It’s time to reap the context-aware harvest

Last week, I was fortunate enough to be asked to speak at two ubicomp-related conferences in San Francisco:

• APPNATION, which focused on the mobile apps explosion — I was on a panel with friends and colleagues Lars Erik Holmquist (inventor of the Bump technique for device pairing, though he didn’t call it that); Mike Kuniavsky (author of Smart Things); and Mattias Rost (a PhD candidate in the MobileLife lab at the Swedish Institute of Computer Science).
• Intel Developer’s Forum, where Justin Rattner talked about the future of context-aware services and devices — I was invited to participate in Rattner’s keynote speech, which naturally focused on Intel innovations, but he also described the heritage of the field beginning with PARC work in the late 1980s and early 1990s. Rattner asked me to come out and speak about the motivations and theories behind the concepts.

From these events, here are some of my reflections on ubiquitous computing (“ubicomp”) then and now. [See my previous post for a modern definition of ubiquitous computing especially compared to All Other Paradigms.]

Personal computing vs. people getting things done

It’s ironic that following the invention of the Personal Computer workstation and laptop computers at PARC, researchers would then turn toward making the computer disappear. To most people at the time, having a single “personal” computer was a dream, but Mark Weiser and many others envisioned that we’d soon all have more than just one personal computer in our lives…

The problem? Well, our social scientists pointed out that the goals of “personal computing” still kept the computer as the focus of attention, rather than what people were trying to get done. People were forced to convert their work processes to fit arcane computer programs, as opposed to the other way around.

The solution? PARC envisioned “ubiquitous computing” systems that could leverage the pervasiveness of devices. By sensing the physical situation, these devices could help shape services so they fit the ways people naturally get things done.

Ubicomp’s main objective has always been to empower individuals and expand their abilities to get things done, without slavish dependence to particular IT infrastructures. The needed technology would be conveniently accessible either on a mobile computer or using computers and displays embedded in the environment. That way, individuals can engage in high-level decision-making, rather than the mundane details of routing, formatting information, etc.

First, the gadgets

Since PARC researchers use prototypes to test their assumptions and make such visions a reality, ubicomp gadgets proliferated here in the early 1990s. Most of these early prototypes were very device-centric (and to some people ubicomp seemed to be all about the gadgets).

But ultimately, these gadgets were just a means to an end. Since the hardware simply wasn’t available, PARC scientists had to create devices and networks to enable services very similar to some of what we’re seeing today, for example:

The Active Badge helped find people in the building (much like Loopt and Foursquare today).	The PARC Tab (with Unistrokes input tech) is an early version of today’s smartphones.
The PARC Pad portable computer… not unlike the tablets hitting the market today.	Since wireless networking didn’t exist, we created an infrared wireless network hub.

Enter context-aware services

Coined just a few years after ubiquitous computing, “context-aware services” (1994) were envisioned as services tailored to your current physical situation: where you are, who you are with, and what you are doing.

Before the web, context-aware services did things like seamlessly connect to the display or projector in a meeting room, or find colleagues and nearby equipment. Most of these were enabled by location detection.

Today, we’re beginning to see these ideas achieve commercial success in the form of Location-Based Services for tracking assets like trucks, freight, and hospital equipment; finding nearby restaurants, stores, and other services; and also just for socializing by finding friends or playing distributed games. Why did it take 20 years to reach commercial penetration? It’s taken that long to build out the enabling infrastructure of high-power handheld devices, broadband wireless networks, widely accessible and affordable networked services, and critical mass of users.

From awareness to intelligence

In the meantime, PARC research has moved beyond “context awareness”, to what we call “contextual intelligence”. Our research is no longer focused on creating the physical devices and overcoming the technical challenges of sensing location, sound, people, weather, or other aspects of the physical environment — today, the devices are commercially available and we focus more on figuring out the right thing to DO with that kind of information.

Contextually intelligent systems can do better than be triggered by current context; they can:

• combine information about the current situation from calendar, email, and location — along with statistical models of past behavior — to proactively predict future situations; and
• accurately target information to personal preferences, interests, and activities.

The result? Services that can help us reach each other at convenient times to talk, conserve energy resources, remember meetings and other important events in our lives, coach us about healthy life choices, and so on.

The seeds were sowed two decades ago but the ground wasn’t fertile enough at the time. With the advent of GPS-enabled smartphones, wideband wireless, and web-based services, it is time to harvest that bounty. For us, context awareness isn’t about devices and location sensing – it’s about people getting things done.