Bill van Melle, Thea Turner, and Eleanor Rieffel contributed to this post
FXPAL’s work on the MyUnity Awareness Platform has received considerable attention from the popular press and the Internet blogosphere in recent weeks, following a nice write-up in MIT’s Technology Review. That article, despite its misleading headline, correctly relays the core motivation for the work: to improve communication among workers in an increasingly fragmented workplace. However, some writers who picked up on that article focused instead on the sensational aspects of having technology monitor people’s behaviors and activities while they are working. They incorrectly described some of the platform’s technical details, overstated what the platform does and what it is able to do with the data it collects, and failed to mention the numerous options we offer users to control their privacy. We thought we should clear up some of these misconceptions and clarify the technical details.
Some articles and blogs incorrectly suggested that cameras and computers tracked people’s activities. In fact, the cameras in the MyUnity system do not see in detail what people are doing, and they do not provide a video feed of a person’s workspace to others. The camera video feed is privately processed by a set of computer vision algorithms, which classify the type of motion in the video frame and then throw the video away. The output of these algorithms is simply 0, 1, or 2+, an estimate of the number of people in the space. Likewise, the computer activity sensors simply monitor whether a worker is actively using his or her computer, not the specific applications or activities the worker is engaged in. This is the same information as is used to put a computer or screen to sleep when it isn’t in use.
It is also important to point out that MyUnity was designed from the ground up to allow users to choose a level of participation that satisfies their privacy concerns. In fact, this ability is central to the novelty claims we have made in the patent filings related to this technology. If a user is not comfortable with a camera in his or her office, no camera is installed; the algorithm that determines a user’s overall presence state simply does not use that input. The system can similarly operate in the absence of WiFi or Bluetooth location data for any user who chooses not to share it. Users can elect to provide a great deal of information about themselves or very little. In our studies of the platform’s use over more than a year, we have found that the majority of users choose to have multiple channels of information collected about them, because they value the benefits they receive by providing their colleagues with a greater sense of their presence, but the system works for those who chose more restrictive privacy settings as well. It is difficult for others to tell what privacy settings any particular user has selected.
All the misinformation aside, the articles do stress an important consideration that research in this area must address: balancing legitimate, beneficial use of presence information against inappropriate, harmful use. In this area there is still much work that needs to be done, and this is a central focus of our research going forward. Deployed systems such as MyUnity offer a great opportunity to learn through observation and iteration how to best design controls for sharing presence information. We should also point out that we are not alone in this direction of research. Our peers at Microsoft Research, Carnegie-Mellon University, the University of Pittsburgh, and elsewhere are also working diligently towards a better understanding of how to maintain this balance between utility and privacy.
It’s great, however, that this topic has received attention from the media. These are important issues that need to be debated and understood broadly. Besides, there is no such thing as bad press.