Ever wondered how journalists make sense of the deluge of information posted online in a news event?
We used to talk about the information superhighway, that metaphor became outdated as the volume of information grew. The volume of information uploaded now is overwhelming, as quoted in the film it’s more than an hour of video on YouTube, and 58 photos on Instagram – per second. (This was recorded in November 2012, the Instagram figure may have declined).
There’s also a huge growth in what people can do with images and film, so how does a journalist find the best image for a story AND confirm that the image is real. Turns out the best way to validate the image is to find the source, and validate whether the source is trustworthy.
The Nolan explores three degrees of difficulty in assessing this;
- the source has an online persona and a reputation online that you can trust, the example given was around photos from Superstorm Sandy. When the source was confirmed as to known Manhattan food bloggers the image was accepted.
- look at whether other people online find this person credible, this can work across language barriers to identify a source that might be worth interviewing. In the case of the Egyptian revolution this type of analysis was done on twitter data, looking for ‘nodes’ who were retweeted. Then looking at those nodes as possible credible sources.
- if the source cannot be verified can other details of the image or video be confirmed to corroborate the story. He takes apart the details of a rather gruesome video to assess its validity. And cross-references what he finds to assess the credibility of 3 sources.
Most of the tools used are free, online, and available to everyone. Which creeped me out about three seconds before he said “Given a couple of clues, I could probably find out a lot of things about most of you in the audience that you might not like me finding out.
I like the cyber-detective aspect of his work, not sure I want to see some of the video/image content that comes across his desk.
Change Meme 