Sunday, February 26, 2006

Protecting Sources in the Internet Age

Earlier this month, the Washington Post ran an interview with a shadowy young hacker who breaks into insecure computers and harnesses them to run a variety of money-making projects like distributing E-Mail spam for his clients.

The story in the Washington Post was carefully written to avoid giving away the identity of the hacker. There were sketchy descriptions of the otherwise unnamed small town where he lived.

The story also included a photograph of him in shadows, so that his face could not be seen.

But there was something else about the photo that the editors of the Washington Post neglected to reckon.

Photographic images stored in digital format contain embedded text annotations (called metadata) that are used to help organize and classify libraries of digital images. Some of the metadata is inserted automatically by the digital camera. Some of the metadata is inserted manually when images are run through PhotoShop. Photo journalists routinely add such annotations to help them keep their image libraries sorted out. In this case, the metadata included the name of the small town in Oklahoma where the photographer had taken the original picture.

Someone who knew about metadata in digital images opened up the photo from the Washington Post's story (as posted on the newspaper's website) and examined the metadata. That was enough to complete the missing parts of the picture. The Post hasn't confirmed whether the sleuthing is accurate. They are remaining mum.


Dave Kennedy said...

Here's a very interesting article from MSNBC.
Metadata is not only a problem for geeky hackers, but for the government and it's secrets as well. I think metadata could have an interesting future.

dairy diva said...

That is all very interesting. Being technologically challenged I am shocked. It is going to be interesting to me now to follow this metadata thing. I did not know that was possible. It could pose problems for a lot of people.