Drastic Differences in Leaking Websites | MIT Center for Civic Media

Drastic Differences in Leaking Websites

I have been researching document processing for my Intro to Civic Media project. Specifically, I have been examining the varying goals as well as processes and tools used in the intermediary steps after a leaked document is received but before it is released. I hope to identify both strategies that work particularly well and difficult areas where few tools or processes are available to help. In the past week, I have been conducting interviews and updating LeakWiki.

So far I have interviewed people from Cryptome and Associated Whistleblowing Press. A few people have requested that I email them questions. I did not expect this initially but it makes a lot of sense as they can provide more detailed information and coordinating interviews is often difficult with time zone differences. I currently have interview questions out to Public Intelligence and Juzne vesti. I am coordinating with a few others to find the best time to interview.

As I complete my interviews, I am working on LeakWiki case studies for each site or major document set and tool profiles for all specific tools mentioned. I currently have a Cryptome case study up and am working on an Associated Whistleblowing Press case study. The case studies are based on information from the interview, the organization's website, and articles or other external information about the organization. The interviews are also available on LeakWiki. I hope to turn LeakWiki into a useful resource for anyone starting or running a leaking website as well as those researching or looking to improve the leaking process.

The first two organizations I interviewed were opposites in many respects. My cryptome interview revealed that Cryptome does very little document processing and is primarily interested in increased transparency and free information. More than just transparency, Cryptome wants to promote education and thoughfulness. Cryptome is also quite centralized and fairly minimal. It does not use any tools or processes beyond basic email (with PGP encryption) and word processing. Sources are asked to protect themselves and everything is published (including my interview). John Young also made some interesting points about leaking as an extension of authority. Opinions on his responses seemed mixed on Twitter and reddit but I can understand his reasoning. It actually reminds me of last class where Emi mentioned that not just hiding information but also releasing it gives organizations power as they can control which information is most visible. The interview provides some more insight into Cryptome's process and philosophy.

The second interview was with the Associated Whistleblowing Press (AWP). AWP receives documents but it is primarily concentrated on finding restricted or secret documents that reveal wrongdoings or are otherwise significant. In other words, it is focused on whistleblowing. AWP works with local nodes (the first of which just opened) to bring documents and articles about those documents to a global audience. It relies on the local nodes to do most of the document processing and analysis which makes sense as the people in a particular country, region, or city will best understand documents pertaining to local issues. Completed articles are published on whistle.is. Unlike Cryptome, AWP tries to protect its sources. To some extent sources must still protect themselves but they use the GlobaLeaks framework and do things like remove metadata from documents to increase security and anonymity.

As far as document processing itself goes, I am finding that there are few good solutions. Having local nodes do analysis is a great idea and some sites seem to be experimenting with document tracking systems and crowd sourcing (for research, translation, and fact checking) but few of these are widely used now.