Intel: “Stop Storing Everything”
The Internet of Things (IoT) isn’t just coming to a factory near you; soon, tech leaders like GE and Intel hope to see connected devices used across industries and even in the home. But to connect so many items on such a large scale, something about today’s paradigm has got to give, or so Intel claims.
In The Data Society Manifesto, published on Thursday, Intel mapped out nine specific areas that need attention from industry and government alike to make the Internet of Things possible. High level topics include bettering educational standards and addressing privacy and security issues, but the company also says that part of reaching this goal requires us to “stop storing everything.”
At the manifesto’s launch, Kumar Balasubramanian, Intel’s Intelligent Solutions general manager laid out the company’s new IoT division while assuring audiences that the Internet of Things is already possible—at least provided we make these tweaks.
“There are a lot of reasons to be excited about the IoT,” he said. “It’s truly not hype. It’s not about IoT as a promise for IoT a decade or five years from now; this is IoT for here and now.”
Balasubramanian gives credit to dropping costs of big data analytics platforms and the growth of mobile for making IoT possible today, but he also warned that in order to make use of so much sensor data industries must change the way they approach the area.
The first prong of this approach, Intel says, lies in training more skilled data scientists who will in turn help stakeholders make more sense of growing data volumes.
In addition, adding transparency data collected from consumers along with clear and trustworthy privacy policies ranked highly on the list of priorities, which is a chief concern in healthcare and retail, but will likely spread to other sectors as more devices become connected. Part of this effort also means keeping data anonymous where possible, Intel adds.
But on the other side, even as analytics tools become more sophisticated and IT infrastructure is less expensive to invest in, the company warns that not filtering out sensor-gathered noise will undermine users efforts to derive meaning from data.