Yesterday I had a chance to read a story that discusses an important aspect of activity tracking; these devices might encourage us to exercise and walk more, but we don’t know yet what their long-term impact might be. The article claims that various institutions (hospitals, universities, healthcare organizations) are conducting trials to determine how digital data trackers affect activity levels, obesity and dietary intake in various populations. I am really looking forward to reading more about these studies.
As it’s quite obvious from my previous posts, I’ve been thinking quite a bit about the large number of activity tracking devices that have been released in the last few months. Literally, it seems like a week doesn’t go by without the release of a new wristband, or watch, or mobile app that claims to track our activities and help us get healthier. But the truth is, we have a lot to learn. This is one of those cases where products are being developed because we have the technology to build them, and people seem interested in buying them.
There are two ways of looking at this. One one hand, we don’t have a solid scientific foundation in place to support this trend quite yet, at least not in the traditional sense. At the same time, as many self-quantifiers would say, the rules of the game are changing, and n=1 is what counts.
Mark Abel, a colleague and principal investigator at the Intel Science and Technology Center for Pervasive Computing, sent out an email with a link to an article about add-ons and accessories for smartphones. The author of the article makes one point that I think is particularly relevant:
There’s no need to duplicate what a smartphone already does in a new CE device under development. In some cases, there’s no need to develop hardware at all. A smartphone already exists as a hardware platform, on which you can design — in software — whatever functions you want it to perform.
I have been thinking about this point in the context of ubiquitous computing research for some time. Many believe, rightly so, that we are in the middle of a hardware renaissance of sorts. Over the last decade, a number of factors have greatly reduced the time and cost that goes into designing and prototyping a custom hardware device. At the same time, we see certain devices, particularly smartphones, that are becoming universal hardware platforms. Why build a new item from scratch if you can simply build an add-on to the smartphone people already own and (hopefully) love?
The answer, of course, depends on the exact type of research that one is pursuing. Today, there is a large group of researchers actively developing and experimenting with custom hardware (e.g. the UIST community). At the same time, I do see an upward trend towards leveraging existing platforms. For example, highly respected researchers in the community such as Tanzeem Choudhury, Andrew Campbell, and Deborah Estrin, have recently centered their work much more around existing mobile platforms and apps than custom implementations and sensor networks. Computer vision researchers like Dieter Fox have been actively exploring the XBox Kinect depth camera in a variety of ways. It is worth noting that there is an obvious level of opportunism at play here – if the hardware is available and interesting, why not use it? – but I believe my point stands.
There is tremendous value in the “imagine and realize” mentality of custom hardware. Innovation is what one usually associates with this process, as in “build new things”. But there is another angle on custom devices that I find particularly promising – it has to do with access. When the overall cost of building hardware goes down, it is possible to build not only a better device over time, but also take an existing device that costs $100, such as the Fitbit activity tracker, and reinvent it for $10. Why? To turn it into an almost “disposable” device that you can make available to lots of people. This results in the ability to scale studies and experiments.
On the other hand, there is an enormous amount of low-hanging fruit around existing and upcoming platforms – we are just scratching the surface of what can be done, even with mobile phones. Hello wearables. It would be unwise to ignore the benefits that these platforms provide, not to mention the aspect of studying how people use and appropriate existing devices and infrastructure.
No matter how you look at it, whether you break new ground with new hardware implementations, or gather data from commodity hardware provided to hundreds of people, one thing is for sure: this is a great time to be a ubicomp researcher.
Do you still wear a wristwatch on a regular basis? Many people, myself included, don’t wear a watch nearly as much as they used to. Blame (or thank) the mobile phone. Over the last few years, however, a growing number of body-worn devices like the FitBit and Nike’s FuelBand have started to counterbalance the equation in favor of “wearable devices” again. Today, there’s even excitement about the idea of wearing a wristwatch as a companion to the mobile phone, which presents a very interesting HCI design space. Another piece of evidence of this trend is Project Glass, envisioned by Google, which offers the promise of bringing the 20-year-old MIT wearable concept to the masses. Remember that going back in the history of wearable computing, the pioneers at the MIT Media Lab were forced to carry large, heavy backpacks and wear awkward face-mounted displays. No longer.
But why talk about wearables? First of all, a core element of my research agenda is to study different mechanisms for capturing human activity, and FitBits and FuelBands fall in that category. More interestingly, there is an emerging class of outward-facing wearable cameras devices that give researchers direct observation into a person’s everyday activities. The Looxcie and the AXON Flex are two examples of such cameras, the former for the consumer market and the latter for specific applications (i.e. police officers). The SenseCam, a tool for image capture for Microsoft Research’s MyLifeBits project, is now sold as the Vicon Revue.
Researchers are beginning to explore the potential of these wearable cameras. Kelly et al. are using them to investigate active versus sedentary behavior (PDF), while Mingui Sun is interested in the assessment of diet of physical activity with a custom device he and his team built, called eButton.
Lately, as part of my research in indirect health inference through infrastructure-mediated sensing techniques, I’ve been investigating options for remote data acquisition. It would be wonderful if I could find a platform that let me take an analog signal as input, send it through a pipeline of signal processing and machine learning algorithms, and submit results to a remote server.
At the high-end, there are netbooks. Any BestBuy can sell you a complete netbook system for less than $300. For certain applications in data sensing, processing and communication, $300 is good enough. Unfortunately, a netbook is a bit too big and power hungry. Not to mention that its screen, graphics card and other features might go unused, inflating the cost of the device unnecessarily considering the job it’s been designated to do.
At the other end of the spectrum are platforms like the Arduino, which is small, inexpensive, but might not provide the processing power one might need. Are there any other alternatives? There are plenty of single-board computers out there, one of which is the Chumby Hacker Board, or CHB for short.
Recently, I’ve been following the development of an ARM-based platform called Raspberry Pi. The goal is to develop the cheapest possible computer with a basic level of functionality, for around $25. The team is already showing a prototype board, the size of a credit card, running Ubuntu:
Lots of details can be found here and they are expected to be shipping in December. I am really looking forward to experimenting with them.