Gerald Ajam

How I work

From evidence to changed practice.

Research rarely tells us exactly what to build. Practice requires moving from observation to explanation, from evidence to judgement and from judgement into forms that survive real constraints.

The Atlas & Loom framework

A framework for making sense of challenges that won’t hold still.

Futuring expands what could be. Complexity changes how we understand interaction. Different ways of thinking change how a situation can be framed. I’m working out how the three combine into one framework for making sense of challenges that won’t hold still.

Explore the framework

Questions 01 to 05 of 10

  • 01 Surface the assumptionWhat must currently be believed for the present arrangement to make sense?
  • 02 Change the boundaryWhat happens when we include actors, effects, or timescales currently treated as outside the problem?
  • 03 Change the scaleDoes the problem look different at the learner, classroom, school, system, or societal level?
  • 04 Find the feedbackWhat consequence of an action later changes the conditions that produced the action?
  • 05 Read the dispositionWhat is this situation already inclined to do, without us?

Notes

Arguments being developed

Notes usually begin with a tension, observation, or distinction that seems worth examining more closely. Each ends on what transferred.

2026 · 09The room is an interfaceA classroom organises what learners can do the way a screen does, so it can be designed and questioned like one2026 · 09The robot that became a social contractWe set out to automate rearranging classrooms, and testing showed the hard problem was agreeing how the room is used2026 · 09The unglamorous leversThe cheapest items in a classroom change behaviour most, because each removes a specific friction or dissolves a specific norm2026 · 09Build as you goMost school spaces should be built in increments, trialled and extended as use cases emerge, except the things that must be done properly once2026 · 09A myth is a finding that travelledMost of what education gets wrong began as something true, and the fix is to measure how far it has travelled2026 · 09Every rule of thumb has an opposite numberA rule of thumb is right inside its conditions and wrong outside them, and the conditions can be named2026 · 09A catalogue of faults is half an instrumentCatching a bad argument settles nothing. Arguing better is a short set of moves, and they can be taught2026 · 09Nobody in the room has to be wrongCollective work has failure modes of its own, and more minds do not cancel them2026 · 09A bias is a design briefAwareness is the weakest counter, and the product controls the conditions2026 · 09Ease of use is not ease of learningThe effort that feels like failure is the effort that teaches2026 · 09What design is, and how it gets practised in edtechDesign is a category of decision, taken mostly by people without the title2026 · 09A principle is not a findingA canon that cannot grade itself is a reading list2026 · 07A design system is not a component libraryComponents scale reuse, and principles scale judgement2026 · 06A thinkpiece is a designed objectA writing genre with a job, an audience and a failure mode2026 · 05Design for translationResearch → principles → tools → changed practice
All notes

In the works

Named, not yet written

Listed here so the rest of the site has no dead ends. Each one moves to its proper place when it’s worth reading.

Practice cases

  • When ease of use is not ease of learning · Digital platform
  • From educational principle to usable resource · Curriculum and resources
  • Why a good intervention changes when it scales · System implementation

Notes

  • What counts as evidence of learning? · Evidence of learning
  • Completion is not competence · Evidence of learning
  • Measure the change, not the state · Evidence of learning
  • What dashboards cannot tell you · Evidence of learning
  • Engagement is evidence of activity · Evidence of learning
  • Designing questions that reveal thinking · Evidence of learning
  • Why delayed measures matter · Evidence of learning

Guides

  • Privacy, security and ethics · Craft guide 17, written, not yet online
  • Designing learning spaces · In progress
  • Using immersive media · Planned
  • Designing interactive experiences · Planned

To take away

None of this is finished, the method included.

If you’ve tried one of the questions or moves on a real problem, tell me where it broke. That’s how the next version gets written.

Singapore