object, such as a whiteboard or game
Create an interface for uploading and choosing between different maps and avatars (will require also learning to use the Phaser3 API)
Add screenreader support - generate a textual representation of the map and what players can do to interact with it
Support real persistence: store data in a database (e.g. Postgres + GraphQL), allow users to register and save a profile (e.g. using Auth0)
Add direct messaging, image messaging, and other chat features

Please note that multiple teams might choose to propose the same feature, or a variation of that same feature - this is OK.

When considering your project, please keep in mind that you will be allowed to publicly post your project online: while your immediate audience for the project is the course staff, if you are ultimately looking for software engineering jobs or co-ops, this project can be a useful piece of your portfolio. If you build a sufficiently maintainable feature (i.e., if your project is particularly well architected and tested), we will also consider pull requests to merge your feature into the main Covey.Town codebase on GitHub, allowing you to also tell recruiters that you have contributed a feature to an open source project on GitHub!

The project plan will include:

Introductory problem statement
User stories and acceptance criteria: high level description of how users will interact with your new feature.
Work breakdown: Define engineering tasks that will be necessary to implement your new feature. Map each task to a sprint.

Your assigned TA mentor will review your project plan and provide you with feedback on the scope and details provided in this plan. Your team will self-organize, as agile teams do, and will enhance and adapt its plan during the project lifecycle. As such, the primary goal of this document is to begin the planning process, and not to produce a detailed plan that must be followed precisely. The course staff will provide feedback on your project to help ensure that the scope of your project is appropriate.

We list page maximums for each section as general guidance of what we are willing to grade. Please do not feel compelled to use all of the pages provided, and remember that a diagram or table can be as expressive (or more) as a comparable amount of text.

Problem Statement, User Stories and Acceptance Criteria (max 4 pages)

Your project plan should begin with a 1-3 paragraph introductory problem statement: what problem in Covey.Town does your (proposed) feature solve? Provide a paragraph (or two) that describes why you are interested in building this feature.

Given the problem statement, develop three user stories that show how a user would interact with the feature. User stories are requirements specified in the format “As a < type of user >, I want < some goal > so that < some reason >.” My conditions of satisfaction are < list of common cases and special cases that must work >.

You should include three different user stories to describe how users will interact with your feature. Your three user stories should cover the key behavior that your feature will provide. Do not provide more than three user stories. Your problem statement and description of user stories and conditions of satisfaction should be between 2-4 pages.

Work Breakdown (max 10 pages)

Given the project concept that you have chosen and the functionality that you expect to implement to satisfy your user stories, define a breakdown of the work that will be necessary to complete the project.

A work breakdown includes all of the tasks necessary to accomplish the project, and will be an artifact that we will refer back to throughout the project to evaluate whether you are making satisfactory progress. Consider all of the kinds of tasks that your team will need to perform, including knowledge acquisition, design, implementation, testing and documentation tasks. It is hard to say (generically) how many work items are necessary.

Each task on the work breakdown should be assigned to exactly one team member, who should provide a “T-Shirt” estimate for how long it will take (along with a justification for that estimate). Consider the dependencies between tests: perhaps an API needs to be designed and specified before implementation can begin; perhaps your development environment needs to be configured before anything else can proceed. Assign tasks to sprints considering these dependencies.

Given the preliminary nature of your project, we do not expect that you will know all of the details about precisely how to implement your feature such that you could break it down into tasks that you feel could be implemented in a day or two. However: Large tasks (those which you can not provide a responsible estimate for) must be accompanied by smaller “research” tasks that can be performed early on in the project, providing clear deadlines by which the task must either be refined into smaller tasks (based on new knowledge gathered), or reworked/abandoned. You might consider even scheduling some of these research tasks to take place during Sprint 0 (immediately after submitting this document).

For example: Consider if you were proposing the “Viewing Area” project (the individual project), without the experience of having completed it. It might be difficult to consider how to break down a task like “Implement the frontend components for sychnronized video playback” into something that you could commit to doing within a day or two. Given that this is a task that can be delayed until the end of the project (no other tasks depend on it), it would be wise to consider having some tasks early on in the project, such as: “Find react components that embed YouTube videos,” and “Implement simple video player that does not synchronize playback.” Completing these smaller tasks early would let you both demonstrate that some forward progress is being made, and also allow you to create a much more responsible estimate for how that last, otherwise insurmountably large task would take.

Be realistic, and leave time for contingencies (including the time around the midterm exam on Nov 4). Remember that you will need to have a demo prepared of your feature by 11/30 - just 6 weeks from the due date of this assignment. If you are uncertain that some tasks will be feasible, then be sure to include evaluation tasks earlier-on in the project that will allow for “go/no-go” decisions to move forward on a task or drop it.

We understand that it is quite difficult to estimate the technical complexity of a new project (as you are doing in the case of this course). We will provide you with feedback on this preliminary project plan, which you will use to produce a revised project plan (due Oct 26). Throughout the project period, teams will meet regularly with their dedicated TA Mentor, who will help track progress on a week-to-week basis and help to determine when adjustments to the project scope are needed.

Each work item should contain the following information:

Task to be performed
User story (or stories) that this task relates to
Team member responsible for completing the task
T-shirt size estimate of how long will be needed to complete the task, using the following buckets:
- Small: Can likely be completed by one team member in one sitting of less than 3-4 hours
- Medium: Likely to require involvement of multiple team members, over the course of 1-2 days
- Large: Currently unable to provide a responsible estimate.
A brief (1-2 sentence max) justification of how you reached the size estimate of the task
Milestone for delivering the task, chosen from one of the following:
- Sprint 0: Oct 12-Oct 26
- Sprint 1: Oct 27-Nov 2 (Sprint 1 is just a single week)
- Sprint 2: Nov 3-Nov 16
- Sprint 3: Nov 17-Nov 30

Your work breakdown should take the format of a simple textual list.

Submission

Your project plan should be submitted as a single PDF in GradeScope to the assignment “Preliminary Project Plan.” Each team submits a single document to GradeScope: when uploading the submission, GradeScope will ask you who your teammates are, and then will associate this submission with every member of your team. Be sure to tag your team mates. This assignment is due Oct 12 at 11am.

Grading

The project plan will account for 10% of your project grade, and will be graded out of 100 points. The grading of the project plan is further broken down as follows:

Introductory problem statement (5 points):

Receive full marks if there is a narrative consisting of 4-10 sentences that describes a specific problem that your project aims to solve.
Receive partial credit if the narrative is present, but does not describe a problem that the project aims to solve

User stories (15 points):

Each of the three user stories will account for 5% of your grade on this assignment and graded as follows:

Receive full marks if:
- The user story fits the problem statement
- The user story satisfies the INVEST criteria for good user stories (construed quite broadly)
- The user story includes conditions of satisfaction that cover the “normal” expected behavior of the feature, and any relevant error cases

Work breakdown (80 points):

Your work breakdown will be evaluated holistically on the following rubric:

Coverage of tasks needed (20 points):

Receive full marks if the work breakdown includes all (reasonably expected) tasks to implement your feature, considering these kinds of tasks:

Background research
Design of interfaces and data types
Deployment of third-party services
Implementation
Testing
Documentation

It is not possible to state generically for all projects whether all of the above types of tasks are necessary. However, we believe that this list is exhaustive (we do not expect other kinds of tasks).

Assignment of tasks (10 points):

Receive full marks if:

Each element on the work breakdown is assigned to one team member
The distribution of tasks of each size are roughly similar between the whole team (no single person is assigned significantly more or fewer tasks of one size)

Sizing of tasks (40 points):

Receive full marks if each element on the work breakdown:

Has a size estimate (small, medium, or large) that is provided by the team member assigned the task.
Has a responsible justification for that estimation
Every “large” task:
- Is accompanied by a reasonable explanation of why the team is unnable to provide a responsible estimate
- Is accompanied by at least one small or medium task, scheduled well-before the “large” task is due to be completed. We would expect that most of these research tasks are scheduled to sprint 0, or perhaps sprint 1.

Scheduling of tasks (10 points):

Receive full marks if each element on the work breakdown:

Is assigned to a sprint
There are no obvious constraint violations (tasks that logically must happen before others should be scheduled before them)
There are no “Large” tasks scheduled in sprint 0