Washington Post: Rogue U.S. Service Members Charged With Months-Long Spree of Unprovoked Attacks

Reluctance of military officials to investigate when a soldier's father told them what was happening is disturbing...

The premiere of a documentary produced by U of Minnesota's Bell Museum of Natural History, on pollution of the Mississippi by agricultural practices, appears to have been canceled by the "University Relations Department." See also this from a U Minn bio professor.  He writes that the Vice President of University Relations is married to the president of a PR firm that represents the Minnesota Agri-Growth council.   I know,  I know, correlation is not causation...  Still, I expect outrage from the local Tea Warm Milk and Kahlua party...

Unfortunately (or perhaps fortunately) for the humorist in all of us, Justin Wolfers posts to the New York Times' Freakonomics blog, but is not actually an author of the recent Freakonomics sequel, Superfreakonomics.  (If you picked up on my reluctance to appear implicitly favorable toward the Freakonomists in my previous post, it's largely because of what I've read about Chapter 5 of that sequel, on geoengineering and global climate change---see, for example, this.)  Nevertheless, the paper by Wolfers and his Wharton colleague Betsey Stevenson, titled Subjective and objective indicators of racial progress, looks interesting.  A rough summary would be that in the US, black or African-American people's self-reported level of happiness has increased substantially since the mid-1980s.  Although it is still lower, even controlling for other circumstances like average income, than for white Americans, the gap has narrowed.  The overall gap (not controlling for income) has also narrowed, but the narrowing appears not to be due to increases in income, but appears due to other factors;  Wolfers and Stevenson suspect, but have no direct data to indicate this, that it is due to the decrease in racism black people encounter in day to day life.

There's some bad news for some of us in the conclusion to their Section II:

Comparing these various estimates, we find that controlling for measurable differences in the lives of blacks and whites explains about one-third of the black-white happiness gap in the 1970s and much of this is due to the differences in income between blacks and whites. Turning to the trends over time we see that little of the change over time is explained by the controls. In all specifications the black-white happiness gap—measured relative to the standard deviation of happiness—is closing at a rate of about 0.5 per century. However, this relative change is composed of both a decrease in the happiness of whites and an increase in the happiness of blacks; the decrease in the happiness of whites is larger once controls for objective indicators have been taken into account. Finally, while the racial happiness gap remains large, around two-thirds of this can be explained by differences in observable characteristics.

White folks, it seems, have been getting less happy---an effect that is only made more pronounced by controlling for other measurable factors.  Actually, Stevenson and Wolfers (2009) claim that this is mostly due to a decrease in happiness among white women.

Reading the paper is recommended as a look at a simple piece of modern social science research, how it tries to control for potentially confounding effects, and how tricky it is to deduce correlation from causation. (An example of the latter: "one reason that married people report substantially greater happiness in a cross-section is that happy people are more likely than unhappy people to marry (Stevenson and Wolfers 2007)".)

Just a couple of quick notes on some good wines I've had recently.  Perrin's 2008 Reserve Blanc, a white Cotes-du-Rhone made from 50% Grenache Blanc, 20% Bourboulenc, and 10% each Marsanne, Roussanne, and Viognier, is what a white Cotes-du-Rhone should be, and one of the best ways you could spend ten bucks on a white wine (I paid US$9.98 at Ta Lin in Albuquerque).  I've posted before on Perrin's excellent reds; this shares some characteristics with them, notably a certain almost glyceriny smoothness, but without stickiness... and a characteristic that I'd call "watery" except that sounds bad, and this is good---it's probably associated with being relatively low in alcohol, and not extremely high (nor excessively low) in acid or tannin.  Mainly, it's got intense, but refreshing rather than tiring, flavors, flavors akin to those of a good chardonnay-based wine from the Maconnais, but maybe slightly fruitier and slightly more floral.  And much cleaner, more balanced, and more intense than most of the Macon-Villages and such I come across (in North America, anyway) in this price range.  Some resemblance to nice Spanish whites, like Muga's white Rioja.

A favorite wine at a party I recently attended was the 2006 Clos du Bois North Coast Cabernet.  Nothing super-complex, just a very drinkable, fairly rich, but balanced, Cab.  It Googles up at $12-14, which I'd say is reasonable although not a steal.  (A 2007 Acacia Pinot Noir, tasted afterwards, seemed excessively jammy and just not enjoyable to drink, which surprised me as their Pinots have seemed decent value even in the $20 range, in the past.  Perhaps not fair to judge based just on one bottle.)

Haven't been blogging much recently, but I certainly need to post this link on Deolalikar's claimed proof of P ? NP:  Scott Aaronson saying (nearly a month ago) that a consensus has developed that it's fatally flawed.  Scott's post is worth reading in part as a preliminary guide to what you'd need to learn about to really tackle this problem---and what to do to increase your --- and others' --- confidence you'd actually solved it.

I think I already linked Richard Lipton's less conclusive "Fatal Flaws...?" post.   Lipton wrote another good post a few days later (a week after the proof was posted).  These lack a definitive statement that "it's dead" (one section is labeled "it ain't over till it's over".  Scott seems to have concluded on the basis of most of the same evidence that it was almost certainly over.  The fact that googling "Deolalikar proof" turns up nothing later than Aug. 12 on the first page (I see a broken link from Aug. 18 on the second page) also suggests it's dead.

Mainly storing a link to this interesting paper that raises some questions:

William Press, "Strong profiling is not mathematically optimal for discovering rare malfeasors".  Weak profiling apparently is---under the assumptions of his model, which seem to me a rather imperfect fit, but not wholly irrelevant, to most real-world terrorist-screening scenarios.   No implied view from me on the correctness or relevance of these results, pending more careful reading and thought.

Aha, just realized it's that Bill Press--former (pre-2004) deputy director for Science and Technology at LANL (curently at UT Austin, on leave from LANL).

Oh yeah, link is from The Yorkshire Ranter, who got it from Bruce Schneier.