Skip to main content

Pre-Semester Launchpad: Answer Key and Solution Outlines

These outlines calibrate self-assessment for the semester-level checkpoint, exam, and project. They are intentionally concise: a correct submission may use different examples or wording, but it must show the reasoning and evidence described here.

Checkpoint Gate Solution Outline​

Preserve your first attempt before opening these criteria. Each numbered item corresponds to the checkpoint: award 1 only when the whole requested explanation and evidence are present; otherwise 0 and name the gap. Require 12/15 and every separate progression gate. Equivalent correct methods count. A new case after solution exposure must change the reasoning demand, not just copy a filename.

  1. A specific learning outcome, capacity arithmetic, start plan and fallback fit together. More planned work than available time is not feasible.
  2. Two separately dated actual observations distinguish prediction, result and help, with an independent learning check. Simulated records do not count.
  3. A concrete access/resume choice addresses an observed obstacle, with a limitation. Do not score health, personality or workspace aesthetics.
  4. Minutes/pages measure activity; an explained changed-case result measures usable learning. Neither proves long-term retention by itself.
  5. The review uses observations to justify an adjustment or unchanged method; recovery is feasible and honestly labeled actual/rehearsed.
  6. The explanation identifies both the folder-relative path and the full location, and distinguishes file contents from navigation. The saved file is actually inspected.
  7. The fixed log's done filter preserves the source; duplicates remain in the record count but not in the distinct-topic list. For the exam fixture: three records, two topics.
  8. Before/after and reopened saved content establish the intended replacement. An unsaved editor display cannot establish on-disk content; explain an applicable undo/recovery method.
  9. A fresh child shell loads the small saved fixture. Removing an alias definition from the file does not erase the already-running alias; remove it from the session or start a clean child, then verify.
  10. Quoting preserves a space-containing argument. The known input, missing-file error/status and separate streams are explained. wc -l counts newline characters; a nonempty unterminated line can yield zero.
  11. Working, staged and committed versions are separately identified. Selective staging commits the chosen change; ignore evidence shows the scratch file is not tracked.
  12. The two edits diverged from one baseline. Show both intentions in the resolved file and merge history; abort returns from an in-progress merge, not an arbitrary undo of all later work.
  13. Specific feedback narrows the claim to the tested same-device boundary, with preserved proposal and revision. Solo role-play is labeled accurately.
  14. An identified earlier state is actually recovered and inspected. Revert adds a reversing commit; stash contents depend on options, and reflog is local and subject to expiry, not a permanent backup. A plan alone fails.
  15. Fetch updates downloaded objects/remote-tracking information without normally integrating the current branch. The stale push would lose remote history; fetch plus deliberate integration preserves both changes. Local clones do not prove remote-device resilience or human review.

Semester Project Solution Outline​

Use the five required sections in the project: study/recovery; small reproducible environment; Git work; preparation evidence; and assessment/handoff index. Link actual artifacts and label unknowns. Do not award completion for decorative screenshots, automation complexity or essay length. Missing observations remain pending; targeted technical failures remain repair required.

The author example packet shows a complete worked illustration, including a deliberate not-qualified status because it is not a learner attempt. It is calibration material after your own work, not an answer to submit.

Semester Exam Answer Key / Solution Outline​

Apply the task-specific calibration below and the exam's four criteria per task. Preserve criterion scores, evidence and reasons. A technical output does not by itself establish the independent explanation criterion. Use the original 25/35/25/15 weights, 75% overall floor and four 70% part floors.

Self-Calibration Procedure​

  1. Attempt the checkpoint, exam, or project review without looking at this page.
  2. Mark every answer or deliverable as meets, partial, or missing against the outline above.
  3. Repair only the highest-risk gaps first: conceptual misunderstanding, unverifiable claims, or missing core deliverables.
  4. Re-grade with the rubric and record the final decision: pass, conditional pass with remediation, or repeat.

Workload and advancement contract​

Use Workload and Progression for the complete hour allocation, prerequisite order, evidence reuse, score thresholds and repair rules. Its estimates do not establish learner-tested duration. All required gates apply together; conditional pass means repair remains, not permission to skip a gate.

Preparation calibration​

Use reading, communication, sustainable-study, digital recovery and AI learning calibration after an attempt. A successful copy with missing intended content fails the restore check. A feasible schedule with fabricated calendar access still needs correction. An observation promised for next week remains pending. These are separate gates alongside the original weighted exam.

Practical exam task calibration​

Use the criterion-level 0/1/2 scoring in the exam. Record each of the four criteria for every task; neither confidence nor speed earns points. Each task percentage is its sum divided by 8, multiplied by 100. Apply the task weights within each part, then the 25/35/25/15 part weights. These are author-designed scoring rules, not validated psychometrics.

  • A1: grouping and compression are different; a successful archive operation can omit an intended file. An independent inventory/content check is needed. Require a new example and an honest attempt/correction record.
  • A2: the resume note identifies the actual file/state and next check. Pausing or using an accessibility aid is not a failure. No induced noise, health disclosure or immediate restart is necessary.
  • A3: the learner explains a chosen process, unsaved versus saved state, and working/staged/committed versions. A concrete correction is better evidence than “I know everything.”
  • B1: notes/plan.txt, notes/notes.txt, logs/session.log and inbox/draft.tmp retain their given content. Moves can be reversed into the original scratch locations. Losing the draft fails preservation even though it was not moved.
  • B2: three records match; two distinct topics are files and reading. Repeated files records still count separately. One valid POSIX pipeline is grep '^status=done ' sessions.log, followed separately by grep '^status=done ' sessions.log | wc -l and grep '^status=done ' sessions.log | sed 's/^status=done topic=//' | sort -u. Explain each stage. This is a fixed teaching format, not a general log parser. Source bytes must remain unchanged.
  • B3: output lines are Open the study folder., Save the study note., Check the saved copy.. Verify the saved file, not only the editor buffer. Equivalent editor workflows earn equal credit.
  • B4: wrong path or an unsaved edit are plausible initial explanations. The supplied files demonstrate a location mismatch. Inspect paths before claiming the root cause; do not delete the draft to hide it.
  • C1: status/history and the ignored-file check support the explanation. A screenshot of a folder does not prove a commit exists.
  • C2: both branches change the same baseline line. A valid resolution says Next: review paths and copies; both intentions survive and no conflict markers remain. The history and final content must be inspected.
  • C3: a same-device copy does not establish device-loss recovery. Require the original diff, specific feedback, corrected claim and truthful reviewer label.
  • C4: demonstrate an actual content recovery. git revert is one valid choice for a committed change; alternatives require explaining their consequences. A plan to recover later is incomplete.
  • D: an inspectable handoff can be brief. Missing linked module evidence remains pending even if the written plan sounds excellent.

For each failed criterion choose a fresh filename, record set, branch edit or source claim before reassessment. Preserve the old result. Do not publish a repository or contact an external maintainer solely to satisfy these tasks.