A Little Remembered Fact? White Lynchings

Matthew Yglesias blows up an American Spectator piece on the lynching of Sherrod's father, but he includes a graph showing lynchings from 1882 on.  In the first 10 or so years, the graph seems to show more whites being lynched than blacks (caution: my memory is that statistics on lynchings were very hard to gather so need to be taken with a grain of salt).  Of course, in proportion to population, the rate for blacks was always higher, but white lynchings remind us just how violent a country we used to be.

On the Limits of Transparency

One of the problems with "transparency" is: who cares?  The data may be out there or available, but unless there's someone with enough interest in the submit to dig into it and make a story out of it, there's little impact.  Part of the solution can be auditors/IG's.  See this piece from the World Bank blog.

To cite one example, we did a big, RCT study on what reduces corruption in community programs. Whereas my entire team thought that increasing participation and transparency would be most effective, in actual fact increasing the frequency of locally publicized audits had far greater effects.

Either I'm Blind or Opengovtracker Disses USDA

I don't see USDA listed on this site.

Dealing with People, Lawmakers Will Keep This Secret?

Here's a discussion of a guide to dealing with "distressed constituents", by which is meant, not constituents who have been impacted by a natural disaster, or the recession, but those, not to put it mildly, who are nuts. (My term, not theirs.)  Anyone who has dealt with the great American public knows they're out there.  H. L. Mencken thought much of the public was.  But it's not something any Congressperson is going to advertise: "hey, I've had my staff read this great guide so now we'll be able to deal with you better."

Audience Approval of What in Sherrod's Speech?

William Saletan does a careful analysis of the audience's reaction to Shirley Sherrod's speech.  My own reaction, which I've hesitated to state because I don't have a lot of experience listening to majority black audiences responding to black speakers and ministers, was that they were giving the audible feedback which seems to be the norm in such settings.  Saletan's analysis is much more careful than that, and convinces me that Breitbart's claims are wrong.

Change in the Armed Forces

Women/mothers make Sailor of the Year.  Turns out we have 4 sailors of the year and women won all 4 spots.  Article says they're about 16 percent of the Navy.

What's Possible and What's Not

The Priest/Arkin series on the post 9/11national security bureaucracy is filled with interest.  As a sidelight, two small scenarios show the difference between what's possible in IT and what isn't.  Here's what's possible:

To understand how these firms have come to dominate the post-9/11 era, there's no better place to start than the Herndon office of General Dynamics. One recent afternoon there, Ken Pohill was watching a series of unclassified images, the first of which showed a white truck moving across his computer monitor.
The truck was in Afghanistan, and a video camera bolted to the belly of a U.S. surveillance plane was following it. Pohill could access a dozen images that might help an intelligence analyst figure out whether the truck driver was just a truck driver or part of a network making roadside bombs to kill American soldiers.
To do this, he clicked his computer mouse. Up popped a picture of the truck driver's house, with notes about visitors. Another click. Up popped infrared video of the vehicle. Click: Analysis of an object thrown from the driver's side. Click: U-2 imagery. Click: A history of the truck's movement. Click. A Google Earth map of friendly forces. Click: A chat box with everyone else following the truck, too.

And here is what's not possible,  from the first article:

The practical effect of this unwieldiness is visible, on a much smaller scale, in the office of Michael Leiter, the director of the National Counterterrorism Center. Leiter spends much of his day flipping among four computer monitors lined up on his desk. Six hard drives sit at his feet. The data flow is enormous, with dozens of databases feeding separate computer networks that cannot interact with one another.

So, in one case IT is able to correlate information from different sources into one presentation; in the other it's unable to.

Why the difference? This is just speculation, but I see two key differences. In the first instance different kinds of information are being brought together and the sources of data probably were created within the last 5 years.  In the second instance similar kinds of information from different bureaucracies are coming in, and probably all of them had deep historical roots.  (For example, FBI's case file system dates back to J. Edgar Hoover's prime in the 1920's.)

Best Sentence of July 23

"But who expected in 2003 that in 2010, the president of the United States would have "Hussein" in his name but the president of Iraq wouldn't?"  Tom Ricks at the Best Defense

Curves

My mind's on curves. Not female curves, but something more nerdy. 

• On the one hand, there's the curves of economics.  I'll probably get the terms wrong, but there's a supply/demand curve that shows for any price there's a market clearing point at which supply and demand are in equilibrium.  And implicit in that, or maybe something else, is a curve that shows the marginal cost of producing something declines as production goes up--mass production saves money in other words.
• There's also the learning curve.  The more time you've spent making stuff/learning a subject, the easier it is. 
• Finally in my thoughts is the Pareto 80/20 rule.  Now it's not usually talked of as a curve, but if you visualize it you can see it.  To me, it's the exact opposite of the supply/demand curve and the learning curve.  I used to use it in discussing software development.  It would be easy to do software to handle simple cases, but as the cases got more difficult it would become harder and take longer.

It seems this week I've got more questions than answers; because in this case I can't figure out how the curves work together, if they do.  Or is it just a case of different tools for different situations.

Must Reading for Those Discouraged by the State of Dialog on the Web

See this post of Dan Drezner's.