Hardware hacking for data scientists
The Data Sensing Lab is a project sponsored by O'Reilly Media with the tagline "Hardware hacking for data scientists". By deploying custom wireless hardware at tech conferences like Strata and Google I/O the team is looking to advance what real-time sensor network data collection, analysis, and visualizations will look like in the near future.
"We will soon begin to move in a sea of data, our movements monitored and our environments measured and adjusted to our preferences, without need for direct intervention. What will this look like? How can we create and shape it? How can we introduce the relevant hardware to people who already possess data analytics skills?"
Developed by Alasdair Allan, Kipp Bradford, Robert Faludi, Kim Rees and Julie Steele the team uses Arduinos, XBee radios, 3d printed enclosures and cloud database tools to track temperature, humidity, noise, light, air quality, audience engagement, and foot traffic levels in real-time for each of the conference venues.
The sensor deployments gives conference attendees a better glimpse into their surrounding environment and the team wanted to remind them 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."
More details about the hardware and project details can be found at: Datasensinglab.com or by watching the team and sensor motes in action at the 2013 Strata conference in the video below.
Via: Alasdair Allan