Francis Galton [...]

A Victorian polymath, he is best remembered in educational history for founding the field of psychometrics, a subfield of Galton’s larger interest in anthropometrics.Francis Galton in the 1850s

He invented the use of percentile grades for showing distribution of phenomenon. The height charts that you saw as a kid (at age three 25% of children are above x inches, 50% of children are above y inches, etc.) are a Galtonian invention. This method was a simple but useful way of showing normal distributions.

While Galton’s work in measurement and data visualization was groundbreaking, it also represents the beginning of the Eugenics Movement. Galton himself coined the term after his studies indicated that many desirable traits were hereditary in nature.

For a complete treatment of the history of statistics, see History of Statistics.

Galton is also known for Galton’s Distribution, more commonly called the Log-Normal Distribution.

Galton’s Anthropometric Laboratory at The 1884 International Health Exhibition was an early intersection of data collection and entertainment. [ pdf]

Suicide Clusters [...]

Suicides are not purely stochastic events — one suicide can (and does) often influence others.  Influence can extend not only to probability but also method of execution.

A recent example has occurred in Palo Alto, where a cluster  has taken the lives of four students in the Palo Alto Suicide School District. All four died by suicide along the Caltrain corridor. A previous cluster of five suicides happened in Palo Alto in 2008-09.[ cite]

Cluster suicides are responsible for between 1 to 5% of all suicide deaths.[ cite] Adolescents are most at risk.

The fact that this happened in Palo Alto twice within ten years makes it a rare incidence of an Echo Cluster.

Goffstown NH in the 1990s provides another example.[ cite]

While most people think of suicides as the result of long reflection, the vast majority of sucides are impulsive. See Most Suicides Are Impulsive

A related issue is suicide contagion, which includes Media Influence on Suicide.

Weird Twitter

Genesis of the OER Term [...]

The idea of open educational materials goes far back, and in fact predates most of the technology we associate with the movement today. The use of the particular term “open educational resources” however, has a more recent provenance, and can be traced to a 2002 UNESCO resolution. But what about the initialism “OER”? Where does that come from?

The initial UNESCO report does not mention the initialism. Here is the core of the resolution:

Participants then adopted a Final Declaration (Annex 6) in which they “express their satisfaction and their wish to develop together a universal educational resource available for the whole of humanity, to be referred to henceforth as Open Educational Resources.” [ source]

At the same time, the Hewlett Foundation was looking at the ways in which technology had and hadn’t impacted education and educational practice.  Under the leadership of John Seely Brown and others a new strategic initiative: Using Information Technology to Increase Access to High-Quality Educational Content. This project also used, initially the full term. However, by 2007 the term OER was prominent enough that it formed the title of their report: A Review of the Open Educational Resources (OER) Movement.[ cite]

The impact of Hewlett funds was quick and visible in the Open Education world. To give a sense of scale, between the initial report at the end of 2002 and 2007, a period of five years, Hewlett invested $68 million in OER projects, and it is likely their use of the term had broad impact.

Additionally, early debates tended to pair off OpenCourseWare with Open Educational Resources — explaining that Open Educational Resources was the larger lens. It’s likely that the constant use of the OCW initialism invited a parallel use of OER.

Report: Forum on the Impact of Open Courseware for Higher Education in Developing Countries (Paris, 1-3 July 2002)   [ pdf]

Paris Declaration on OER (2012). [ pdf]

See also The Hewlett Wave


First Women-Directed Theaters in Paris [...]

By the early 1920s it became common for women to act as directors of theaters in France. Sarah Bernhardt and the comic actor Gabrielle Rejane were two of the first, but others soon followed. The following note is from a Charlotte News article from 1921, titled “Parisian Women in Business Do Well”:

07Rejane_1856-1920PARIS, July 18.—The entrance of women into the field of directing French theatrical enterprises has taken on the aspect of real invasion. Mme. Sarah Bernhardt and the late Mme. Rejane were the first French women to direct their own theatres. Today Mlle. Rolle directs the Dejazen theatre. Mlle. Cora Laparcerie, who formerly directed the Bouffes-Parisians, is now at the Renaissance. Mlle. Constance Maille directs the Marigy and Mme. Rasini the Ba-Ta-Clan, besides supervising numerous revues. Jane Renonardt’s new theatre is nearing completion and Maud Lott is about to assume the directorship of the Imperial.

To date most of the enterprises conducted by women have been quite successful.[link t=”Newspaper Article. Subscription Required” l=]

Under Mme. Ragini the Bata clan produced a play by Jean Cocteau. See Cocteau’s Bataclan Play

The Hirschsprung Family [...]

Numerous people have tweeted and blogged this 1881 family portrait as “People Ignoring People Before Cell Phones”. It was rather difficult to track down details on it, so we capture them here.


It was painted by Peder Severin Kroyer, a famous Dane known for painting scenes of 19th century Danish life.

It was commissioned by Heinrich Hirschsprung, a tobacco manufacturer and patron of the arts at that time. He became good friends with Kroyer, and Kroger would have known all of these family members he painted quite well. The painting was meant to show a happy engaged family for which he had a deep affection. [ cite]

Idea for future page: It strikes me right now that the big sin that people are reacting to with cell phones is not engagement with something else among others, but engaging with distant others, over the people in front of you. This violates the “natural order of things”, in a way that interacting with knitting, newspapers the view do not.

Analytics of Empathy [...]

Can data science be used to encourage better user behavior? A number of experiments with League of Legends show perhaps it can:

But Beck and Merrill decided that simply banning toxic players wasn’t an acceptable solution for their game. Riot Games began experimenting with more constructive modes of player management through a formal player behavior initiative that actually conducts controlled experiments on its player base to see what helps reduce bad behavior. The results of that initiative have been shared at a lecture at the Massachusetts Institute of Technology and on panels at the Penny Arcade Expo East and the Game Developers Conference. (post)


The first change they made was to turn off cross-team chat as a default. This dramatically reduced negative chat while keeping use of cross-team chat stable.

The second thing they did was to compile dictionaries of words the negative players would use that were not used by positive players. “It turns out that if you use the dictionaries, you can predict if a player will show bad behavior with up to 80 percent accuracy from just one game’s chat log,” Lin said.

The third thing they did was make the banning process more informational, showing banned members precisely what they were banned for, and what level of agreement was shown on the ban by the judges. This allowed people to use temporary bans as a learning opportunity, knowing the specific behavior that triggered it, as well as helped them understand it was not a single opinion that led to the ban, but a majority opinion.

Of the remedies, the change to internal group chat as a default is the most interesting. It means that players who wanted to say something to the other team would have to explicitly think “This is meant for the other team”.  That small difference is apparently enough to cause teams to filter language, and a testament to the power of defaults (and the importance of making certain things in UX less frictionless).

Open networks can be strangely polarizing. See Streams Don’t Merge

Making bannings more transparent is an application of the Fair Process Effect

The Love Bench is an example of this in architecture.

Opioids, Alcohol, Suicide [...]

In contrast to every other major demographic, death rates for middle aged whites are rising, and rising fast. While part of this is attributable to increasing suicide rates, stunning new research indicates much of the increase is due to drug and alcohol abuse. The effect is centered in the poorest populations, and seems to be related to increased use of various legal and illegal drugs to curb pain. (nyt)


Analyzing health and mortality data from the Centers for Disease Control and Prevention and from other sources, they concluded that rising annual death rates among this group are being driven not by the big killers like heart disease and diabetes but by an epidemic of suicides and afflictions stemming from substance abuse: alcoholic liver disease and overdoses of heroin and prescription opioids. (nyt)

The researchers note that the magnitude of the increase is almost unprecedented: “the mortality rate for whites 45 to 54 years old with no more than a high school education increased by 134 deaths per 100,000 people from 1999 to 2014.”

The only precedent for this rapid a rise in death rates in a demographic? AIDS.

While the ramifications of the finding are enormous, the finding was discovered primarily by accident, during research into increases in the suicide rate.

Despite middle-aged mortality in the U.S. being one of the most studied demographic phenomena in recent history, the pattern had been largely missed. While many had looked at the data, no one had happened on the right slice to show the pattern:

Dr. Preston of the University of Pennsylvania noted that the National Academy of Sciences had published two monographs reporting that the United States had fallen behind other rich countries in improvements in life expectancy. One was on mortality below age 50 and the other on mortality above age 50. He coedited one of those reports. But, he said, because of the age divisions, the researchers analyzing the data missed what Dr. Deaton and Dr. Case found hiding in plain sight. (nyt)

Kevin Drum notes that why the headline is about middle-aged whites, the truth is that the trends are evident in all age groups:

But the paper is being misreported. It’s not just middle-aged whites. It’s all whites. The chart below tells the real story:every age group from 30 to 65 has shown a steep increase in mortality. So why focus just on middle-aged whites? “The midlife group is different only in that the sum of these deaths is large enough that the common growth rate changes the direction of all-cause mortality.” In other words, the midlife group makes for a more dramatic chart. But every age group has shown a similar trend.

To support this he shows his own chart of increases in white mortality at different age groups from the Opioids, Alcohol, and Suicide trend:

Mortality of Americans by age group (source)
Mortality of Americans by age group (source)

In fifteen years death by overdose, suicide, and liver conditions associated with drug and alcohol abuse have doubled in the white American population. So much so that they have changed the shape of all-cause mortality in middle age, but they are impacting other age brackets in profound ways as well.

The suicide trend among middle-aged white men is well known. See Suicide Belt, and Rising Suicides

“More than half of Americans now report a personal connection to painkiller abuse, 16 percent know someone who has died from an overdose, and 9 percent have seen a family member or close friend die.” [(Link)]( quote)

By 2002, the trend was already terrifying and visible. See Opioid Increase 1997-2002

Standards bodies were involved with the increase. See JCAHO and the Opioid Epidemic