Section 318 of 440
Complete canonical tutorial. This reader section contains the same teaching body as PWR-136 · Iterative invention. Open the Power dossier.
PWR-136 · SELF GUIDED full tutorial
Build three paper prototypes, test one uncertainty and revise one feature
Iterative invention turns an idea into information through small reversible tests. This lesson uses three rough paper or card concepts for the cable-reach problem. The learner names the next uncertainty, applies the same test to parallel concepts, records unexpected effects, chooses keep, adapt or reject, changes one meaningful feature and preserves version history. Nothing powered, load-bearing, medical, transport, fire, chemical or safety-critical is built.
1 · Permission and limits
Know exactly what you may do
2 · Get ready
Gather what you need and check the starting conditions
What you need
- Cable-reach brief, paper, card, removable tape, string used only on the tabletop and scissors used in the ordinary safe way.
- Three prototype cards, shared test rubric and version log.
- Criteria for one-movement reach, snagging, cable thickness, removability and new problems.
Before you start
- Confirm prototypes are non-functional representations and no electrical device is modified.
- Write hard stops for damage, trip hazard, pinch, sharp edge and inaccessible use.
- Choose one uncertainty that the first cycle will answer.
3 · The method
Follow these steps in order
- Name the next uncertainty
Write one question, such as whether the cable end can be reached in one movement without extra snagging.
Why: A prototype should answer a question rather than look finished.
Check: The question has an observable answer and stop rule.
- Build three rough alternatives
Make a desk-edge guide, counterweight representation and storage pocket at equal low detail before refining any one.
Why: Parallel prototypes reduce commitment to the first idea.
Check: Three mechanisms are testable with comparable effort.
- Predeclare a common test
Define success, failure, hard stops and how each concept will face the same paper-cable walkthrough.
Why: A common test supports comparison and limits hindsight.
Check: The rubric is fixed before trials.
- Run and preserve raw evidence
Test each once, record result, unexpected effect and hard-stop status without repairing during the run.
Why: Unedited evidence shows what the version actually did.
Check: Each prototype has a complete result row and photo or sketch.
- Choose keep, adapt or reject
Apply the rubric and write the decision for every concept. Do not rescue a hard-stop failure with a high average.
Why: Explicit decisions keep iteration evidence-led.
Check: Every decision cites a result and gate.
- Change one feature
For one adapt decision, alter one meaningful feature only and label version two.
Why: A single change makes the next result interpretable.
Check: Version two differs in one recorded feature.
- Retest and stop intelligently
Repeat the same test, compare criterion and new problems, then stop when the uncertainty is answered or further work is not justified.
Why: Iteration is not endless polishing.
Check: The final log states improvement, no improvement or termination and why.
- Check delayed repeatability
After 48 hours, rebuild the chosen paper version from the lineage and test record without using the original object. Run the same rubric once more and explain any changed result.
Why: A reconstructable invention record should preserve the critical feature beyond one lucky build or immediate memory.
Check: The delayed version matches the declared feature, and any score change is traced to build variation, measurement noise or a real failure.
4 · Worked example
See the whole method used once
Scenario
Three paper concepts represent a desk-edge guide, counterweight and fabric-like storage pocket.
Walkthrough
- Nico asks which concept makes the paper cable end reachable in one movement without snagging.
- He builds all three roughly before decorating any one.
- The guide succeeds on reach but catches thick cable; counterweight adds desk-edge movement; pocket needs two movements.
- He keeps the guide, rejects the unstable counterweight and adapts the pocket only as a backup.
- For guide version two he widens one channel feature, retests the same paper cables and records improved thick-cable passage with no new hard stop.
Result
Nico completes a traceable one-variable improvement in a paper fixture. He has not built, electrically tested or approved a real product.
5 · Right and wrong
Compare correct or safer execution with the common wrong version
| Moment | Right / safer | Wrong / riskier | Why it matters |
|---|---|---|---|
| Prototype purpose | Answer one named uncertainty. | Make the concept look finished. | Finish quality can hide that no question was tested. |
| Breadth | Build three rough mechanisms first. | Polish the favourite before alternatives. | Early commitment reduces information from the search. |
| Revision | Change one feature and keep version one. | Redesign everything and overwrite the old file. | Multiple changes destroy attribution; deletion destroys the decision trail. |
| Stopping | Stop when evidence answers the question or cost exceeds value. | Iterate indefinitely because invention requires persistence. | Iteration without a decision rule wastes effort and can add complexity. |
6 · Common mistakes
Spot the error and apply the correction
| Mistake | Fix |
|---|---|
| Test criteria change after a favourite fails. | Freeze the rubric for the full cycle and revise only on a fresh cycle. |
| Unexpected effects are omitted. | Give every run a “new problem” field even when blank. |
| Hard failures are averaged. | Use pass/fail gates before scored criteria. |
| Version two changes shape, material and test. | Alter one feature and preserve the same test. |
7 · Practice
Turn the steps into a usable skill
First session
- Name one uncertainty.
- Build three rough paper alternatives.
- Run one common test.
- Choose keep, adapt or reject.
- Make and retest one single-feature revision.
Repeat plan
Complete one two-cycle benign project every two weeks for six weeks. On the final project, transfer the question–prototype–test–decision method to a different harmless brief and have another reviewer reconstruct the versions.
Progress when
- Each prototype answers a named uncertainty.
- One-variable revisions produce interpretable change.
- Version history and attribution are complete on a new brief.
Do not progress when
- The build becomes powered, load-bearing, clinical, transport or otherwise hazardous.
- Real people would be tested without consent and authority.
- Iteration continues without a remaining decision-relevant uncertainty.
8 · Check the result
Measure what changed
Criterion change and error correction across versioned benign prototypes.
How: Record common-test results, hard stops, unexpected effects and reviewer-rated value, feasibility and error correction for versions; compare only matched tests.
Good result: A single-feature version improves the predeclared criterion on two matched runs without a new hard-stop failure, and the trail is reconstructable.
This does not prove: It does not establish objective originality, real user value, engineering safety, manufacturability or commercial invention.
Self-check
- What single uncertainty is being tested?
- Did all three concepts face the same rubric?
- What evidence caused keep, adapt or reject?
- Did version two change only one meaningful feature?
9 · Stop, adapt or get help
Keep the safety boundary practical
Stop and get help
- Stop before connecting to electricity, body, vehicle, pressure, load, heat, flame, chemical or live service.
- Stop when sharp edges, trip hazards, privacy risk or inaccessible use appears.
- Seek domain, safety, rights and affected-user review before functional prototyping or implementation.
Accessibility and adaptations
- Use storyboards, cardboard, tactile mock-ups or accessible interface wireframes.
- Allow a maker assistant while the learner controls the hypothesis and decision log.
- Use low-force tools and pre-cut materials without changing the design question.
10 · Evidence and limits
Why these instructions are here
- primary research
The Fusion Innovation Test decomposes real-world creative problem solving into measurable generation, selection and revision features but does not establish general trainability.
Introducing the fusion innovation test as a new paradigm for studying realworld creative problem solving - primary research
An experimental study found leader behaviours aligned with innovation requirements affected performance, showing that configured process conditions matter.
Aligning Leader Behaviors With Innovation Requirements Improves Performance: An Experimental Study
Limits
- Only benign, reversible representations are built.
- Matched improvement answers one prototype question, not overall product success.
- Real invention requires domain validation, rights, safety, users and implementation evidence.
Open the complete canonical research register
- Primary empirical supportLimiting / contraryIntroducing the fusion innovation test as a new paradigm for studying realworld creative problem solving
Wu CT; Kern FB; Su YS; Street M; Chao ZC · 2025 · Primary research
- Primary empirical supportLimiting / contraryAligning Leader Behaviors With Innovation Requirements Improves Performance: An Experimental Study
Gerlach F; Heinigk K; Rosing K; Zacher H · 2020 · Primary research
- Limiting / contraryOfficial boundary contextCopyright and Artificial Intelligence
United States Copyright Office · 2025 · Official guidance
- Limiting / contraryOfficial boundary contextTraditional Cultural Expressions
World Intellectual Property Organization · 2026 · Official governance
Read the complete evidence interpretation on the Power dossier.
Tutorial delivery controls
Learn, adapt, troubleshoot and resume
Progress is saved only in this browser on this device.
Step-by-step learner mode
Each activity includes its success check, a nearby accessible alternative and an “I’m stuck” correction path. Alternatives preserve the target where possible; when they change the task, Titan labels them as related rather than equivalent.
Name the next uncertainty
Write one question, such as whether the cable end can be reached in one movement without extra snagging.
A prototype should answer a question rather than look finished.
The question has an observable answer and stop rule.
I’m stuck on this step
Reset: Re-read this authored instruction — “Write one question, such as whether the cable end can be reached in one movement without extra snagging.” — and its success check, then attempt only this step.
Possible snag: The result from “Write one question, such as whether the cable end can be reached in one movement without extra snagging.” does not yet meet this declared check: The question has an observable answer and stop rule.
Correction: Return to the start of “Name the next uncertainty”, reduce complexity or pace, and repeat only the part needed to satisfy: “The question has an observable answer and stop rule.”
Stop / get help: Stop before connecting to electricity, body, vehicle, pressure, load, heat, flame, chemical or live service.
Build three rough alternatives
Make a desk-edge guide, counterweight representation and storage pocket at equal low detail before refining any one.
Parallel prototypes reduce commitment to the first idea.
Three mechanisms are testable with comparable effort.
I’m stuck on this step
Reset: Re-read this authored instruction — “Make a desk-edge guide, counterweight representation and storage pocket at equal low detail before refining any one.” — and its success check, then attempt only this step.
Possible snag: Hard failures are averaged.
Correction: Use pass/fail gates before scored criteria.
Stop / get help: Stop before connecting to electricity, body, vehicle, pressure, load, heat, flame, chemical or live service.
Predeclare a common test
Define success, failure, hard stops and how each concept will face the same paper-cable walkthrough.
A common test supports comparison and limits hindsight.
The rubric is fixed before trials.
I’m stuck on this step
Reset: Re-read this authored instruction — “Define success, failure, hard stops and how each concept will face the same paper-cable walkthrough.” — and its success check, then attempt only this step.
Possible snag: The result from “Define success, failure, hard stops and how each concept will face the same paper-cable walkthrough.” does not yet meet this declared check: The rubric is fixed before trials.
Correction: Return to the start of “Predeclare a common test”, reduce complexity or pace, and repeat only the part needed to satisfy: “The rubric is fixed before trials.”
Stop / get help: Stop before connecting to electricity, body, vehicle, pressure, load, heat, flame, chemical or live service.
Run and preserve raw evidence
Test each once, record result, unexpected effect and hard-stop status without repairing during the run.
Unedited evidence shows what the version actually did.
Each prototype has a complete result row and photo or sketch.
I’m stuck on this step
Reset: Re-read this authored instruction — “Test each once, record result, unexpected effect and hard-stop status without repairing during the run.” — and its success check, then attempt only this step.
Possible snag: Unexpected effects are omitted.
Correction: Give every run a “new problem” field even when blank.
Stop / get help: Stop before connecting to electricity, body, vehicle, pressure, load, heat, flame, chemical or live service.
Choose keep, adapt or reject
Apply the rubric and write the decision for every concept. Do not rescue a hard-stop failure with a high average.
Explicit decisions keep iteration evidence-led.
Every decision cites a result and gate.
I’m stuck on this step
Reset: Re-read this authored instruction — “Apply the rubric and write the decision for every concept. Do not rescue a hard-stop failure with a high average.” — and its success check, then attempt only this step.
Possible snag: The result from “Apply the rubric and write the decision for every concept. Do not rescue a hard-stop failure with a high average.” does not yet meet this declared check: Every decision cites a result and gate.
Correction: Return to the start of “Choose keep, adapt or reject”, reduce complexity or pace, and repeat only the part needed to satisfy: “Every decision cites a result and gate.”
Stop / get help: Stop before connecting to electricity, body, vehicle, pressure, load, heat, flame, chemical or live service.
Change one feature
For one adapt decision, alter one meaningful feature only and label version two.
A single change makes the next result interpretable.
Version two differs in one recorded feature.
I’m stuck on this step
Reset: Re-read this authored instruction — “For one adapt decision, alter one meaningful feature only and label version two.” — and its success check, then attempt only this step.
Possible snag: Version two changes shape, material and test.
Correction: Alter one feature and preserve the same test.
Stop / get help: Stop before connecting to electricity, body, vehicle, pressure, load, heat, flame, chemical or live service.
Retest and stop intelligently
Repeat the same test, compare criterion and new problems, then stop when the uncertainty is answered or further work is not justified.
Iteration is not endless polishing.
The final log states improvement, no improvement or termination and why.
I’m stuck on this step
Reset: Re-read this authored instruction — “Repeat the same test, compare criterion and new problems, then stop when the uncertainty is answered or further work is not justified.” — and its success check, then attempt only this step.
Possible snag: The result from “Repeat the same test, compare criterion and new problems, then stop when the uncertainty is answered or further work is not justified.” does not yet meet this declared check: The final log states improvement, no improvement or termination and why.
Correction: Return to the start of “Retest and stop intelligently”, reduce complexity or pace, and repeat only the part needed to satisfy: “The final log states improvement, no improvement or termination and why.”
Stop / get help: Stop before connecting to electricity, body, vehicle, pressure, load, heat, flame, chemical or live service.
Check delayed repeatability
After 48 hours, rebuild the chosen paper version from the lineage and test record without using the original object. Run the same rubric once more and explain any changed result.
A reconstructable invention record should preserve the critical feature beyond one lucky build or immediate memory.
The delayed version matches the declared feature, and any score change is traced to build variation, measurement noise or a real failure.
I’m stuck on this step
Reset: Re-read this authored instruction — “After 48 hours, rebuild the chosen paper version from the lineage and test record without using the original object. Run the same rubric once more and explain any changed result.” — and its success check, then attempt only this step.
Possible snag: Test criteria change after a favourite fails.
Correction: Freeze the rubric for the full cycle and revise only on a fresh cycle.
Stop / get help: Stop before connecting to electricity, body, vehicle, pressure, load, heat, flame, chemical or live service.
Correct versus incorrect execution
These accessible process diagrams are built from the tutorial’s own right/wrong teaching. They are not anatomical illustrations and do not add technique beyond the canonical tutorial.
Answer one named uncertainty.
Make the concept look finished.
Build three rough mechanisms first.
Polish the favourite before alternatives.
Change one feature and keep version one.
Redesign everything and overwrite the old file.
Stop when evidence answers the question or cost exceeds value.
Iterate indefinitely because invention requires persistence.
Method-structure checklist
10 of 10 structural checks present
- Ordered, Power-specific instructions — present
- Every activity has a success check — present
- Materials or supplied records are declared — present
- Measurement or assessment rule is present — present
- Tutorial-specific troubleshooting is present — present
- Stopping or escalation boundary is present — present
- Every activity has an adjacent alternative — present
- Correct-versus-incorrect comparison is present — present
- Evidence context is bound to the Power record — present
- Planning metadata is present — present
The method-readiness band and presence checklist assess tutorial presentation and are separate from evidence quality for the underlying Power. They are automated editorial aids, not human approval.
Manual editorial sign-off: Pending. This tutorial must not display a human-approved state until an identified editor signs the exact content hash.