Questions I keep returning to

Some questions persist because the answer is difficult. Others persist because answering them too quickly closes useful possibilities. These keep surfacing across projects, reading, conversations and design work.

Status Open question — Deliberately unresolved.

Ideas

Related argument
Operationalise it
Evidence
See it in practice
Follow the concept
Challenge it

?
Question Q01
What should a learner have to do, rather than merely see, inside a digital learning experience?
- Related argumentInteractive learning experiences
- Challenge itA product becomes educational when its interaction changes what the learner must think about.
?
Question Q02
When does difficulty contribute to learning, and when is it simply poor design?
- Related argumentEase of use is not ease of learning
- Challenge itRemoving friction from an interface and removing difficulty from a learning task are different design operations.
?
Question Q03
What can be standardised without suppressing professional judgement?
- Related argumentA design system is not a component library
?
Question Q04
What should educational technology optimise for if engagement is not learning?
- Related argumentEvidence of learning
- Challenge itEngagement is evidence of learning
?
Question Q05
When does personalisation increase agency, and when does it remove productive choice?
- Challenge itPersonalisation is inherently learner-centred

?
Question Q06
What counts as convincing evidence that learning has changed?
- Related argumentEvidence of learning
?
Question Q07
Which behaviours are meaningful signals, and which are merely convenient to measure?
- Operationalise itLearning analytics (opens in a new tab)
- Challenge itThe usefulness of an educational metric decreases when the metric becomes a substitute for the construct it was meant to approximate.
?
Question Q08
Can a school measure something without changing it?
- Related argumentThe analyst inside
- Challenge itThe usefulness of an educational metric decreases when the metric becomes a substitute for the construct it was meant to approximate.

?
Question Q09
How can a system learn from local variation without eliminating that variation prematurely?
- Challenge itA scalable intervention must specify what may change as clearly as what must remain invariant.
- See it in practiceProbing with options open: complexity in practice
?
Question Q10
What happens when a good intervention enters a system that changes it?
- Operationalise itEducation reform: why change is slow and what scales (opens in a new tab)
- Related argumentDispositions before predictions
- Challenge itSuccessful pilots should be replicated faithfully
?
Question Q11
What should remain after an innovation has disappeared?
- Related argumentAsk what persists
- Challenge itThe most durable output of an innovation may be the capability the organisation develops while implementing it.
?
Question Q12
How can a school tell which parts of a plan are complicated and which are complex, before the plan commits it?
- Operationalise itWhen it is needed, and when it is not
- Challenge itMost school plans are part complicated and part complex, and they fail when one logic is used to run both.
?
Question Q13
When should a school stop keeping its options open and commit?
- Related argumentProvisional commitment and reserved optionality
- Challenge itUnder uncertainty, a plan should be judged by the options it keeps open as well as by the outcome it targets.
?
Question Q14
Who pays for the boundary a plan draws, and who gets to say so?
- Related argumentResponsibility for framing
- Challenge itA plan’s boundary is a decision, and the plan should say who it leaves out and who bears the cost.

?
Question Q15
Which decisions should AI make, which should it support, and which should remain deliberately human?
- Operationalise itAI in learning products (opens in a new tab)
?
Question Q16
What should education make scarce when information and generated answers become abundant?
- Operationalise itFutures of education (opens in a new tab)

?
Question Q17
Under what conditions does adding more people improve a decision?
- Related argumentDesigning collective thinking
- Challenge itA group needs structures for independent thought before it can benefit fully from collective thought.
?
Question Q18
How should disagreement be designed rather than merely encouraged?
- Related argumentNobody in the room has to be wrong

Singapore

Questions I keep returning to

Learning and design

Evidence

Systems

AI

Collective thinking