When Open Government Information Awareness first launched, I found it as an interesting idea — enable citizens to keep track of their government. It seemed the perfect retaliatory reaction to TIA, the once called Total Information Awarenes and now Terrorist Information Awareness, the contraversial DARPA program to help the government keep track of people/terrorists. Quoting GIA’s FAQ
The system presents itself to users as a Web site, but is actually a suite of information technologies that actively peruse data, accept contributions, and post alerts about government. The system will accommodate information of almost any type, allowing users to sort through volumes of information which would otherwise be unusable. More importantly, the system allows for people to submit any information, while retaining anonymity, but while also being identified as a consistent source.
This system allows citizens to post information to help others understand what is going on with their government and their elected officials. Its like a bulletin board system for people interested in what their government is up to and some of the tools on the site also allow it to be a specialised Google for aggregated government-related content. But the hard problem, is how do you know what people are contributing is true? What if they are total rumours, or worse yet, outright lies? The people running GIA could take legal responsibility for that mis-information. And I think worse yet, the public is getting bad information from which they are making judgements.
They hope that following the Napster approach will get them round this problem. Instead of storing the data on a single server, so-called peer-to-peer networks hold data in a number of locations around the internet, from where it can be downloaded directly.
While this type of P2P solution would alleviate the legal problems the creators of the system may face (similar to the Napster vs. Gnutella architectural decisions), it also has the possibility of tainting the quality of the information that gets out to the public. Instead of building a reputation engine into GIA which would allow people to rate, in a P2P fashion, whether the contributed information is correct or not and whether the contributor is trustworthy or not, GIA is using P2P to spawn off more copies of itself. This seems to provide more opportunities for the public to get bad knowledge of what their government is up to. A trust system on one centralized system is hard enough to get right — but a trust system spread out amongst lots of autonomous servers is even harder, if not impossible.
Think of it this way. From the iTunes Music Store, you can download music that you know will be right. Its going to be of good quality and it will (with very good probability) be the piece of music that you think you are downloading because you are getting it from a reputable source (one that takes the time to verify). If instead you go onto Kazaa and try to download the same piece of music, you have no guarantee the song you are downloading is the song you wanted. It could be mislabeled. It could be a horrible recording of it. The creators of GIA strive very hard to be neutral in this “information battlefield”, but if they want to provide a service to the public, they have to be neutral in a “NPR” sense and at least do some work in verification of information, deliver some means of proving credible sources, or at least marking which information they have no way to proving.
Spawning off many versions of GIA to be controlled by different people seems to be a way of taking the legal burden of getting any of the information right off the site maintainers. But this seems to be very self-serving to the creators, and not serving at all to the public. The problem really is, unlike the iTunes and Kazaa example, that downloading uncredible information about the government is not simply annoying — it is very dangerous to the public as a whole.