Smart Cities SIG

Lesson 5 — Teach It Back

Mission: explain the SIG's semantic capabilities framework clearly enough to onboard a newcomer.

Teach It Back

Capstone — no new capabilities. The goal here is explaining, not just classifying: for each observation, write out loud (or in text) what you'd say to a newcomer, then check yourself against the answer key.

How to use this lesson: for each scenario, (1) split it into sub-observations, (2) assign each one's Primary Capability and any Companion Capabilities, (3) write one plain sentence per sub-observation explaining why — as if a newcomer just asked "wait, why is that one and not the other?" Then open the answer key and compare. Where you disagree with the key, that's exactly what to bring to your teacher.

Scenario A — a streetlight cluster

"Cabinet C12's control unit reported that its cluster of 6 luminaires averaged 85% of rated lumen output over the last night, drawing 2.4 kWh total, while the street below stayed under-lit because three overgrown trees block the light — a known, permanent condition at this site. The cabinet's cellular signal has been weak all week, so the luminaires are currently running their stored nightly schedule instead of waiting for commands."

Reveal answer key

Sub-observation	Primary	Companion(s)
Cluster averaged 85% of rated lumen output	Infrastructure Output	Observation Scope (cluster of 6, not one luminaire), Temporal Semantics (over the last night)
Reported by Cabinet C12's control unit	Provenance	Observation Point (the cabinet is also where this sits physically) — Provenance is primary here only if you want to stress which system reported it over a person; otherwise Observation Point is the safer default per the Lesson 3 rule.
Drew 2.4 kWh total	Resource Consumption	Observation Scope (total across the cluster)
Street stayed under-lit	Service Outcome	—
Trees blocking light, permanent at this site	Physical Context	—
Cellular signal weak all week	Network Operability	—
Running stored nightly schedule instead of waiting for commands	Fallback Behaviour	Network Operability (the weak signal is why)

Note the Provenance row is a genuinely close call — flagging that out loud to a newcomer, instead of pretending it's obvious, is itself good teaching.

Scenario B — a water pressure zone

"The Digital Twin predicts water pressure in Zone 9 will drop below the 40 psi minimum within the next 3 hours, based on a model rather than a live reading, because a heat wave is driving unusually high consumer demand today. Zone 9's booster pump, owned and maintained under contract by AquaCo since 2019, is already at 95% of its rated flow rate."

Reveal answer key

Sub-observation	Primary	Companion(s)
Pressure predicted to drop below 40 psi minimum	Service Outcome	Temporal Semantics (a forecast, within the next 3 hours)
Prediction comes from the Digital Twin, model-based not a live reading	Provenance	Observation Method (model-derived/predicted)
Heat wave driving high consumer demand today	Operational Context	—
Booster pump at 95% of rated flow rate	Infrastructure Output	—
Owned/maintained under contract by AquaCo since 2019	Asset Management Context	—

Worth calling out to a newcomer: the same sub-observation ("prediction from the Digital Twin") carries both Provenance (who/what supplied it) and Observation Method (how it was produced) as tightly linked companions — a model-derived value almost always needs both to be interpreted safely.

Now try one yourself

Write your own 2–3 sentence compound observation from a domain you know — traffic signals, waste collection, EV charging, whatever's familiar — and classify it the same way. Then bring it to your teacher and ask them to check your split, your Primary/Companion calls, and whether your one-line "why" for each would actually land with a newcomer.

This is the lesson to repeat with real cases from your own work, not just once. Onboarding fluency comes from doing this with observations you didn't write yourself — ask your teacher to pull a real example from [[smartcities-sig-lwm2m-mapping]] and quiz you cold.

← Lesson 4 — Operational Semantics Cheat sheet →