New Version Planning

As we're still getting started, please feel free to insert notes and comments where you like - just try to avoid changing or deleting things you yourself didn't write. We'll come up with a shared structure as we go. Also consider using the "discussion" tab at the top if you aren't sure where to comment or want to have a dialogue.

These items at the top are on the agenda for discussion, but don't have a clear place they belong yet:

• POST handler/API reachability
• Chat capability
• Containers
• NSJail, chroot?
• Physical systems: controlled via ordinary SSH, provisioned with PXE or other net boot methods?
• How to model shared resources and contention?
• Staged/prerequisite content in guide/scenario
• How to refer to observation/logging instruments

Functional requirements:

• 20-300 students simultaneously? Semi-simultaneously? ~150 concurrently, 500+ at large institutions?
• Telemetry layer for instrument/orchestration health
• Students are provided a learning environment that is safe to explore/attack without threatening the application host infrastructure

Ubiquitous language:

(See What is Ubiquitous Language? | Agile Alliance, Cosmic Python or DDD Resources - Domain Language)

• Scenario
• Learning objectives
  • e.g.
    • "Students will be able to use ls to list the contents of a directory"
    • "Students will be able to log in to the exercise via SSH"
  • Milestone
    • Do these have sufficient resolution to see why/how students arrived at a correct answer?
  • What are indicators of student progress on learning objectives?
    • How do events (such as those matched by milestones) correlate with questions?
      • What are students asking about in help requests?
      • Can we select semantics from log data (commands, chat, questions)?
  • How do we know tangibly that a student is completing learning-related tasks?
• Questions (are exercises distinct?)
• Guide
  • Guide section
• Student
• Instructor
• Hint
• Intervention
• Prediction
• Learning environment
  • Underlying system state?
    • What denotes/distinguishes a command?
• Observability/instrumentation? (Name?)

User stores:

• How do we assess correctness/domain truth in a model of learning objectives:
  • Navigating a filesystem:
    • "Can a student navigate the filesystem?"
      • "Can the student compose an absolute path?"
        • "Does the student use / to begin an absolute path?"
        • "Does the student refer to directories present at the root of the filesystem?"
      • "Do they correctly use find?"
      • "Does the student map correctly between paths and the purposes of the file hierarchy?"
        • "Does the student look for binaries in /bin/?"
      • "Does the student use ~ to refer to their home directory?"
    • What artifacts help to assess this?