GIA to use P2P to Avoid Litigaton

DrMorpheus writes "With the recent demise of the Bush administration's controversial Terrorist Information Awareness (TIA) programme to monitor everyone in the US, citizens now have a chance to get their own back. A website to be launched later in 2003 will allow people to post information about the activities of government organisations, officials and the judiciary. The two MIT researchers behind the project face one serious problem: how to protect themselves against legal action should any of the postings prove false. The answer, they say, is to borrow a technique from the underground music-swapping community. Instead of storing the data in one place, they plan to distribute it around the internet in a similar way to the notorious Napster software that got music file-sharing under way."

Comments

Rumor mill run wild.

[Hawthorne01]

Maybe it's just me, but without a system in place for fact-checking and followup, it seems to me this would quickly turn into a breeding ground for conspiracy theories of the worst sort, be they from the Right or the Left or anywhere in between.

Validity

[Boing]

Does it seem to anyone else like the GIA program is going to drown in its own chaff? I mean, if it's really supposed to be giving people "insider" information about the government, how are they going to confirm that any submission is true? Think of all the just-shy-of-slanderous commercials that air around election day... and that costs money. If people can anonymously, cheaply spread whatever "information" they want about their least favorite candidate, we'll never know whether anything we read is accurate.

Re:Validity

[rsborg]

Does it seem to anyone else like the GIA program is going to drown in its own chaff? I mean, if it's really supposed to be giving people "insider" information about the government, how are they going to confirm that any submission is true?

I have an awesome idea... They should allow a certain percentage of the users to "Moderate" in a rotating fashion, depending on their usage statistics.
Ah, you say... who will police the police? Good question!
They should also have a group of "Meta-moderators" that "Moderate" the "Moderators", thus preventing malicious "Moderation"... Wow, this is an incredible way to improve the signal to noise ratio!... I should patent it!

Unless, of course there's prior art :-/

Solution looking for a problem?

[kawika]

What makes them think that they would somehow be responsible if they were to do something simple and straightforward like a discussion forum? The high court has already ruled on several of these kind of cases and free speech won. For example, they struck down the CDA and they also said that Yahoo didn't have to rat out an anonymous Yahoo Groups poster.

Why use an obscure technology? That will have more of a chilling effect on active participation than any other factor.

Demo site available at MIT

[Jonah Hex]

It is based on a site that Chris Csikszentmihalyi and Ryan McKinley of the Massachusetts Institute of Technology's Media Laboratory set up in July. That site encourages members of the public to post information about organisations, officials and politicians, such as their business links and the source of their campaign donations.

The demo site mentioned above is pretty damn cool, even offers monitoring of C-SPAN/C-SPAN2 as well as a "robot" that watches and records appearances by "people, pundits, or politicians who have recently been entered into our facial database". Hopefully the system looks this polished when it moves to being a P2P network!

Jonah Hex

Use anonymous reputations to confirm data

[lawpoop]

It seems to me that there's several ways to confirm the validity of the data.

One, use anonymous, public keys. Create a reputation system where anybody gives a personal rating to anyone else. (Sorta like Slashdot "friend or foe"). Reputations build over time, and some sources will be more reliable than others. You can decide who you will trust to tell you what's going on. You can rate sources, and you can see how others rate sources.

Two, use what's called 'triangulation' in interview journalism. If three or more 'independant' sources agree on a datum, is more likely to be true than if just one says it's true. (if there is some kind of sinister collaboration to hack the triangulation system, fall back on the reputation system).