Find the regression hidden in a revision
Does an improvement for one group hide harm to another?
Entering midway? Revisit Fast for whom? for the preceding worked example, or start here and use the explanation below.
Start with a concrete question
Replay fixed keyboard and touch tasks on three fictional prototypes. Development cases are reused and openly visible.
Try the initial inputs, then write a prediction before changing one setting. All calculations run locally on your CPU.
What do you expect?
Before changing a setting: Does an improvement for one group hide harm to another? Write what you expect and why.
Change one thing. Inspect what happens.
- Compare the combined totals before and after the first revision.
- Inspect keyboard and touch separately.
- Test the second revision against the same cases.
- Write acceptance conditions and propose fresh cases for a later independent check.
Your settingsMechanism
Calculation belowEvidence
Compare outcomes
Compare the current results
Every bar starts at zero and uses the same numeric scale in this chart. The upper bound automatically adjusts to the largest current value (at least 1), so compare the numbers when changing inputs. Units and quantities come from each label; different kinds of quantity are not interchangeable. Displayed values are rounded to at most six significant digits.Follow the calculation
- Keyboard group: baseline 6/10; revision 6/10.
- Touch group: baseline 9/10; revision 9/10.
- The same fixed fictional tasks are replayed for each prototype; this is development evidence, not a fresh independent study.
A revision can improve one group while harming another. Inspect subgroup results and repair regressions before declaring success.
Write a revision and retest record
Record baseline and revised outcomes by group. Identify a regression, propose a repair, and name a fresh task for later testing.
83/8,000 characters. Drafts save locally and are included in backups. Importing or opening a draft never runs it.
Compare with my earlier draft
Combined successes increased, so the revision passed. No further testing is needed.
Explain what you changed and why in your observation notes. Text differences do not establish quality.
Inspect an example solution after your attempt
Keyboard increased 6→9 but touch fell 9→7; total 15→16 hides harm. Repair the touch regression and replay both groups. Then test a fresh form and unfamiliar users; the reused fixture is development evidence.
This is one worked example. Test another input and explain its limitations.
Captures keep your inputs, prediction and observation. Results are recalculated from those inputs. Resetting inputs keeps your captures and challenge history.
Explain the mechanism
Find the regression hidden in a revision
Replay fixed keyboard and touch tasks on three fictional prototypes. Development cases are reused and openly visible.
Follow the mechanism
The first revision raises keyboard successes from 6 to 9 but reduces touch successes from 9 to 7.
An aggregate increase from 15 to 16 hides the touch regression.
Repeated cases support debugging; they do not establish performance on unfamiliar users or tasks.
Worked example
Keyboard group: baseline 6/10; revision 6/10.
Touch group: baseline 9/10; revision 9/10.
The same fixed fictional tasks are replayed for each prototype; this is development evidence, not a fresh independent study.
Calculation
Combined success = keyboard success + touch success when task cohorts are disjoint.
Try a controlled comparison
Compare the combined totals before and after the first revision.
Inspect keyboard and touch separately.
Test the second revision against the same cases.
Write acceptance conditions and propose fresh cases for a later independent check.
What this does not establish
- All messages, people, documents and outcomes here are original fictional teaching fixtures; results are not measurements of real organizations.
Try a fresh case
A revision raises one group from 5 to 8 successes but lowers another from 9 to 7. What is the net change in combined successes?
This is a public practice check. Retries support learning; matching a number does not certify independent proficiency.
Take the investigation further
Choose any extension. These are project prompts for you to carry out; they are not automatically assessed.
School investigation
Create a new case with different numbers and explain your decision to a classmate.
College investigation
Implement or analyze a fresh case, document assumptions and compare at least two alternatives.
Workplace investigation
Prepare a decision artifact with evidence, operating constraints, an owner and a follow-up measurement.
Keep the evidence, choose your next step
Completion records your own decision that you have worked through this investigation. Practice checks and independent project evidence are separate.