Zoom In

Zoom In

Organisation: National Geographic (United States)

Publication Date: 06/22/2017

Description

To display feeds of single-topic or single-story coverage, many news outlets rely on tag-driven, chronological “hub pages.” These feeds, however, fail to balance breadth, depth, and context. As a result, readers find them minimally useful, and publishers rarely consider them works of journalism unto themselves. How can we make this mainstay of the news website better—by making it into a dynamic explainer? We propose that the user “zoom in” or “zoom out” of a story feed, via a slider interface that shows more or less content based on the user’s preferences. Importantly, the feed is organized via a context-first taxonomy that’s baked into a publisher’s CMS. Our target audience is the subset of readers who wish to quickly understand the gist of a story but also explore how the latest news fits into the bigger picture. When a user is fully “zoomed out,” the user sees a single paragraph of text, each sentence of which represents an important component of the story or topic. Moving the slider transforms these sentences into headers, under which are two to three sentences of extra detail. Zooming in further reveals one-sentence summaries of the latest stories related to each header. Finally, the reader reaches a list of all latest stories with key sentences and quotes. We envision that a new story would be tagged as belonging to a topic (e.g. climate change) and subtopic (e.g. rising temperatures). In the near future, text-summarization algorithms will reliably identify summaries and key quotes. (We experimented with the text summarization algorithm Aylien, but in our submitted prototype, we ultimately chose to select passages manually. We likely will experiment other technologies, such as Google's TensorFlow, in the future.) Creating and maintaining site-wide taxonomies for Zoom In could also become automated. Higher-order statistical analyses could cluster stories within a given topic into subtopics, which algorithms could then summarize with descriptive sentences. To ensure that summaries are up-to-date, these algorithms would give more weight to more recent stories. Since Zoom In would improve the usefulness of an article feed, the feed’s search standing, social reach, and overall engagement would improve—all of which contribute to display-ad revenue. Moreover, placing a story within an improved context could invite increased click-through rates, additional engagement, more ad revenue, and possibly lower bounce rates.

Technologies used for this project:

HTML, CSS, Javascript, taxonomy in Adobe Experience Manager (AEM).

Video

Follow this project
Wait
Comment

Comments (0)

You have to be connected to contribute

You have to be connected to follow

Leave this project and no longer be informed about this project

By joining this project, you will be informed by email when an update or a new contribution is posted on the website.

Thank you for your active participation !
Best,

The GEN Community Team