Sunday, May 29, 2016

Why America is bad at building new infrastructure

Sometimes small stories capture large truths. So it is with the fiasco that is the repair of the Anderson Memorial Bridge, connecting Boston and Harvard Square. Rehabilitation of the 232-foot bridge began in 2012, at an estimated cost of about $20 million; four years later, there is no end date in sight and the cost of the project is mushrooming, to $26.5 million at last count.

This glacial pace of implementation does not reflect the intrinsic technical difficulty of the task. For comparison, the Anderson Bridge itself was originally completed in just 11 months in 1912. General George Patton constructed nearly 40 times as much bridging in six months as American soldiers crossed the Rhine to win World War II. And even modern-day examples abound; for instance, in 2011, 14 bridges in Medford were fixed in just 10 weekends. ...

In order to adhere to strict historical requirements overseen by the Massachusetts Historical Commission, the Massachusetts Department of Transportation had to order special bricks, cast by a company in Maine, to meet special size and appearance specifications from the bridge’s inception in 1912.

...once construction had already started on the bridge, the contractor, Barletta Heavy Division, discovered that an existing water main would need to be relocated. With the subsequent change order and additional Massachusetts Water Resources Authority permitting processes, an additional 357 days were tacked on to the original contract completion date.

To cap it off, after resisting for years the inclusion of pedestrian underpasses in bridge rehabilitation, MassDOT changed course in 2014 and agreed to revise the design so as not to preclude the construction of an underpass in the future. The contractor then had to move a major utility pipe so that an underpass could fit underneath; meanwhile, another 256 days of delay were added to the project. The entire project is now 22 months behind schedule.

Delay, then, is at one level the result of bureaucratic ineptitude and the promiscuous distribution of the power to hold things up. At another level, it is the failure of leadership to insist on reasonable accountability to meet reasonable deadlines. Perhaps, at a deeper level, it is the failure of citizenry to hold government accountable for reasonable performance — a failure that may in part reflect a lowering of expectations as trust in government declines. ...

The Anderson Bridge is approximately one-sixth the length of the bridge Julius Caesar’s men built across the Rhine in 10 days in 55 BC. Caesar’s feat is admired not just for its technical mastery but also for its boldness. An allied tribe had offered boats to carry Caesar’s troops across the river, to avoid the difficult task of bridge-building. Yet Caesar rejected this offer, on the grounds that it would not be “fitting for the prestige of Rome.”
--Lawrence Summers and Rachel Lipson, Boston Globe, on American sclerosis

Saturday, May 21, 2016

Analytics: Short words are better in Scrabble

Nigeria is beating the West at its own word game, using a strategy that sounds like Scrabble sacrilege.

By relentlessly studying short words, this country of 500 languages has risen to dominate English’s top lexical contest. ...

Once, almost all of Scrabble’s champions hailed from North America or Europe. Most stuck to a similar “long word” strategy—mastering thousands of seven- and eight-letter plays like QUIXOTRY, a 365-point-move in American Michael Cresta’s record-breaking 830 point win in 2006. ...

Global competition and computer analytics have brought that sacred Scrabble shibboleth into question, exposing the hidden risks of big words.

Risk one: Every extra letter on the board is another opening for an opponent to land their own seven-letter blockbuster.

Risk two: Every letter played gets replaced by a random tile from the bag. A bad draw can—and often does—leave players stuck for several turns without vowels or decent letter combinations. After millions of computer-simulated games, Scrabble strategists have concluded that bad draws happen more frequently than previously assumed.

So while Scrabblers still fancy bingos, they increasingly hold off on other high-scoring moves, such as six-letter words, or seven-letter terms that only use six tiles from the rack. Instead, by spelling four- or five-letter words, a player can keep their most useful tiles—like E-D or I-N-G—for the next round, a strategy called rack management. ...

Also, thanks to a design quirk, the board is oddly generous to short words. Most of the bonus squares are just four or five letters apart. ...

Nigeria’s Scrabble ambitions date to the 1990s, when several local fans convinced the dictatorship of Gen. Sani Abacha to make the game an official sport, a designation that brings funding. Nigeria was ostracized from the world then. Scrabble offered one area where the country could redeem its image abroad.

Nowadays, the country of 187 million stages daylong tournaments in stadiums on an almost weekly basis, often with small prizes on the line. Dozens of Scrabble clubs scout high schools for talent, sometimes poaching players. Several of Nigeria’s 36 states have a Scrabble coach on the payrolls.

Of them, Prince Anthony Ikolo was the first to glimpse the potential of the shorter-word strategy. In the late 2000s, the university mathematician had two apps—Quackle and Maven—that let him simulate tens of thousands of possible game scenarios that would result from a given move. The data showed how often a long word would leave the player vulnerable to a counterstrike or a series of bad draws.
--Drew Hinshaw and Joe Parkinson, WSJ, on less being more. HT: Jeff Mosenkis. I also noticed for the first time that the WSJ has fully succumbed to the singular "their."

Sunday, May 15, 2016

Why Koreans don't have smelly armpits

With the weather getting warmer, one of the most common complaints heard from expats is that quality deodorant is near impossible to find in Korea. ... What’s with the lack of quality deodorant in Korea?

Basically, most Koreans don’t actually need deodorant.

Several years ago, scientists discovered a gene that was dubbed “ABCC11”. The team who discovered the gene found that it was a key determinant in whether a person will produce dry or wet earwax. Since then, it has also been discovered that people who produce “wet” ear wax also produce chemicals in their armpits that cause underarm odor when metabolized by bacteria. Those who produce dry ear wax lack these chemicals.

“This key gene is basically the single determinant of whether you do produce underarm odor or not,” Ian Day, a genetic epidemiologist at the University of Bristol, told ...

More than 97% of people who are of European or African descent have the version of the gene that causes smelly underarm sweat, while most East Asians and, according to the study, almost all Koreans do not. 30-50% of People from areas such as Southern Asia, the Pacific Islands, Central Asia, Asia Minor and indigenous Americans exhibit the mutation in their ABCC11 gene that saves them from stinky armpits as well.
--Jane Mahoney, 10 Magazine, on the genetic roots of B.O. HT: JC

Tuesday, May 3, 2016

You actually like the sound of your own voice

So you hate the sound of your own voice. This complaint has become something of a cliché...

...when you hear yourself talk, the sound also comes from an extra speaker of sorts: the bones of your skull. This is known as bone conduction, meaning that when your vocal cords vibrate to produce speech, that movement also causes the bones of the skull to vibrate, and this, too, is registered in the cochlea. Bone conduction transmits lower frequencies as compared to air conduction, so this is one reason why your voice sounds so unfamiliar when it’s played back to you. When you hear the sound through your own head, your brain perceives it as being lower-pitched than it really is, because the transmission via the skull made it sound that way. ...

In a fascinating study from 2013, researchers at Albright College and Penn State Harrisburg played their study participants a variety of different voices and asked them to rate how attractive they thought the unseen speaker would likely be. The twist, however, was that the experimenters did not tell the volunteers that they would also be rating recordings of their own voices. Their results showed that people tended to unknowingly prefer their own recorded voices; they rated their own voices as being more attractive as compared to the other voices they heard, and their ratings for the attractiveness of their own voices were on average higher than the ratings that other people gave them. The researchers note, by the way, that the volunteers were informed afterward that one of the voices they heard was their own, and that they were surprised at the knowledge.