What is the Data Sensing Lab?
At the O'Reilly Strata Conference in New York in October 2012, we gave attendees a taste of the super-connected world that's ahead of all of us by instrumenting the conference environment with basic sensors and wireless mesh networking. We built more than 40 sensor motes using Arduino Leonardo boards, XBee radios, and a handful of off-the-shelf parts, including a PIR motion detector, a temperature and humidity sensor, and an electret microphone amplifier. These motes were distributed around the conference venue, and reported back during the conference. The data was made publicly available online.
At the O'Reilly Strata Conference in Santa Clara in February 2013, we added more sensors (which included rebuilding some that had been damaged in transit, for a total of around 50 sensor motes), real-time visualization, and a new interactive feature for attendees: the “Awesome Button,” a giant red button outside of each session room, which attendees were encouraged to push as they exited if they thought the talk they had just seen was awesome.
The six of us who run the Lab occupied a booth in the Expo Hall during the conference, making ourselves available to interact with attendees and answer questions about what those blinky lights all over the conference venue actually were. Makers Alasdair Allan and Kipp Bradford manned a couple of soldering stations for assembling motes, as well as a MakerBot Replicator that was printing cases for our sensors; Rob Faludi of Digi sat next to the gateway that was routing our data to the iDigi backend, and presided over the data stream; Kim Rees and Andrew Winterman of Periscopic had a large screen showing the real-time data visualizations they were building; and editor Julie Steele talked with attendees about Allan and Bradford’s new book, Distributed Network Data (O’Reilly), which shows readers how to recreate everything we’ve done in the Data Sensing Lab.
Why Run the Data Sensing Lab?
Many people stopped by the booth at Strata wanting to know why we were doing this. That was a wake-up call: to us, the answers were obvious, but not so to everyone else. We realized that we might need to tell this story more broadly. So here it is.
We’re bringing whimsy backO’Reilly Media's “special sauce” is its capacity for dreaming. This is part of what allows us to see so far into the future, over and over again (this future sight has been both our superpower and our kryptonite; a consistent challenge of ours has been products and conferences released too soon). A big portion of that whimsy has lived in our Make division—now a separate company. There were good reasons for spinning off Maker Media, but shedding our capacity to dream was not one of them. The Data Sensing Lab is a great example of us doing what O’Reilly does best, in a way that no one else does it.
We’re taking data scientists to their roots
So many data scientists begin their work with data that has already been collected. It’s in a spreadsheet, or a CSV file, or in the cloud somewhere: it’s already virtual. But the truth is that all data has to be collected. It comes from somewhere messy, somewhere real. Even when the data consists of social media streams, it is often describing—and meant to answer questions about—the physical world.
We wanted to remind attendees at the Strata conference what actual data collection looks like: what it means to build your own sensors from scratch, deploy them into an environment where they might be kicked or stolen or partially blinded by a wall, fine-tune the reporting rates to work within the available bandwidth, and collate data from multiple sources reporting at differing rates. These things are the roots of data science, and we wanted to go back to them.
We’re providing public data sets
There is still a lack of public data sets, particularly data sets that haven’t already been sanitized in some way, according to someone else’s rulebook. A strong motivation for the Data Sensing Lab is to provide a data set that outsiders can interact with: clean and munge as they wish, and experience the inherent difficulties of data collection and cleaning. That's an important step in understanding what's involved in making data public, and what can be gained from it. Already we know this: that it's much harder to get others to work on your data than you might imagine.
We’re looking to the future
Do you remember when the first mobile phones came out? Until the late 1980s, they were the size of several laptops stacked together, and had shoulder straps for carrying. They seemed like ridiculous toys for Saudi oil barons and other elites. A quarter-century and three mobile generations later, they seem even more ridiculous. But looking back, they were the obvious first steps of the technology that would change all of our lives.
Those awkward phones with their behemoth batteries weren’t the thing—smartphones are the thing. But the former was necessary to achieve the latter. In the same way, the kind of DIY sensor motes we’re building in the Data Sensing Lab aren’t the thing; they’re only a precursor to the smart dust that will change all our lives in another decade or two. Sure, the motes we’re deploying in the Lab are awkward, fragile, and a little clunky. That’s okay. We’re looking forward and paving the way to something amazing.
Who runs the Data Sensing Lab?
The O'Reilly Data Sensing Lab has been a labor of love by an extended group of O'Reilly and Make employees, as well as several key people from outside of those companies. The active members of the Data Sensing Lab are as follows.
Alasdair Allan is the author of Learning iOS Programming, Programming iOS Sensors, Basic Sensors in iOS, Geolocation in iOS, iOS Sensor Apps and Arduino, and Augmented Reality in iOS. Last year, he and Pete Warden caused a privacy scandal by uncovering that your iPhone was recording your location, all the time. This caused several class action lawsuits and a U.S. Senate hearing. He isn’t sure what to think about that. From time to time, he stands in front of cameras, and you can often find him at conferences run by O’Reilly Media.
Alasdair runs a small technology consulting business writing bespoke software, building open hardware and providing training, including a series of workshops on sensors. He sporadically writes blog posts about things that interest him, or more frequently provides commentary about them in 140 characters or less. Alasdair is a former senior research fellow at the University of Exeter. As part of his work there, he built a distributed peer-to-peer network of telescopes which, acting autonomously, reactively scheduled observations of time-critical events. Notable successes included contributing to the detection of the most distant object yet discovered, a gamma-ray burster at a redshift of 8.2.
Kipp Bradford is an entrepreneur, technology consultant, and educator with a passion for making things. He is the founder or cofounder of start-ups in the fields of transportation, consumer products, HVAC, and medical devices, and holds numerous patents for his inventions. Some of his more interesting projects have turned into kippkitts.
Kipp is the author of Distributed Network Data (hardware hacking for Data Scientists, with Alasdair Allan) and is one of the cofounders of the Data Sensing Lab. Kipp also co-founded Revolution By Design, a non-profit education and research organization dedicated to empowerment through technology, and co-organizesRhode Island’s mini Maker Faire. He is one of the USA Science and Engineering Festival’s “Nifty Fifty”. Kipp was the Demo Chair of the 2013 Open Hardware Summit and has been recognized as a leading innovator at Frost & Sullivan’s GIL 2013. As the former Senior Design Engineer and Lecturer at the Brown University School of Engineering, Kipp taught several engineering design and entrepreneurship courses. He serves on the boards of The Rhode Island Museum of Science and Art, The Providence Athenaeum, and the community arts organization AS220. He is also on the technical advisory board of MAKE Magazine, is aFellow at the College of Design, Engineering and Commerce at Philadelphia University, and is an Adjunct Critic at the Rhode Island School of Design.
Robert Faludi is the Collaborative Strategy Leader in R&D for Digi International, with a mandate to forge stronger connections with the community of innovators, discover outstanding new work, contribute to outside projects, and support the people making that work. Faludi is also a professor in the MFA program at the School of Visual Arts in Manhattan and in the Interactive Telecommunications program at NYU. He specializes in behavioral interactions through physical computing and networked objects. Rob is the author of Building Wireless Sensor Networks, with ZigBee, XBee, Arduino and Processing (O’Reilly 2011). His work has appeared in The New York Times, Wired Magazine, Good Morning America, BBC World, the Chicago Museum of Science & Industry, and MoMA, among others. He is a co-creator of LilyPad XBee wearable radios, and Botanicalls, a system that allows thirsty plants to place phone calls for human help.
Kim Rees is a founding partner of Periscopic, an award-winning information visualization firm. Their work has been featured in the MoMA as well as several online and print publications, including CommArts’ Interactive Annual, The Information Design Sourcebook, Adobe Success Stories, CommArts Insights, Infosthetics.com, FlowingData.com, and numerous websites, blogs, and regional media outlets. Periscopic’s body of work was recently nominated for the Cooper-Hewitt National Design Awards.
Kim is a prominent individual in the information visualization community. She has published papers in Parsons Journal of Information Mapping, was an award winner in the VAST 2010 Challenge, and is a guest blogger for Infosthetics.com. Kim has been featured on CommArts Insights and has presented at several industry events including Strata, the Tableau Software Conference, AIGA SHIFT, WebVisions, CERF Biennial Conference, and Portland Data Visualization, among others. Recently she has also been an advisor on an upcoming documentary film and is the Technical Editor for Visualize This by Nathan Yau. Kim received her BA in Computer Science from New York University.
Veton Saliu is a graduate from Brown University with a degree in Biomedical Engineering. His interests lie in computer programming, robotics, assistive technologies, and engineering design. Veton is an engineer at kippkitts LLC, where he is responsible for software/hardware development and testing. Furthermore, he is a research assistant at the Brown Institute for Brain Science, developing a neurorehabilitative video game platform for children with cerebral palsy. His work will be presented in a research paper he coauthored for the Journal of Accessibility and Design for All.
Julie Steele is the Content Editor for Strata at O’Reilly Media. She is co-author with Noah Iliinsky of Beautiful Visualization (O'Reilly 2010) and Designing Data Visualizations (O'Reilly 2011). She finds beauty in exploring complex systems, and thinks in metaphors. She is particularly drawn to the visual medium as a way to understand and transmit information. Julie holds a Master’s degree in Political Science (International Relations) from Rutgers University in Newark. She lives in New York City, where she cooks, reads, designs, and practices yoga. You can find her blogging occasionally for O’Reilly Radar, the O'Reilly Strata blog, or on Twitter.
Who's Your Friend?
Pronounced "sensei," like the Japanese word for a teacher or other elder who is wise in a particular profession; it literally means something like "one who was born before the rest." This name was submitted by two readers of our Google+ page, and we love it. (Hat tip to Sensey's creator, Mark Paglietti of O'Reilly Media, for the wonderful logo design. )
While the electronic sensors we use in the Data Sensing Lab are relatively new, the things we are measuring (humidity, temperature, light, noise levels) are elemental; we are hoping to learn and grow our modern world by understanding the things that have always been there. We feel that this name for our little sensor buddy, therefore, is highly appropriate, and a good reminder that gaining wisdom and insight is the purpose of the Lab.