Is Twitter Becoming a Research Funder?

Twitter has been increasingly share-y with its enormous data set. Given the recent $10 million grant to MIT to analyze and put social media data to use, how grand are Twitter's research aspirations? 

Twitter reports about 500 million tweets are sent every day, from 271 million users (the population of the United States is 317 million) in more than 35 different languages. That’s shared information on everything from the latest episode of Scandal to protests in Hong Kong. Even though it’s become this entirely new universe of public discussion, the company is still feeling its way toward what exactly it is, figuring out its business model and what greater role it might play in media and society. 

Related: IP Profiles of Jack Dorsey, Evan Williams, and Chris 'Biz' Stone

Twitter is generally more open than its main competitor Facebook, but this year it’s been inviting outside, academic researchers to grapple with some of these questions about what’s to be done with hundreds of billions of little online interactions. In the spring, the company awarded inaugural "Data Grants," handing over its archive of every tweet to select recipients. And just recently, the company awarded $10 million to the MIT Media Lab to analyze patterns in how people share opinions and organize online. A spokesperson has said that the company plans to invest more into academic research.

All this suggests that Twitter is following in the footsteps of Google, which has a large community of researchers inside the company, but also courts outside academics studying fields related to the search engine's interests through grants and fellowships. 

Related: Google’s Quiet, Multimillion Dollar Academic Research Program

With such a huge, constantly expanding body of information, the potential for researchers to play around in the entire Twitter sandbox is tremendous. The company could become a big-picture data research program well beyond its origins as a way to keep up with friends. 

Consider that this month, Twitter announced that it will establish the Laboratory for Social Machines at MIT. The initiative seeks to find new ways of analyzing information posted to online media platforms, beginning with, but not limited to, Twitter’s entire catalog of tweets, which the company is handing over. The project will also look at other public venues such as Reddit and Wikipedia. The center seeks to build new technologies to channel such information for good, teaming up with journalism outlets and social and political reform movements. In part, the lab wants to find ways for the firehose of opinion and information to be used for constructive purposes (beyond its frequent use for protest or disruption).

And in the Twitter Data Grants pilot project, the company awarded its dataset for free (normally not cheap) to be used by six researchers, chosen out of 1,300 applicants. Harvard won with a plan to study the spread of food-borne illness. The University of California at San Diego won for its proposal to measure the happiness of cities based on tweets. The University of Twente in the Netherlands will study the dataset to explore effectiveness of cancer early detection campaigns. And about 1,294 other researchers presumably had some pretty good ideas about how to use all those tweets.

All of this does suggest that Twitter is genuinely interested in how 140 characters can be a force for good, and not just profit. Still, in the same way that Google’s grant program telegraphs its business aspirations, the MIT program is a touch cozy with its plans for building revenue. 

The head of the new MIT lab is Deb Roy, the founder of Bluefin Labs, a startup that analyzes the connections between Twitter activity and what’s airing on television. Twitter bought the company in 2013 for $90 million reportedly, and gave Deb Roy a part-time gig as the company’s chief media scientist. The company states that the research at MIT will be independent of Roy's work advising Twitter, and has established Conflict Management Guidelines to handle the balance. 

Twitter also makes quite a bit of revenue by selling its data for commercial purposes, about $70 million in 2013 through licensing of the full stream of tweets. So the line between Twitter the for-profit company and Twitter the research funder is a tricky one to navigate. You can easily imagine how such mastery of online public opinion could cause worry about less-than-noble intentions. (You saw Captain America: Winter Soldier, right?) 

But suppressing our paranoia, and seeing how instrumental Twitter has been in pivotal moments like the Arab Spring and the Ferguson protests, you can also imagine several exciting academic projects if Twitter continues to head down this road and let more outside researchers in. 

Related - Google: Grants for Science Research