Use lessons 01–03 or equivalent skills before this exercise. You can revisit them from the lesson list.
A spreadsheet or calculator; optional local Python. Text-only AI can explain a plan but cannot establish that calculations ran.
Start with the idea
Compare options using usable output and repair effort.
Test comparable tasks and record quality failures. Include the time needed to fix them. Unknown repair work prevents a complete savings claim.
Paired tasks: Corresponding tasks used to compare methods under comparable conditions.
See the example and check your understanding
A made-up teaching example
Method A takes thirty minutes for three passing outputs. B initially looks faster, but one output needs six more minutes of repair. Corrected B effort is twenty-four minutes, with a recorded initial failure.
Ask yourself
Can you report usable-output savings while repair time is unknown?
Show the explanation
No. Keep the failure and missing effort visible. A small fixture does not establish unattended reliability.
Start with these checks
- Include rework.
- Preserve initial quality failures.
- State configuration and comparison limits.
Use the different practice case below and check every requirement in its full instructions.
Full learning objectives
- Separate quality, time and failure rates.
- Identify sample imbalance and confounding.
- Preserve configuration and evaluation cases.
Captions appear here during playback.
Compare alternatives without overstating causation
Compare options using usable output and repair effort.
Video playback is unavailable. Read the descriptive transcript below, go directly to practice, or open the video file.
Source-based teaching example · Fictional data · Reviewed 2026-10-03. Microsoft Mark · locally synthesized. On a small screen, use fullscreen for diagram details or read the transcript.
Read the descriptive transcript
0:00 · What you will learn
Compare options using usable output and repair effort.
What appears on screen
Compare options using usable output and repair effort.
0:06 · The idea
Test comparable tasks and record quality failures. Include the time needed to fix them. Unknown repair work prevents a complete savings claim.
What appears on screen
Test comparable tasks and record quality failures. Include the time needed to fix them. Unknown repair work prevents a complete savings claim.
0:18 · One useful term
Paired tasks. Corresponding tasks used to compare methods under comparable conditions.
What appears on screen
Corresponding tasks used to compare methods under comparable conditions.
0:27 · A worked example
Method A takes thirty minutes for three passing outputs. B initially looks faster, but one output needs six more minutes of repair. Corrected B effort is twenty-four minutes, with a recorded initial failure.
What appears on screen
Method A takes thirty minutes for three passing outputs. B initially looks faster, but one output needs six more minutes of repair. Corrected B effort is twenty-four minutes, with a recorded initial failure.
0:44 · Your turn to think
Can you report usable-output savings while repair time is unknown? Pause and decide before continuing.
What appears on screen
Can you report usable-output savings while repair time is unknown? Pause the video if you need more time.
1:02 · Why this answer works
No. Keep the failure and missing effort visible. A small fixture does not establish unattended reliability.
What appears on screen
No. Keep the failure and missing effort visible. A small fixture does not establish unattended reliability.
1:12 · Check 1
Include rework.
What appears on screen
Include rework.
1:16 · Check 2
Preserve initial quality failures.
What appears on screen
Preserve initial quality failures.
1:22 · Check 3
State configuration and comparison limits.
What appears on screen
State configuration and comparison limits.
1:28 · Now try a different case
Open the lesson and try its different practice case. Use every check in the full task. Watching alone does not mark the lesson complete.
What appears on screen
Open the lesson Try the new case Check and revise The full task and its requirements are in the linked lesson.
Watching explains the example. Complete the lesson practice to check what you can do. Explore all visual lessons →
The method
Write evaluation criteria before trying the alternatives. Use the same task set, inputs and review standard; record product, model label, harness, tools and date. Compare usable outputs rather than response speed alone. Keep individual failures so an average cannot conceal a critical requirement violation.
If cases or operators differ, describe the comparison as limited observational evidence. Small trials cannot establish all-task reliability or causal benefit. Repeat representative cases and include adverse inputs when practical. More samples do not rescue an invalid measure. When results are mixed, report the tradeoff rather than forcing a single winner.
Choose your model, harness and offline route
Text-only assistant
Use supplied text. Ask for a plan or draft; do not claim browsing, code execution or file access that was unavailable.
Tool-enabled application or harness
Record which tools actually run, allowed files and actions, approval boundaries and observable results. Tools are optional unless the task needs them.
Local or cloud deployment
Inspect inference, retrieval, connectors, logs and storage separately. Use only authorized environments for real data; training fixtures are fictional.
Higher-capability or frontier model
Evaluate the exact task and configuration. Capability does not supply missing evidence or grant permission. No provider ranking or guarantee is implied.
Offline alternative
Use the supplied inputs and inspect the worked example manually. Builder exercises use local deterministic Python. No paid account or live model call is required.
Record the exact product/model label when available, tools actually used, inputs, permitted actions, observed output and checks. Live accounts or APIs may have external costs; they are optional and are not used by this page.
Read the general workflow guide →Worked example
Supplied input
Constructed paired results: method A takes 8, 10, 12 minutes and all outputs pass. B takes 5, 6, 7; middle output misses a required condition and needs 6 more minutes.
A checked result
A total 30; corrected B total 24. B has one initial quality failure. This fixture shows an effort difference after correction, not a universal benefit or unattended reliability.
Why this works
Time and quality are separate outcomes. A usable comparison includes rework and the failure trace.
Try a different case
Independent paired cases: manual 10, 12, 14; assisted 7, 8, 9. Assisted case 3 fails the required source check and has no repair time recorded. Recommend the next evaluation step.
Download this task and checklist →Interactive practice / Fictional fixture
Account for unusable output
Use the supplied independent case above. Check the bounded decisions, then review your written artifact against the full acceptance criteria.
Everything runs in this browser. No AI model is called and no work is uploaded. Bounded checks cover specific facts and rules. Your writing is not graded: you review it yourself. Practice never marks a lesson complete automatically.
Include the failed case, configuration and a matched quality standard.
Try the task, run the checks, then revise. Checks are teaching feedback, not an instructor grade.
Observed browser simulation
This trace comes from the bounded local checker. It is not a Python run, model result or production tool execution.
Self-review
Locked. Write every written part, then run the checks. A few words is not an attempt.
This is self-review, not grading. Read our answers, then tick what your own work does. Ticks are optional. They stay on this page and are not saved or sent anywhere. A different answer can be just as good if it meets the lesson criteria.
Your next evaluation step
A strong answer
Initial assisted effort was 24 minutes against 36 manual. That is first-pass effort only. Case 3 failed the required source check and no repair time was recorded, so I cannot state a total. Next step: rerun case 3 and record the minutes to repair it, or the decision to reject it. Then run a matched set where both methods must pass the same source check, with the tool version and settings written down. Report first-pass time, repair time and pass rate as three separate numbers.
A flawed answer
Assisted is 33.3% faster. Case 3 was an outlier, so drop it and compare the other two cases next time.
What is wrong with it: 33.3% compares an unfinished output with a finished one. Dropping the failed case removes the one result that shows the cost. A failure is data, not an outlier.
Keep or restore your work
Drafts are not saved automatically. Save only fictional practice data. A browser draft may be lost if storage is cleared; download a copy to keep it.
Your steps
- Report the observed times without treating all outputs as usable.
- Identify the missing repair or rejection cost.
- Define a repeat trial including a source-check failure case.
Check your work
- Manual total 36; initial assisted total 24.
- Do not report a complete usable-output saving while repair cost is unknown.
- Preserve case 3’s failure and configuration limits.
After your attempt: review notes and a common mistake
“33.3% faster” based on 24 versus 36 omits unresolved quality. Report initial effort separately and require repair or rejection evidence.
Different wording or a different justified decision can pass. Check your work against every requirement; these notes are not an automated grade.
Explain your approach
Which failed criterion would override a favorable average?
Keep your artifact, one correction or decision, its evidence and one remaining limitation. Nothing on this page sends or uploads your work.
Check your understanding
Choose an answer to see feedback. This is a knowledge check, not instructor assessment.
Review your independent practice
Complete the supplied task before recording progress. Use the acceptance criteria and reveal the review notes after your attempt. Keep your artifact yourself; nothing is uploaded.
Not yet self-reviewed. Completion stays in this browser and does not issue a badge.
Optional deeper study
Explore the related general-course lesson → This opens the existing canonical lesson; it is optional depth, not another copy of this focused task.
Revision 2026-10-03. Fictional examples, not observed provider results. Knowledge checks and completion records are self-study aids.