Metadata has had a bad rap lately, with disclosures tying its collection to government spying programs. But those bits of information lurking behind our phone calls, photos and online chats can be useful in other ways if they’re harnessed properly.
Metadata that reveals when and where photos and videos were taken can help establish trust in eyewitness footage documenting events as they unfold. Without it, establishing what’s real and what’s not can be hard, if not impossible.
How do news organizations determine which bystander’s footage is legitimate before they show it on the nightly news? How do attorneys find the most reliable footage of an assault to build a case?
“There’s something powerful in people’s capacity to use the devices in their pockets to document what’s around them, but there’s a frustrating gap between that and accountability,” said Sam Gregory, program director at the nonprofit group Witness, during a panel this week at RightsCon, a conference on digital rights and the open Internet.
Witness was founded in 1992 after the Rodney King, Jr. beating, which was videotaped by a bystander. The group aims to further the use of video as a tool for human rights campaigns, partly by better verifying its authenticity through metadata.
Metadata, the group argues, can enable greater verification and trust and enhance data mapping. From Syria alone, the group says, there are more than 500,000 videos online that could provide documentation for human rights violations, but many are not sourced to an identifiable creator.
The proof is in the picture
One of the projects Witness is involved with is a mobile app developed in collaboration with the Guardian Project. Called InformaCam, it uses the smartphone’s built-in sensors to validate the date, time and location of capture.
The app uses digital signatures and encryption to try to ensure media hasn’t been tampered with, and it wants similar capabilities to be offered natively in other camera apps and in key platforms for distributing photos and video such as YouTube, Google+ and Facebook.
The group has pushed for an opt-in “eye witness” or “proof” mode that users could select before creating or sharing media files. On capture, the mode would incorporate and preserve rich metadata, perhaps using the J3M standard, providing a way to check file integrity. Or there may need to be a way of confirming that the data in a video was captured with an app that meets certain verification standards, Witness has said.
They’re lofty goals, and Witness and the Guardian Project acknowledge they face an uphill battle. Challenges include an increasing discomfort with metadata, as more documents are leaked showing governments’ use of it for surveillance.
There are concerns the groups’ sensor technologies, which can locate nearby cellphone towers and Bluetooth addresses, might work too well, and pick up information about people who don’t want to be included in an analysis. “Accepted social norms in one region may not be acceptable elsewhere,” said Nathan Freitas, founder of the Guardian Project.
Still, metadata offers unique opportunities from a human rights perspective. “It’s all there anyway,” Freitas said, “so why don’t we use it for something other than tracking us down?”