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
Feb. 2, 2012, 2 p.m.

topheadlin.es: Aggregating both news and news judgment

The experimental mobile app from The Wall Street Journal skims homepages for the top news from a universe of sources.

There’s a morning routine that seems to unite almost everyone in the media world: that first, groggy glance at the smartphone before leaving bed. It’s a universal experience born out of a basic need: Tell me what I need to know right now.

But we get that news in different ways. Some turn to Twitter; others turn to email. Some jump right to their favored news site; others dive into RSS. It’s a matter of choice, but one that usually hinges on three important properties: timeliness, relevance, and readability.

When Jeremy Singer-Vine was building topheadlin.es, a mobile-ish news aggregator, this was the specific use case he had in mind. “For me, I was trying to think: What part of the day am I most wanting a different experience?” he said. “For me, it’s waking up in the morning and trying to catch up with whatever was happening.”

Singer-Vine is a programmer/reporter for The Wall Street Journal, and topheadlin.es is a sort of stealth experiment in mobile news and aggregation for the company. What it does sounds fairly simple: Every three minutes, it scans a list of 10 news sources and delivers the one top headline from each, in an unclothed layout best fit for a mobile web browser. The Journal is no different from other news outlets that offer a handful of device-specific apps for news, but in this case, Singer-Vine said, they wanted something that was specifically mobile and worked across devices. “We just wanted to experiment with the idea of doing something that wasn’t a mobile site but was mobile minded,” he said.

Sounds simple, but there’s an underlying issue under the hood. How exactly do you figure out, in code, the top headline in The New York Times, The Washington Post, BBC News, Politico, The Wall Street Journal, and more? (That list also includes the Drudge Report, Vulture, ESPN, the Associated Press, ProPublica, and a smattering of regional papers, sports, and financial sites) In other words, how do you make an aggregator with a little more finesse?

“It aggregates news judgment,” Singer-Vine told me. “We’re looking at what are, in this case, homepage editors of big news sources think are the most important stories of the day or of the moment.” In the same way that services like News.me, Zite, or our own Fuego aggregate social news judgments (like Twitter patterns) or personal news judgments (like user behavior), topheadlin.es aggregates editorial news judgments.

Picking out a news site’s top story is an easy task for any human eye; most news homepages have a geospatial familiarity to them inspired by newspapers and bred further on the web. There is order, or at least a method, to the sameness that puts the top story in a given slot. That may speak to larger questions about design of news sites, but in this case was quite useful. What Singer-Vine did was create a way of IDing how those top slots are defined within a site’s underlying HTML and CSS and then create a rule for the app to follow when it looks for new headlines. So, for instance, topheadlin.es knows to look in the top-left corner of WSJ.com for the biggest story of the moment.

“Every site is designed differently, but within those designs they follow very consistent patterns,” he said. “We take advantage of those patterns to ID what is the top story.”

But another interesting thing happened in creating topheadlin.es: Singer-Vine inadvertently made a tool for gauging the diversity of news coverage. In a way, it’s flaw-as-feature. When topheadlin.es gives you 10 stories and five of them are about Mitt Romney winning the Florida Republican primary, that repetition isn’t exactly the most useful snapshot of news at that moment. But it is a useful sign of the intensity with which the collective news universe views one story as dominating that particular rotation of the daily cycle.

Singer-Vine said he’s still tweaking the app, and that “clumping” effect of getting the same story from different sites is one thing he’s looking to address. He also wants to see what a topic-based version of topheadlin.es would look like, with sports, money, and technology spun off into different bundles. For the time being, the project will likely remain a quiet side project; there was no official launch or announcement, and the project has spread largely through word of mouth from Journal employees, Singer-Vine said. Which more than likely means it’s picked up many of its eyeballs from heavy news consumers, the type whose media diet begins every morning similar to Singer-Vine’s. One goal for topheadlin.es to reach a broader audience is finding a way to appeal to readers who may not have as much zeal for keeping on top of the day’s news. Then again, there’s a growing market for compartmentalized, smartphone-friendly aggregators like Flipboard, Pulse, Zite and more. Singer-Vine thinks topheadlin.es can easily fit into that world.

“Even if you’re not news obsessed, it’s another way of watching the news, of bringing together disparate subjects,” Singer-Vine said.

POSTED     Feb. 2, 2012, 2 p.m.
Show tags
 
Join the 60,000 who get the freshest future-of-journalism news in our daily email.
From shrimp Jesus to fake self-portraits, AI-generated images have become the latest form of social media spam
Within days of visiting the pages — and without commenting on, liking, or following any of the material — Facebook’s algorithm recommended reams of other AI-generated content.
What journalists and independent creators can learn from each other
“The question is not about the topics but how you approach the topics.”
Deepfake detection improves when using algorithms that are more aware of demographic diversity
“Our research addresses deepfake detection algorithms’ fairness, rather than just attempting to balance the data. It offers a new approach to algorithm design that considers demographic fairness as a core aspect.”