Metadata displayed on InformaCam testbed. Image courtesy of The Guardian Project.

When Sharing More Metadata Is a Good Thing

In early 2014, reports of rhinos being hunted down by poachers in South Africa thanks to tourists’ Instagram postings struck a nerve. Safari operators warned adventurers to turn off their geotagging settings and to refrain from mentioning the locations of rhino sightings on their social media feeds.

But it’s not just rhinos that are at risk from over-sharing. It soon emerged that the supposedly anonymous app Whisper has been tracking users. In both cases the threat came from metadata — such as geotags, timestamps, dates, IP addresses and mobile tower data — that is associated with every piece of media we create and every electronic communication we write.

In the post-Snowden world, metadata has taken on ominous connotations, whether on a wildlife safari or in association with mass violations of people’s privacy.

But metadata — particularly in images and video — can be harnessed for good. It is especially important for the tens of thousands of citizen witnesses who document rights violations, cruelty, and injustice, and want that material to be found and trusted. If ordinary citizens and activists had a better understanding of metadata and more control of all the communication tools we use, the benefits could be myriad. It would be a win for people making choices about being seen and heard, for news companies looking for the latest verified citizen media, and for human rights advocates searching for critical evidence.

People in Ferguson, Rio de Janeiro, Aleppo and beyond, share images as a call for action, as evidence of wrongdoing, and an attempt to mobilize their communities.

But they do this amid a flood of content. Well over a 300 hours of video are uploaded to YouTube each minute, and there are more than 250 billion photos on Facebook. More than half a million media items showing rights violations in Syria have been posted on the Internet since protests began in 2011. So a critical image may not be seen, it may not stay around long on a commercial platform (either because something else takes its place on a feed or because it gets taken down), and it may not be trusted even if it is seen.

How can human rights organizations such as WITNESS and technology companies best support people who want to be citizen witnesses to wrongdoing?

WITNESS and other similar organizations need to work to improve filming techniques, distribution strategies, and methods for organizing and preserving material so that more people can be effective agents for justice. Further, they need to help people find out what tools are available to both protect their identities and amplify their messages. When activists produce a large volume of images — such as in the Syrian conflict — it’s easy to lose track of individual videos, and also details about each of them: who and what are depicted, and where and when the video was shot. Proving the provenance of a video using metadata can be crucial to that footage being useful in an investigation or future trial.

The Internet and mobile communication and media creation are being used increasingly by activists. We believe that individuals should be able to easily choose between sharing media with greater personal visibility or sharing it anonymously.

Platforms should give individuals better choices about how much data to add to their media and what they share, so that activists’ and citizen journalists’ media can be located and verified when they want it to be. At WITNESS we’ve advocated for an easy opt-in “proof” or “witnessing” mode that helps facilitate this, and for easier ways to identify where media has additional metadata. Such a mode would be useful to the citizen journalist witnessing an extraordinary news event, or even a person who’s just had a fender-bender.

Conversely, tech companies need to provide better and safer options for what we call “visual anonymity” in the design of their platforms and apps. When it comes to photos and video, this means that YouTube, Facebook and others should offer options to obscure metadata and do things like blurring faces when people need to be anonymous. This can help citizens choose whom they identify when they share images. The key in both cases — sharing more data or obscuring — is the choice being made by the individual.

There are also roles for policy-makers whose decisions on mobile and Internet privacy set the ground rules for how people can use tools and how others can inhibit, watch or suppress them. And for journalists too, in an age of information overload, we need to work out how newsrooms, activists and the rest of us locate and then verify media that offers crucial evidence, makes a compelling argument or describes a pattern of abuse. And we need better and more rapid ways to do this.

What we need is a deeper knowledge of the potential good that metadata can do when we enhance individuals’ ability to control it. Without it, we leave open the possibility for false images and stories to inflame, distort and jeopardize the credibility of genuine and courageous witnesses on the ground. We increase the possibility that critical images and voices go unseen, unheard, unnoticed, unfound, and undervalued.

Sam Gregory (@samgregory) is WITNESS’ Program Director.

Originally published at The statistics in this piece have been updated to reflect public numbers as of May 2015.

Show your support

Clapping shows how much you appreciated WITNESS’s story.