Sunday, January 31, 2010

Three Kinds of Information Sensitivity

OK, a naive meditation on three modes of paying attention to the world.

Pretend you are a politician, perhaps a senator or member of congress. What do you pay attention to?

Some would have us believe that poll numbers are the most important. You open your mouth, emit a sound bite, the media disseminates it, people react and respond to polls, and you adjust accordingly. Depending on your point of view, that's either democracy in action or appalling pandering. In either case, the opinions/reactions of "the people" are aggregated via some presumably reliable and accurate method.

Another theory would be that you are listening to powerful interests who have your ear and who donate to your campaign. Your comments are probably a little more proactive than reactive -- they've let you know what they want to hear and so you make sure you say it. But as above the whole thing is a cycle -- we get the initial attention by saying things and then it cycles from there. In this case, though, the method for aggregating the reactions (and pre-actions) of donors is harder to suss out. Tally up the dollars? Is there a pecking order? Or a "one topic each" rule?

A third approach would be that you apply accepted methods of policy analysis and make use of trustworthy data to decide what policies would best achieve desired aims. Here information is aggregated and decisions made using generally accepted (and open) methods. Of course, deciding on those aims is an information problem that can bring us right back into one of the first two approaches, but we'll set that aside for the moment.

My guess is that a system COULD run on any of these three approaches to information processing. What presents a challenge to govern-ability, though, is when one or more of these is the public face of what's going on while another one is what's going on behind the scenes. Or, worse, when the actors themselves don't really have a handle on when they are using one or another to try to ascertain how to govern.

And yes, this could be seen as an attempt to translate direct democracy, some variation on aristocratic pluralist democracy, and technocracy (help me on the terms, polisci friends) into information terms.

Tuesday, January 19, 2010

New Digital News Outlet from KALW

This week KALW is launching its new local digital magazine to complement their broadcast work.  The new site has a way for community leaders to plug in and help them do a better job of reporting on the arts and other community events and issues.  Users can become "community correspondents".  Check it out, help them tell others about it and together we can do a better job of becoming the media we want to create.

Here's the magazine: http://www.kalwnews.org/

This is the community page: http://www.kalwnews.org/community

And here's their FB group to stay in touch: http://www.facebook.com/pages/KALW-News/195280839624

Friday, January 08, 2010

Great Info Blog and Interesting Sounding Conference (NY Feb)

Check out Graham Webster's excellent blog for some insightful essays on topics not unrelated to those I'm writing about here. It's called:infopolitics/

While reading through that blog, came across mention of an interesting conference to be held in February at the New School in NYC: Conference on Information Flow Restrictions at the New School. It actually had me looking at plane ticket prices and thinking "whom do I know that I could stay with...." If you know how little I like to travel during the semester, you get the idea that I was intrigued.

Wednesday, January 06, 2010

Those damn unconnected dots again (rough draft)

An article in the Times, under the headline "Obama Says Plot Could Have Been Disrupted," reprises the metaphor of "connecting the dots" to describe different pieces of information having been in different heads, but never getting put together in one head that could make sense of them.

It is reassuring that Obama's speaking bluntly about organizational performance rather than riding roughshod over the constitution, but, as argued in an earlier piece ("Mind the Gap"), the idea that it's a simple problem of dot connecting is a basic misconception.

How do you hear "connect the dots"?  One version is reminiscent of a detective show or Agatha Christie novel; the challenge is to assemble hints -- pieces of information that, alone, are not conclusive proof of anything -- in such a way that the "answer" emerges as a sort of logical necessity.  The "logic" is in the mind of the beholder, but that's all.

A different version is reminiscent of the we draw lines between stars and come up with "constellations."  Two things are important.  One, the stars are not really next to one another -- the viewer is the one who sees them as points on a plane and interpolates and extrapolates the other vertices of the figure.  Two, there's no there there -- the crab in cancer or the warrior in Orion has to be brought to the observation by us.









The first requires us to have all the pieces on the table and be open to what they "tell us" when seen together.  The challenge for intelligence agencies is to put the information from various sources onto the same table.

The second requires us to decide what to pay attention to and what to ignore (left), how to connect and not connect (middle), and what to add that's not there (right).

If we increase the degree of information sharing we fill up our field of view with more and more points and the dots get harder and harder to connect.

On the other hand, if we ask the different agencies to filter the information then we are back in hot water because none of them know what they are looking for.

The president was furious about the failure of the system to see "the red flags" and intelligence agencies are reported to have said that the information they had was "vague but available."  The problem is that flags are not, in general, a priori red.  Presumably, some smart people are thinking about how systems see and things like that; hopefully, they don't just think of it as "connect the dots."

We observe with some irony that the actual policy response to the problem -- at least the response that's been announced -- is in fact to gather more information via increased screening.

Oh, and if we look up "connect the dots" in Wikipedia you get a short article about a children's game. It bears a Wiki-warning: "This article may require cleanup to meet Wikipedia's quality standards."