Public Intelligence Interview
Below is an interview about Public Intelligence's leaking/disclosure process conducted by M.C. McGrath with Michael Haynes from 11/26/2012 to 12/08/2012.
 Part One
M.C.: What is the ideal outcome of a release of a document or set of documents for public intelligence?
Public Intelligence: The ideal outcome is simply to do justice the material being presented. Our particular intent is not to force transparency on anyone, but simply to inform people. We feel that access to information is a fundamental human right, allowing individuals and societies to orient themselves, make decisions and have some modicum of control over their own lives. Thus, our only intention in publishing a document is to provide the material to the widest audience possible, presenting it in a manner that is clear and straightforward so that readers can understand and benefit from the information that it conveys.
M. C.: Could you walk me through the process your documents go through from when you receive the documents (or even before) to release of the documents? I am interested in the whole leaking process but particularly focused on document processing steps like verification, removal of names, analysis of a document, etc. What are the most time consuming and/or difficult parts of this process?
Public Intelligence: The majority of the material published on Public Intelligence is, as the name implies, obtained from public sources. The purpose of the site has always been to make information more accessible to people by aggregating original source material from the public sphere that is restricted, censored or otherwise difficult to obtain. This means that we often know the source of a document directly because we obtain the documents ourselves and can thus determine their authenticity, provenance and purpose by understanding the nature of the source from which we obtained them.
Once a document or other information is acquired it is reviewed to determine its relative value and whether more thorough analysis is warranted. The intensiveness of this process varies from document to document based on a number of factors from personal interest to the availability of time. However, the function of our analysis is always aimed towards making the material more accessible to people by elucidating key points or other useful pieces of information that the document may contain. We do not want to function like an intelligence agency or a major media company as these forms only further the current system where information is wielded by institutions to benefit a few, rather than benefiting all. The most important function that we can perform is simply to get the information out to people. They must create context, meaning and relevance for the information themselves.
We almost never perform redactions to a document, as this too would defeat the purpose of making information freely available. The only circumstances where we have done so is to remove intensely personal and/or legally protected information that serves no educational function, like a person's social security number. We also have, on a few occasions, performed redactions to ensure that the source of a document is not discoverable. Over time, this allows for more information to enter into the public domain than if we were to simply post the document with nothing removed.
The most time-consuming part of this process is always the acquisition of material that is worth publishing. People are rarely, if ever, going to risk their lives and/or livelihoods to come and drop material in your lap. This is one of the things most people don't understand about so-called "leak" sites: they receive far fewer "leaks" than you would imagine.
M.C.: What tools or software do you use throughout the leaking process and for what? How well do these tools work? Are there any tools or improvements to the tools you use that could make the process easier?
Public Intelligence: We try to incorporate free and open source software as much as possible into the publishing process. This is intended to demonstrate the degree to which people with minimal technical experience and small amounts of money can operate publishing platforms capable of disclosing information and helping to inform those around them. So, we use Wordpress to publish the site and other free software for communication, encryption and similar functions.
M.C.: How many people generally help with document processing? Do you ever ask experts in the area relevant to the documents to help with analysis? Have you ever tried to crowd source any part of the leaking process? Have you ever worked with media partners? If so, what sort of media?
Public Intelligence: Not many. We don't provide the documents we obtain to experts because this would defeat the purpose of our project which is to provide information to the public, not create a cartelized collection system for profitable media companies. We have tried to crowd source parts of the analysis of documents, but have found that it is difficult to attract individuals who are willing to contribute meaningfully. We have never provided documents to media outlets or reporters prior to their publication on our site. At times, we have discussed material with reporters who had questions about its provenance or implications. This can sometimes be an intensive process.
M.C.: Do you think of leaking and whistleblowing as synonyms or different terms? How would you define each of them (or one of them if you think of them as synonyms)? And how do you think of Public Intelligence in terms of whistleblowing and leaking?
Public Intelligence: I don't really like either term. Leaking sounds like an accident and is a favorite tactic of those in government. Whistleblowing sounds like something done by a hall monitor in grade school. Telling the truth shouldn't require special terminology. We should endeavor to build a culture that is built upon truth-telling and openness, rather than one that lauds these activities as praiseworthy or unusual in nature. Public Intelligence publishes information that is of value and then tries to defend that information from attempts to censor it. That is all. Call it what you will.
M.C.: How effective has Public Intelligence been in achieving its goals and ideal outcomes with release of documents? What are or have been the main obstacles to achieving your goals? Is there anything that has or could help Public Intelligence be more effective?
Public Intelligence: We have had tremendous success in our primary purpose: to publish information of value that is difficult to acquire or not otherwise publicly available. In fact, we do this on a daily basis. The problem comes in trying to get people to pay attention to the material. We struggle often with how to make complex material accessible and relevant to audiences that are not necessarily motivated to understand it.
 Part Two
M.C.: You mentioned that you tried to crowd source parts of document analysis. What parts did you try to crowd source and how? Did you have any success or was it just too difficult to find people who would contribute?
Public Intelligence: We tried to aggregate readers' efforts to find interesting material in a set of documents from NATO and the UN regarding the war in Afghanistan. We used simple, publicly available forum-based tools to try to do this. But, not many people ultimately contributed.
M.C.: How do you determine the value or significance of a document?
Public Intelligence: There are a number of factors in the appraisal of material that are difficult to articulate, but they generally revolve around the public worth of a document. Does the document describe or reveal something about a public process? Does the document relate information that is informative or educational in a broader societal context? Who produced the document and what are the rules/laws governing its disclosure or publication? Has someone else already published the document? Does the publication of the document contribute in some way to the body of available information on a topic of public importance? These are some of the questions that might be asked.
M.C.: Do you have any particular guidelines or practices you follow during analysis? For example, how do you avoid imposing context while still making the information accessible?
Public Intelligence: We do what good journalists are supposed to do: explain the material in a manner that attempts to limit our own personal biases. Sometimes we are more successful at this than others. Journalism/leaking/disclosure/whatever you want to call it as much an art as it is a craft.
M.C.: Have there been any releases people have paid more attention to than normal? To what extent do you think the attention the material receives is due to the analysis and accessibility versus the nature of the content?
Public Intelligence: In the thousands of documents we have published since we began doing this several years ago, hundreds have received coverage in various types of media all over the world. However, the most popular stories are always the ones that are most marketable due to some salacious or attention-grabbing quality, not necessarily their true informational value. For example, we published a document and an article earlier this year on an IARPA program called Catalyst that is designed to act as an "entity extraction" and "semantic integration" system across the entire intelligence community. The system is designed to parse large-scale datasets and automatically extract and organize meaningful information for storage and analysis in later investigations or intelligence gathering. It's a dry topic, but has tremendous implications for a number of issues like widespread electronic surveillance. Basically no one paid any attention to this. On the other hand, a few months prior we got a story into the mainstream media that was so popular that it was reported everywhere from Indian TV to Good Morning America. It was about how apparent gibberish in a early version of an al-Qaeda magazine was in fact cupcake recipes. There were headlines like "Al-Qaeda Mag Attacked in Operation Cupcake." That should give you an idea of what I'm talking about.
M.C.: Do you use any tools internally for tracking documents you are analyzing or otherwise waiting to release? If so, what tools? Similarly, do you use any software to aid in your analysis process? And is there any software or tool that would be helpful in your analysis process?
Public Intelligence: No, not really.
M.C.: Generally, how do you obtain your information? I don't want information on specific sources but you noted that you obtain the documents yourselves so am interested in how you get difficult to obtain or restricted material.
Public Intelligence: It's all around, you just have to look for it.