Nieman Foundation at Harvard
HOME
          
LATEST STORY
From shrimp Jesus to fake self-portraits, AI-generated images have become the latest form of social media spam
ABOUT                    SUBSCRIBE

The future of news is anticipation

“Our interests are temporal, as is the news cycle. But those two don’t always align perfectly.”

One of the most important trends going into 2014 is the wave of sophisticated algorithms and processes that will forever change how journalism is both created and consumed. They are inherently social, but not in the way you may think. And they rely on the vast repositories of data we generate each time we connect, whether that’s searching Google for a restaurant, wishing friends happy birthday on Facebook, or posting an in-line annotation on Medium.

amy-webbThis past year, we saw the first anticipatory computing opportunities in Google Now, which originally launched as part of the Android operating system, and an app called MindMeld, created by former MIT researcher and current Expect Labs CEO Tim Tuttle. In short, apps like Google Now and MindMeld observe the last few minutes of your thought process in order to predict the next 10 seconds.

In the case of Google Now, if you’ve been searching Google for information about the new movie Inside Llewyn Davis and then ask it “Where is it playing?” Google Now assumes you want to know the nearest movie theater, times, and perhaps even directions on how to get there from where you’re standing. MindMeld offers something even more exciting. When users are connected, it listens in and begins to populate its dashboard with contextual information to help you have a more informed conversation. So if you’re talking with a friend about Llewyn Davis, MindMeld will automatically show history about the 1960s folk scene, the cast, directors, details about the soundtrack, reviews and more.

In the hands of journalists, these and the other emergent anticipatory computing applications can be harnessed as powerful reporters’ assistants. Google Now can query calendars, traffic, weather, and news to deliver just the right information at the right time. Example: It might see that there’s heavy traffic along your route and alert you to leave seven minutes earlier than you normally would. MindMeld has the power to deliver contextual content to reporters as they’re conducting an interview. Suddenly that usual last question “Is there anything else I should ask you?” seems irrelevant while using MindMeld, since it would have already unearthed ambient information you might have missed during the conversation.

If apps can anticipate our next thoughts, then algorithms should allow for two interesting possibilities in 2014: predicting breaking news and delivering highly personalized content to each consumer.

Already, social feeds can be mined and crunched to show crises as they start to erupt. This is especially obvious during sporting events, when fans tend to post sentiment, then photos or videos of fights and eventually the details of large brawls after games. Some companies, such as IBM, have been harnessing big data along with artificial intelligence and machine learning to infer changes in the real world. This isn’t about predicting the news a few days from now — it’s about seeing a breaking news event just as it’s about to unfold.

News organizations have long hoped for meaningful personalization — the more directly a news product can be tailored for each individual, the more likely she’ll be to stay with that brand and to use it over and over. Personalization experiments have failed at numerous big media brands, but it’s because the approach has always been too rudimentary. Users would tick a few interests, like “sports” and “world news,” and depending on how well the backend system was programmed, stories that kind-of sort-of fit into those categories would be delivered. Selecting a few broad categories never allows for nuances in geography, taste, or even our changing interests as we have fresh ideas and encounter new people.

Our interests are temporal, as is the news cycle. But those two don’t always align perfectly. If you think about it, the exciting promise of Google Now and MindMeld is in the ability to anticipate what content might interest us next. It eviscerates the need for related content, since ostensibly all of the content we’d be delivered is relatable only to each one of us individually.

As a result, news organizations have thrilling opportunities in the months ahead to supercharge the reporting process and to personalize content in ways we have never seen before. The future of news is anticipatory.

Amy Webb is the founder of Webbmedia Group, a digital strategy agency, and cofounder of Spark Camp.

                         
Updating regularly through Friday, December 20