The Tutor Handbook Vol No.0060

The Tutor Handbook · Volume 0060 · Series ID THB-0060

The new route is working.

The learner uses it first.

The tutor reinforces it.

The parent has stopped maintaining the old routine.

The old route is now fallback only.

So when does fallback stop being useful insurance and become unnecessary legacy?

Alicia still has the old seven-step Mathematics checklist in a folder even though her shorter three-check route has now survived school papers, mixed questions and timed work.

Beatrice still keeps the visible paragraph frame that once organised her writing even though she now plans from underlying purpose, evidence and development without needing the frame.

Denise still has the moderate-timing fallback that protected accuracy during recovery, even though her rebuilt whole-paper timing route has now held across several representative papers.

Emily still has the old dashboard login even though her rolling three-day study system has become the actual source of truth.

Faith still has the physical prompt card that once helped her begin unfamiliar work, though the same question now appears inside her own self-talk.

The old systems are no longer primary.

But they are not yet gone.

That half-alive state can be useful for a short time.

Left unattended, it creates another problem.

Old worksheets remain easy to reach. Parents revive old reminders during stressful weeks. Tutors return to old prompts because they are familiar. Learners keep duplicate plans “just in case”. A fallback that should have become memory remains capable of taking over operation.

Decommissioning is the deliberate retirement of an old tutoring route after its function has been safely transferred, so the old route stops accepting ordinary learning work, loses operating authority, leaves behind only the records worth keeping, and can no longer reactivate itself through habit, convenience or hidden dependence.

The previous volume, The Tutor Handbook Vol No.0059 | The Cutover, moved operational authority from the old route to the rebuilt route while preserving bounded fallback.

This volume owns the final transition:

When has the learner earned the right to stop carrying the old route at all—and how can the tutor retire it without throwing away the history, safeguards and recovery knowledge that may still matter later?

Quick Read

Cutover makes the rebuilt route primary. Decommissioning removes the old route from ordinary operation once fallback is no longer justified.
The old route should not be decommissioned until the new route has carried the essential function under enough ordinary variation.
Stop new work entering the old route before removing it.
Finish or reconcile any work that still depends on the old route.
Archive useful records before retiring the old operational system.
Remove duplicate cues, checklists, reminders, dashboards and adult behaviours that can silently reactivate the old route.
The learner should know what has been retired and what principle remains.
Parents should know which support must not quietly return during stressful periods.
Tutors should retire functions, not merely artifacts. Removing a card while continuing the same prompt verbally is not decommissioning.
Necessary accessibility support is not legacy clutter and should not be removed merely because another route has matured.
Do not decommission because the old system looks untidy. Decommission because its function has transferred and its ordinary operation is no longer needed.
Do not keep old routes alive because they once worked. Legacy support consumes attention, creates dual authority and can slow learner ownership.
Records can outlive routes. The learner may no longer use the old system while the tutor still preserves why it existed, what it solved and what failure would justify re-examining that history.
Class 0 to Class 6 tutors retire different kinds of support, control and decision authority.
Learning, studying, training, education and improvement each leave different kinds of useful residue.
Repair, Alignment and Frontier tuition have different decommission conditions.
Three-student tutorials make residual dependence easier to detect because tutor attention naturally moves.
One-to-one tutoring needs deliberate retirement checks because the tutor can keep an old function alive invisibly through live assistance.
Decommissioning is complete when no ordinary learning work depends on the old route, no hidden adult behaviour keeps it active, required knowledge has been archived, and the learner can continue without the legacy system as operational insurance.

1. What This Volume Owns

This volume owns the retirement of a learner-specific tutoring route after a successful cutover.

The broader systems owner is eduKateSG’s How Decommissioning Works | How a System Leaves Service Without Leaving Risk Behind.

That article owns the general lifecycle problem: transfer the function, stop new work, drain what remains, remove residual authority, preserve required records and verify that the old system has genuinely left service.

The Tutor Handbook applies that control logic to teaching without pretending a learner is a machine.

The question here is human and educational:

When a support, routine, scaffold, study system or performance route has genuinely been replaced, how does the tutor let it leave without losing the useful knowledge that replacement created?

2. What This Volume Does Not Own

The Cutover owns the handoff of operational authority while fallback remains available.

The Rollback owns restoration if the newer route becomes meaningfully less reliable.

The Archive owns the wider preservation of useful learning history.

The Retirement owns when a known fragility no longer deserves special monitoring.

This volume is narrower than all four.

It retires an old operating route after replacement.

3. Decommissioning Is a State Transition

Old learning routes should not move directly from active to forgotten.

A useful sequence is:

Primary → Fallback → Restricted → Archive-Only → Decommissioned.

Primary means ordinary work enters the route.

Fallback means the route activates only under defined failure conditions.

Restricted means no new ordinary work should enter it while remaining dependencies are resolved.

Archive-only means its records can still be consulted but the route is not used operationally.

Decommissioned means the old route no longer carries ordinary control, hidden prompts, duplicate planning or active authority.

4. Why Old Routes Stay Alive

Old systems rarely survive because somebody formally decided to keep them forever.

They survive because they are familiar.

The tutor remembers the old prompt.

The parent remembers the old checking routine.

The learner still has the old worksheet.

The dashboard bookmark still exists.

The old study sheet remains in the school bag.

Under pressure, familiarity feels safe.

That is how a fallback quietly becomes a ghost route.

5. The Ghost Route

A ghost route is officially retired but functionally alive.

The tutor says the learner now self-initiates.

But the tutor still asks the old first-step question whenever silence appears.

The family says the learner owns revision planning.

But a parent still keeps a second calendar and sends reminders.

The learner says the new checking routine is enough.

But the old checklist remains beside every paper.

Ghost routes create residual control.

6. Decommission the Function, Not Just the Artifact

Throwing away a prompt card is not enough if the tutor keeps speaking the prompt.

Deleting a study spreadsheet is not enough if the parent continues allocating every task.

Retiring a checklist is not enough if the tutor still performs every check on the learner’s behalf.

Decommissioning is complete only when the old function has transferred or genuinely ended.

Remove the old control, not merely the old object.

7. The Receiver Must Be Proven First

An old route can leave only after something else carries its essential function.

Alicia’s short checking route must catch the error families that matter before the longer checklist can disappear.

Beatrice’s principle-based planning must keep writing coherent across unlike tasks before the visible frame can leave.

Faith’s internal self-question must initiate unfamiliar work before the physical prompt card disappears.

The receiver may be a new routine.

It may also be the learner.

8. Function Transfer Comes Before Retirement

The central decommissioning question is:

Who or what now carries the job the old route used to carry?

If the answer is unclear, retirement is premature.

The job might be:

remembering a deadline;
checking signs;
planning a paragraph;
selecting a method;
initiating unfamiliar work;
allocating revision time;
recovering after a blocked examination question;
recognising when help is needed.

Retire the route only after the job has a verified new owner.

9. Stop New Work Entering the Old Path

Decommissioning begins by refusing new ordinary work.

A learner no longer enters new homework into the old planner.

A tutor no longer introduces the old checklist to new Mathematics topics.

A parent no longer adds new reminders to the old family calendar.

Faith does not use the prompt card on new question types merely because the card still exists.

No new work is the first real retirement signal.

10. Drain the Old Route

Some work may still be inside the old route when retirement begins.

An old study sheet contains unfinished corrections.

A legacy error log contains active weak links.

An old paper-checking routine has one scheduled audit left.

Do not simply abandon those items.

Finish, transfer or consciously close them.

Drain-down prevents the learner from losing useful state merely because the container changed.

11. Transfer State Before Deleting Structure

An old route often contains information the new route still needs.

Emily’s old dashboard may contain three recurring weak links worth moving into the simpler study review.

Alicia’s old checklist may contain one rare high-cost check worth keeping as an occasional audit.

Beatrice’s old writing frame may contain a useful reminder about evidence connection that belongs in the archive as a principle, not a live template.

Transfer the signal.

Then remove the duplicate structure.

12. Records Outlive Routes

The learner no longer needs the old system operationally.

The tutor may still need to remember:

why the route existed;
what problem it solved;
what support level it represented;
why it was replaced;
what failure signal triggered rollback in the past;
what stable gain the old route helped create;
what should never be confused with learner identity.

This belongs to learning history, not active operation.

13. Archive Is Not Operation

An archived route can be readable without being runnable.

The tutor can inspect the old checklist if a similar error pattern returns.

The learner should not automatically resume it.

The parent can remember that close monitoring once stabilised a crisis week.

That does not mean close monitoring should restart after one ordinary difficult evening.

Archive preserves memory while decommissioning removes authority.

14. Remove Duplicate Sources of Truth

If two planners remain active, which one is correct?

If two checking routines remain beside the learner, which one is normal?

If two help rules remain available, which one governs hesitation?

Decommissioning ends that ambiguity.

The new route is authoritative.

The old route is reference only.

15. Remove Legacy Cues

A system can remain alive through cues.

The old checklist is pinned to the wall.

The old formula sheet sits on top of every practice set.

The old prompt card remains face up.

The old calendar keeps sending alerts.

The old dashboard remains the first browser bookmark.

These cues keep the old path cognitively available.

When decommissioning is justified, remove or relocate them.

16. Remove Legacy Adult Behaviour

The hardest legacy system may be a person.

The tutor stops using the old worksheet but still asks every old prompt.

The parent stops keeping the old schedule but still asks every night whether each item is done.

The learner stops carrying the old checklist but waits for the tutor’s facial confirmation after each step.

Artifact retirement without behaviour retirement leaves the function alive.

17. The Residual Dependence Test

Before decommissioning, ask:

If the old route disappeared tonight, what would fail tomorrow?

If the answer is “nothing ordinary”, retirement may be close.

If the answer is “the learner would forget half the deadlines”, the function has not transferred.

If the answer is “the tutor would not know how to detect a rare sign error”, perhaps one audit mechanism still needs migration.

If the answer is “the learner would become anxious because the old object feels safe”, the tutor should decide whether that is a capability issue, a transition issue or merely familiarity.

18. Fallback Frequency Must Approach Zero

A fallback used every week is not decommission-ready.

A fallback used once under an unusual condition may be.

The tutor should look at ordinary operating frequency, not mere existence.

Repeated fallback says the cutover is incomplete or the new route is still missing a function.

Rare fallback after broad success may simply be historical insurance that can now move to archive.

19. The Decommission Gate

An old route is ready to leave ordinary operation when:

the replacement route has already passed the Trial Run and Adoption Gate;
cutover is complete;
the protected invariant remains stable;
ordinary work no longer enters the old route;
fallback frequency is low enough that the old route is no longer part of normal operation;
temporary bridges have been retired or clearly migrated;
parents and tutors are not maintaining duplicate controls;
the learner can operate the new route with appropriate independence;
useful records have been archived;
no legitimate accessibility support is being confused with legacy clutter.

20. Too-Early Decommissioning

The new route works for one week.

The tutor deletes every old support.

Then the learner meets a harder condition and collapses.

The problem was not that the new route was wrong.

The tutor retired the fallback before enough operational history existed.

Cutover creates a period where fallback remains useful.

Decommissioning should come after that period has taught enough.

21. Too-Late Decommissioning

The opposite error is equally common.

The new route has worked for months.

The old route remains available “just in case”.

Every difficult week reactivates it.

The learner never fully trusts the new route because the old route remains easier to reach.

Late decommissioning preserves dual authority and can slow learner independence.

22. Legacy Comfort Is a Real Cost

The old route is familiar.

Familiarity reduces uncertainty.

That can make the learner prefer a weaker old method over a stronger new one.

Preference is worth listening to.

It should not automatically determine architecture.

The tutor should ask whether the old route provides a legitimate missing function or merely familiar comfort.

23. Legacy Complexity Is Also a Cost

Every live route consumes memory.

Which plan?

Which checklist?

Which timing rule?

Which support condition?

Keeping obsolete routines alive can turn learning into route management.

Mature systems get simpler as capability grows.

24. The Principle Must Outlive the Procedure

Decommissioning does not mean forgetting why the old route existed.

A seven-step checklist may retire while the principle “check high-cost errors before submission” remains.

A paragraph frame may retire while the principle “claims need relevant development” remains.

A prompt card may retire while the self-question “what is the task asking?” remains.

Retire the surface when the learner has internalised the function.

25. Decommissioning and Learner Ownership

The most important transfer is often not old method to new method.

It is adult control to learner control.

The parent planner disappears because Emily now allocates priorities.

The tutor prompt disappears because Faith now produces the self-question.

The full checking list disappears because Alicia now recognises high-risk error families herself.

Decommissioning is one of the clearest signs that capability has returned to the learner.

26. Decommissioning and the Learning Budget

Old systems cost time even when they are rarely used.

They require remembering, maintaining, comparing and deciding.

The Learning Budget asks whether legacy support still earns that cost.

When the answer becomes no, retirement frees capacity for current learning.

27. Decommissioning and Intervention Cost

The Intervention Cost applies to old supports too.

A support that once prevented errors can later become an overhead.

A prompt that once opened access can later delay self-initiation.

A family reminder that once stabilised deadlines can later weaken learner ownership.

Past benefit does not create permanent immunity from cost review.

28. Decommissioning and the Decision Record

A consequential retirement deserves a compact record:

old route;
function transferred;
new owner of that function;
evidence that ordinary operation no longer depends on the old route;
records preserved;
artifacts or behaviours retired;
condition that would justify consulting the archive again.

The record protects future tutors from reinventing an old route without knowing why it left.

29. Decommissioning and the Learning Claim

Retiring a support is itself evidence.

It can support a stronger claim about learner control—but only if the learner continues to perform after the support leaves.

“The prompt card was removed” is an intervention fact.

“Faith self-initiates unfamiliar tasks after the prompt card is gone” is stronger evidence.

Support removal and independent capability should not be confused.

30. Decommissioning and the Drift Check

After decommissioning, the Drift Check gains a new job.

It should look for legacy reactivation.

Did the old prompt return during exam stress?

Did the parent recreate a second planner?

Did the tutor rebuild the old checklist after one careless error?

If legacy behaviour returns repeatedly, either the decommission was premature or the new route is missing a function.

31. Decommissioning and Rollback

Decommissioning makes rollback more expensive because the old route is no longer operationally ready.

That is why retirement must come after enough operating history.

But archive knowledge can still reduce recovery cost.

If a similar failure appears later, the tutor can inspect what once worked, what once failed and whether the old cause still applies—without automatically reactivating the old system.

32. Decommissioning and the Scope Boundary

Do not retire legitimate support merely because it looks like extra structure.

Necessary accessibility support may remain part of the learner’s operating environment.

The question is not whether the learner uses support.

The question is whether the support is necessary for access, whether it is appropriately owned, and whether avoidable tutor control can still reduce around it.

The Scope Boundary remains active.

33. Learning Decommissioning

Learning concerns understanding, retrieval, representation and reasoning.

A learning route can be decommissioned when the learner no longer needs a particular external representation, explanation sequence or cue to perform the cognitive operation.

The old representation can remain in the archive.

The learner no longer begins there.

34. Studying Decommissioning

Studying concerns planning, allocation, retrieval, resource use and self-management.

A studying route is decommission-ready when the learner can manage ordinary work without duplicate adult planning, redundant calendars or emergency trackers.

The useful planning principles remain.

The old management infrastructure leaves.

35. Training Decommissioning

Training concerns reliability under repetition, variation, delay and pressure.

A training scaffold can retire when representative practice now carries the capability without needing the easier training condition as ordinary fallback.

Blocked practice may leave as a routine while remaining available as a targeted repair tool if a specific component later becomes unstable.

36. Education Decommissioning

Education places the learner inside school stages, curricula, assessments and institutional demands.

A transition scaffold can retire when the new school stage has become ordinary enough that the learner no longer needs special translation or extra coordination to remain operational.

The tutor should not preserve transition mode simply because transition mode once helped.

37. Improvement Decommissioning

Improvement is the loop from observation through diagnosis, prioritisation, repair, practice, transfer, performance and review.

An external improvement control can be retired when the learner now runs that control reliably enough themselves.

The tutor no longer needs to initiate every review.

The learner notices the drift, identifies the weak link and chooses the next test.

Decommissioning becomes a transfer of improvement ownership.

38. Class 0 · Homework Helper Decommissioning

Class 0 works nearest to task completion and basic routine.

Its legacy systems are easy to see: reminders, task lists, platform checks, submission checks, material checks.

Class 0 decommission rule: retire the adult routine once the learner can locate, begin, monitor and close ordinary work reliably enough without it.

39. Class 1 · Explainer Decommissioning

The Explainer’s legacy system is often repeated explanation.

A concept once needed a full representation sequence.

Later, the learner can reconstruct it from one cue or no cue.

Class 1 decommission rule: retire the explanation pathway as ordinary support when the learner can generate the organising relation themselves; preserve it only as archive or occasional repair reference.

40. Class 2 · Drill Builder Decommissioning

The Drill Builder can keep a completed drill alive for too long because practice is easy to assign.

Once the trained operation survives spacing, mixing and transfer, the dedicated drill may no longer deserve ordinary time.

Class 2 decommission rule: retire dedicated practice when the capability now lives reliably inside richer work.

41. Class 3 · Diagnostic Tutor Decommissioning

The Diagnostic Tutor can also create legacy systems.

A learner remains on a special error taxonomy long after the original diagnostic question has been resolved.

Class 3 decommission rule: retire special diagnostic instrumentation once ordinary performance gives enough signal to monitor the learner without keeping them permanently under investigation.

42. Class 4 · Route Designer Decommissioning

The Route Designer often leaves behind temporary detours.

A prerequisite repair lane remains visible after the learner has rejoined current school work.

Class 4 decommission rule: close the detour when its function has been reintegrated into the main route; preserve the map, not the traffic.

43. Class 5 · Performance Coach Decommissioning

The Performance Coach can leave a learner living permanently in simulation mode.

Full papers, timers, checklists and recovery rules continue long after the immediate examination need has passed.

Class 5 decommission rule: retire temporary performance-season infrastructure when the performance event has passed or the learner has internalised the necessary routines; return capacity to learning and studying.

44. Class 6 · Learning Architect Decommissioning

The Learning Architect is most vulnerable to legacy complexity.

Dashboards, reviews, parent loops, study queues, subject allocation systems and check-in schedules can all outlive the problem that created them.

Class 6 decommission rule: architecture should simplify as capability grows; any component that no longer carries a necessary function should justify its continued existence or leave.

45. Repair Mode Decommissioning

In Repair Mode, special supports are expected.

The danger is repair infrastructure becoming permanent identity.

A repaired weakness no longer needs a dedicated repair worksheet, special alert or full diagnostic protocol once the capability has reintegrated.

Repair Mode decommission rule: retire the repair container when the capability has returned to ordinary use.

46. Alignment Mode Decommissioning

Alignment Mode can create compensatory structures that help the learner stay with school.

Previews, extra summaries and teacher-to-tutor translation may be useful during a difficult period.

Alignment Mode decommission rule: retire compensatory translation when the learner can access ordinary school learning directly enough to remain stable.

47. Frontier Mode Decommissioning

Frontier Mode adds extension beyond immediate requirements.

An extension route can also be retired when it stops earning capacity, when school demand changes, or when a different frontier becomes more valuable.

Frontier Mode decommission rule: extension should remain optional enough that it can leave cleanly without damaging the strong core.

48. Three-Student Tutorials Expose Residual Dependence

In a three-student tutorial, the tutor’s attention moves.

This makes decommission-readiness visible.

Does Alicia still perform the high-value checks when the old checklist is gone and the tutor is with Beatrice?

Does Faith begin without the prompt card while the tutor is helping Ciara?

Does Emily maintain the new study route between tutor interactions?

Small-group teaching naturally reveals whether the old support can leave.

49. Shared Class, Different Decommission Dates

A shared lesson does not require shared retirement.

Alicia may retire a checking scaffold today.

Faith may still need her fallback card for another week.

Denise may retain one targeted timing fallback until another representative paper.

Decommissioning follows individual evidence.

50. One-to-One Tutoring Needs a Behaviour Audit

One-to-one tutors can keep old support alive invisibly because help is always available.

Before declaring an old route decommissioned, the tutor should audit their own behaviour.

Am I still giving the old prompt verbally?
Am I still choosing the first task?
Am I still checking before the learner checks?
Am I still rescuing at the old hesitation threshold?
Am I still keeping a duplicate plan in my notes that silently controls the learner’s plan?

The old system may be inside the tutor.

51. Alicia · Mathematics: Retire the Seven-Step Checklist

Alicia’s seven-step checklist once prevented a family of costly errors.

After rebuild and cutover, the three-check routine—sign, unit, copied value—has survived several school and timed conditions.

For two weeks, the old checklist has caught nothing important that the short routine missed.

Decommissioning begins.

No new practice uses the seven-step version.

The checklist is moved from the learner’s active folder into the tutor archive.

One sentence remains in Alicia’s own system:

Check the errors that cost you most.

The procedure leaves.

The judgement remains.

52. Beatrice · English: Retire the Visible Frame

Beatrice’s old paragraph frame is now more restrictive than helpful.

Her new route begins from purpose and support.

The visible frame moves to archive.

It is not destroyed.

If a future writing collapse resembles the old pattern, the tutor can inspect the frame as historical evidence.

Beatrice no longer writes through it.

53. Ciara · Science: Retire Model Language as the Starting Route

Ciara’s model-answer bank remains useful vocabulary support.

What decommissions is its authority as the first step.

The learner now reconstructs the causal mechanism first.

Then precise scientific language can be selected.

The old route becomes reference material, not reasoning infrastructure.

54. Denise · Additional Mathematics: Retire the Recovery Timing Route

Denise’s moderate-timing route once protected accuracy while she rebuilt whole-paper performance.

Now the rebuilt paper routine has survived several representative conditions without crossing the sign-error or method-selection floor.

The moderate route stops being general fallback.

It remains available only as a targeted repair condition if a specific mechanism later destabilises.

That is an important distinction.

The old route has been decommissioned as a normal system without being erased from the tutor’s toolkit.

55. Emily · Studying: Close the Old Dashboard

Emily no longer uses the old dashboard.

But the login still exists.

The parent still sometimes checks it.

That means the old system is still capable of becoming authoritative again.

Decommissioning transfers the final useful weak-link notes, exports one historical snapshot if needed, stops new entries, removes the family bookmark and closes ordinary use.

One plan remains.

56. Faith · Independence: Retire the Prompt Card

Faith has not used the prompt card for several weeks.

She now asks herself the same question silently:

What is the task asking?

The card can leave ordinary operation.

The tutor records why it once existed and the conditions under which it was useful.

Faith keeps the capability.

The object leaves.

57. Parent Communication: Say What Is Leaving

Parents need to know when support is being retired, especially if the support once felt protective.

We are retiring the nightly parent reminder because Emily’s current planning route has now survived ordinary and heavy weeks without it. The reminder system is not being deleted from history—we know it helped during a difficult period—but it is no longer part of normal operation. If a future problem appears, we will diagnose that problem first rather than automatically restoring the old reminder routine.

This reduces fear without preserving dependence.

58. Learner Communication: Retirement Is Evidence of Growth

The learner should understand what decommissioning means.

This support did an important job. You now carry that job well enough that the support no longer needs to be part of normal work. We are keeping what we learned from it, but you do not need to run the old system anymore.

Retirement should feel like capability returning, not help being taken away arbitrarily.

59. Decommissioning Digital and AI Supports

Digital tools can create unusually persistent legacy routes.

Old templates remain saved.

Old prompts remain bookmarked.

Old AI workflows remain one click away.

If the learner has now internalised the target operation, decommissioning may mean removing the workflow from ordinary access while preserving a bounded record of what the tool once carried.

The aim is not anti-technology.

It is to keep the learner’s cognitive ownership visible.

60. Decommissioning Examination-Season Infrastructure

Exam season produces temporary systems quickly.

Extra full papers.

Closer parent monitoring.

Special timing logs.

Daily error reviews.

When the examination ends, those systems should not automatically survive into ordinary learning.

Retire the emergency infrastructure unless the review identifies a continuing function that genuinely deserves a new route.

61. Decommissioning a Tutor Function

Sometimes the route being retired is not a worksheet or routine.

It is a tutor function.

The learner no longer needs Class 1 explanation as the dominant job.

The learner no longer needs Class 3 diagnosis every week.

The learner no longer needs Class 5 performance coaching after the examination season.

The wider Tutor Classification Model by eduKateSG remains the owner of the classification system.

Decommissioning simply asks when a previously useful tutor function should stop consuming ordinary capacity.

62. The Decommission Card

Old route: What is leaving ordinary operation?
Old function: What job did it carry?
New owner: Who or what now carries that job?
Proof of transfer: What evidence shows the function survives?
No-new-work rule: What stops entering the old route now?
Drain: What unfinished work or state still needs transfer?
Archive: What records or principles must remain?
Artifact retirement: What checklist, sheet, dashboard, prompt or aid leaves?
Behaviour retirement: What tutor or parent behaviour must stop?
Accessibility boundary: What support must remain because it is still necessary?
Legacy reactivation signal: What future pattern would justify consulting history again?
Closure test: What proves the old route is truly no longer operational?

63. The Thirty-Second Decommission Test

What job did the old route carry, who carries that job now, what still depends on the old route, what knowledge must survive, and what can finally stop?

That is enough for many ordinary tutoring retirements.

64. The Closure Test

Decommissioning is complete when operational, behavioural and information states agree.

No ordinary work enters the old route.
No unresolved task still depends on it.
The replacement route carries the essential function.
The learner can operate without routine fallback.
The tutor no longer supplies the old function invisibly.
The parent no longer maintains duplicate support.
Legacy artifacts are removed from ordinary access.
Useful records are preserved in archive form.
Necessary accessibility supports remain intact.
A future difficulty would trigger fresh diagnosis, not automatic legacy reactivation.

65. Research Foundation: Implementation Must Eventually Sustain Without Temporary Infrastructure

The Education Endowment Foundation’s A School’s Guide to Implementation emphasises the sustained use of effective practices in ordinary day-to-day work rather than treating implementation as a launch event.

The Tutor Handbook applies that idea at learner scale.

A rebuilt route has matured when ordinary use no longer depends on the temporary transition machinery that introduced it.

66. Research Foundation: Monitor Progress Before Removing Support

The Australian Education Research Organisation’s Monitor Progress guidance emphasises checking what students know and can do and adjusting teaching in response to evidence.

Decommissioning is therefore an evidence decision.

Support should leave because the learner’s current evidence shows that its ordinary function has transferred—not because adults are tired of maintaining it.

67. Research Foundation: Scaffolds Should Be Adjusted and Removed as Proficiency Grows

AERO’s Scaffold Practice guidance describes supports that are adjusted and gradually removed as learners become more proficient.

Decommissioning is the far end of that same logic.

Once an external support no longer performs a necessary ordinary function, leaving it permanently active can preserve dependence that the scaffold was meant to outgrow.

68. Research Foundation: Self-Regulation Moves Control Into the Learner

The Education Endowment Foundation’s Metacognition and Self-Regulated Learning guidance emphasises planning, monitoring and evaluating learning.

When those controls move into the learner, some external systems should become unnecessary.

A mature learning architecture should show subtraction as well as addition.

69. The Wider Decommissioning Analogy

eduKateSG’s How Decommissioning Works notes that a system can stop doing its job and still leave residual risk, ambiguous ownership and forgotten interfaces behind.

A learner is not infrastructure.

But one analogy transfers carefully:

A route has not truly left service while people still depend on it, while work still enters it, while its old authority can reactivate accidentally, or while nobody has transferred the useful state it contains.

The Tutor Handbook uses that idea to protect learner ownership, teaching coherence and useful history.

70. The Tutor’s Decommission Checklist

Has the new route genuinely completed cutover?
What exact function did the old route carry?
Who or what carries that function now?
What evidence proves the transfer?
How often has fallback been used recently?
What ordinary work is still entering the old route?
What unfinished work or state remains inside it?
What useful history must be archived?
What old artifact should leave ordinary access?
What tutor behaviour still keeps the old route alive?
What parent behaviour still keeps it alive?
What learner habit is still tied to it?
What necessary access support must remain?
What principle should survive after the procedure leaves?
What would prove that decommissioning was premature?
If a future problem appears, will we diagnose it fresh or automatically reactivate the old system?
Does retirement simplify the learner’s operating system without removing genuine capability?

71. The Parent’s Decommission Checklist

What old support is leaving?
What job did it used to perform?
Can my child now perform that job or use the new route reliably?
What should I stop doing at home?
What support still remains appropriate?
What old reminders, calendars or checks should no longer be maintained?
What evidence has shown that the new route holds?
What useful information from the old system is being kept?
If a future problem appears, what happens before the old support is restored?
Is retirement increasing my child’s real ownership rather than merely reducing adult help?

72. The Learner’s Decommission Checklist

What old system am I no longer using?
What job did it used to do for me?
How do I do that job now?
What principle from the old system is still useful?
What object, prompt or reminder can now leave?
What should I do if I struggle again?
What would be a reason to ask for a fresh diagnosis instead of simply reviving the old method?
What part of this change shows that I now carry more of my own learning?

73. What Decommissioning Is Not

It is not deleting learning history.
It is not pretending the old route was bad.
It is not removing support merely to make the learner look independent.
It is not withdrawing necessary accessibility support.
It is not the Cutover.
It is not the Rollback.
It is not the Archive.
It is not the Retirement of a fragility watchlist.
It is not destroying every tool that might someday be useful again.
It is not keeping the old route live forever because it might someday be useful.
It is the controlled removal of an obsolete operating route after its function has transferred and its useful knowledge has been preserved.

74. The Ethical Standard

Support should have an exit.

Not because independence is a slogan.

Because every unnecessary external control occupies part of the learner’s attention, agency and future decision space.

Good tutors build supports carefully.

They should retire them carefully too.

Do not preserve a system simply because it once helped.

Do not remove a system simply because a newer one exists.

Transfer the function.

Preserve the evidence.

Drain the old route.

Remove duplicate authority.

Keep what the learner genuinely needs.

Let the rest leave.

The ethical decommission does not erase the past and does not preserve it as permanent infrastructure. It lets the old route finish its work, transfers the function safely, keeps the knowledge worth remembering, removes the control that no longer belongs outside the learner, and leaves the learning system simpler because the learner has become stronger.

Evidence and Connected Reading

Final Compression

The new route works.

Cutover is complete.

Now ask what the old route still does.

Name the function.

Prove the function has transferred.

Stop new work entering the old path.

Drain unfinished work.

Move useful state.

Archive the history that future diagnosis may need.

Remove duplicate sources of truth.

Remove legacy cues.

Remove legacy adult behaviour.

Keep necessary accessibility support.

Keep the principle after the procedure leaves.

Watch for ghost routes.

Watch for legacy reactivation under stress.

Do not restart the old system automatically when a new problem appears.

Diagnose the new problem.

Let the learner carry what the old route once carried.

Let the learning system become simpler because capability became stronger.

The Decommission is the final act of a successful learning transition: the old route stops accepting work, gives up authority, transfers its useful state, leaves behind only the records worth keeping, and exits ordinary operation so the learner no longer has to carry permanent infrastructure for a problem they have already learned to manage.

That is the Decommission.

That is Tutor Handbook Volume 0060.

Small Group Tutorials

The Tutor Handbook Vol No.0060 | The Decommission — How a Tutor Retires an Old Learning Route After Cutover Without Losing Recovery Knowledge