One of the goals of the Share Wars project is to create a replicable formula for story sharing. The idea is we isolate the factors that make a story sharable and then use them to make content we know will go viral.
Sharing is the ultimate valuation of content so it’s not as cynical as it might seem. We’re simply aiming to create stories that are more valuable to the audience.
It’s an idea that’s starting to take hold in other corners of the digital media world and academia. During the past couple of days, we’ve come across two interesting stories dealing with the virality of content.
The first is about a new addition to the team at Gawker, whose dedication to the audience and anti-elitist, modern take on journalism keep them in front of the digital media pack.
Gawker’s latest secret weapon is one Neetzan Zimmerman, who editor AJ Daulerio hired two months ago to focus exclusively on viral content. The results have been impressive and Zimmerman, a 30-year-old “freak” from Israel, works hard for his page views, writing 13-14 stories per day.
Zimmerman describes his approach as “take everything that’s going on the Internet seriously. Treat it as you would something that you might read in The Economist”.
We agree … it’s about time someone took this Internet thing seriously.
In other sharing news, a team from UCLA and Hewlett-Packard’s HP Labs have analysed the news headlines that share on Twitter. The most prosaic outcome from the trial was the most-shared tweet: “Apple Buddies Up With Cheaper Wireless Partners for iPhone”.
More interesting was the use of two tools: Stanford’s Named Entity Recogniser and the Feedzilla API, which allowed the research team to gather a dataset of more than 40,000 news articles, collected during a nine-day span in August 2011.
As Share Wars’ three-month data capture across 100-odd news sites globally comes to an end, we find ourselves with a whopping 2 million articles to process. We need some automated tools to help us sift through the data. Most valuable for us would be a semantic analysis tool that generates key words or themes from a text article. Anyone out there got one of those handy?