Ubicomp at planetary scale
Wearables tested whether the framework was human-centred. Smartphones and 5G tested whether it scaled. It scaled — and picked up two failure modes that were nowhere in the 2006 text: attention and geolocation.
The thesis was scoped to rooms, buildings and campuses, because that was the size of a deployment anyone could fund. The phone made the deployment planetary without asking, and every assumption about context had to be renegotiated at four orders of magnitude.
The phone as sensor hub
A current handset carries fifteen or more sensors — accelerometer, gyroscope, barometer, magnetometer, GPS, LiDAR — which makes it the most widely deployed ubicomp platform ever built, by a margin nothing else approaches. What makes it a research instrument rather than a parts list is sensor fusion: the readings are useless individually and legible in combination.
Dartmouth's StudentLife study is the clearest demonstration: passive accelerometer and location traces, correlated against sleep, sociability and academic outcomes across a term. It confirmed the contextual-memory argument at a granularity I had no way to picture in 2006, using no instrumentation beyond what the participants already carried.
Invisibility can also mean intrusion
The framework's blind spot is attention. Studies put the average smartphone user somewhere near seventy interruptions a day, and the mechanism producing them is the same context-awareness the thesis was arguing for: the system knows something, so it speaks. Montag and Walla put the cost plainly:
Smartphones create parallel realities where users toggle between physical and digital selves, undermining contextual integrity.
Montag & Walla, on digital distraction
That fragmentation is compounded by sheer volume — notifications, alerts and app demands that disrupt rather than enhance. I optimised for seamlessness and assumed calm would follow. Seamless turned out to be an excellent delivery mechanism for interruption.
5G won latency and leaked location
Sub-millisecond latency did what it promised, and it opened a class of applications the 2006 network could not have carried: location-aware vehicular networks, remote surgery, augmented reality that holds registration while the wearer walks.
Network slicing is the part that matters architecturally — per-context quality of service, so a medical wearable or an autonomous vehicle holds guaranteed bandwidth while everything else degrades. This is the first infrastructure that treats context as a first-class network parameter rather than an application concern.
The same precision is the problem. Around 87% of mobile apps collect GPS data, mostly without granular consent, and millimetre-wave positioning narrows that to sub-metre. The FCC's own framing of the risk:
5G's mmWave precision enables sub-meter tracking — a privacy nightmare for marginalized communities.
FCC, 5G geolocation accuracy, 2023
The reference case is still The New York Times' 2018 investigation. At least 75 companies were receiving nominally anonymous but precise location data from about 200 million phones in the United States, with some apps sampling as often as 14,000 times a day — far more than any user imagined.
The reporting worked because the traces are not anonymous in any operational sense. Lisa Magrin, a maths teacher whose phone was tracked by an app that sold her data, was followed to a Weight Watchers meeting, her dermatologist's office and an ex-partner's home:
It's the thought of people finding out those intimate details that you don't want people to know.
Lisa Magrin, to The New York Times
The Weather Channel app was singled out for sending precise location to dozens of companies, having asked for it to provide local forecasts and traffic alerts — and not disclosing that the same data was analysed for hedge funds. The recipients were not necessarily comfortable either:
When contacted by The Times, some of the companies that received that data described it as 'unsolicited' or 'inappropriate.'
The New York Times, 2018
Senator Ron Wyden, who has pushed privacy legislation in the United States, put the stakes at the level they belong:
Location information can reveal some of the most intimate details of a person's life — whether you've visited a psychiatrist, whether you went to an A.A. meeting, who you might date. It's not right to have consumers kept in the dark about how their data is sold and shared and then leave them unable to do anything about it.
Senator Ron Wyden
Nothing in the standards prevents any of this. At this resolution, consent screens are not a control surface.
Coverage is not access
Detroit's Community Technology Project is the case I keep returning to. Air-quality sensors were installed across the city and published live readings — to residents who in many cases had no smartphone to read them on. The instrumentation was equitable; the interface was not.
My framework assumed the endpoint was already in the user's pocket. Where it isn't, ubiquity means the data exists about a neighbourhood rather than for it. That is a socioeconomic blind spot, not a technical one.
Edge computing, and the energy nobody budgets
Pushing inference to the device is the right architecture on privacy and latency grounds, and it is now the default in wearable silicon — Snapdragon in the Humane AI Pin, the M2 in Vision Pro, both running models locally rather than round-tripping to a datacentre.
It is not free. Dense small-cell deployment is estimated to raise the carbon cost of the same coverage by around 160% against the previous generation. Google's Android Dynamic Performance Framework throttles background work under thermal and grid stress, but that is a platform mitigation sitting one layer above a network that has no equivalent.
Almost no operator runs sustainability-aligned slicing, though the mechanism already exists. If context can prioritise an ambulance, it can deprioritise a background sync during a peak.
Hierarchies of context
Projects like the European Space Agency's 5G-IANA, connecting autonomous vessels, drones and vehicles across continents, need context models that resolve from a room to a continent in one hierarchy. The 2006 model has no vertical axis; it describes a single scale well.
Two phenomena also need naming, because they are structural rather than incidental: temporal fragmentation, where activity splinters into micro-interactions, and digital dualism, where the physical and digital self run in parallel and disagree. Weiser said the most profound technologies disappear.
The most profound technologies disappear… until they demand a charger.
After Weiser
Scale confirmed the design principles and falsified the sociology. Context-aware, privacy-first design still holds up. The assumptions that everyone would have a device, and that a quiet system would produce a quiet life, do not.