MIT Research Project Aims to Corral, Enrich Live Streaming Video

"Deepstream" seeks to iron the creases on one of the Internet's fastest growing trends.

by / January 22, 2016

The growing popularity of live streaming video is a momentous milestone in the evolution of the Web, but the organizational effort remains scattered and immature. A research project at MIT's Center for Civic Media at the Media Lab, however, strives to enrich the medium by providing a vehicle for contextual information. Called Deepstream, the project also serves as a central platform that makes it easier to find live streams around particular events and topics.

Live streaming video is amazing, but it suffers from a couple major problems, said Gordon Mangum, project lead and graduate student at MIT's Comparative Media Studies department, who added that Deepstream would solve both of those.

"One thing it's not that good at is providing backstory or context for what you're watching," Mangum said, explaining that through their research, his team found that people were often confused or lost when tuning into livestreams, because unlike a TV news channel or text-based Web content, there's no additional information. "People try to hack their way around this problem by trying to second-screen live streams. They'll start watching a live stream and they'll start opening up browser tabs or going on their mobile phone to look up a Twitter feed related to what they're watching or news stories, which seems kind of silly because you have to stop watching the thing you were watching in the first place."

Deepstream lets broadcasters group multiple live streams together and overlay data, solving the context problem, so it's easy for users to find related content.

Some "larger city governments" expressed interest in Deepstream, Mangum said, because they see potential in this technology to enrich their open government initiatives and support new campaigns.

"I think there's great potential here," he said, "because you can have people essentially tuning into that live stream from home or wherever they are and not just jumping into it blind and having to listen to what's going on, but having this extra layer of information that says, 'Here's the agenda for tonight, we're going to be talking about parking, here's a link to the map with the current parking plan for the city.'"

Other city leaders told Mangum's team they wanted to use the platform as a way to support their anti-smoking campaign. By allowing citizens to forward their live streams to the city's anti-smoking Deepstream, citizens could watch, participate and contribute to a collection of video streams around a common topic.

"We've got traffic cams that watch intersections," Mangum said. "Cities could potentially aggregate those, put those all together into a particular Deepstream and then add a layer of information like what's the purpose of these traffic cams? Is it to cut down on accidents? You could add information about accident rates and the most dangerous intersections, or reminders to people to wear their seatbelts, or whatever sort of public service campaign the city is engaged in."

Another problem with live streaming is that finding content is difficult. There are many different streaming platforms — YouTube, Meerkat, Ustream and Periscope, to name a few — and as a result, streamed video content is insulated. There's no central location to reliably search for live streams around a particular topic or event. Live streaming is mostly an ad-hoc affair, and streams are easy to miss.

"The text-based Internet, we know how to search that, and Google is great at searching that," Mangum said, "but it's much more difficult to search livestreams, the reason obviously being that they're so ephemeral."

That's why Deepstream's second major function is to serve as a search aggregator for other live streaming services. And in some cases, Mangum said, the search platform can find relevant content better than the original service's search function can.

Funding prospects for the project beyond the current research phase are uncertain, and Mangum said when he graduates from his master's program in June, he intends to explore whether Deepstream can work as a business.

"I'm incredibly passionate about this platform and about the potential for live streams to change the way we engage, especially with breaking news and civic events, protests, rallies, gatherings, town hall meetings," Mangum said. "I think there's an incredible potential there."

Editor's Note: this article was edited on Jan. 23 to fix the spelling of the researcher's name.

Colin Wood former staff writer

Colin wrote for Government Technology from 2010 through most of 2016.