(p. 14) For scholars and popular writers alike, the Great Depression has long been a kind of economic Rorschach test. Free marketers look at the economic disaster and blame the Smoot-Hawley tariff, which inaugurated a global trade war; monetarists attack the Federal Reserve for its tight-money policies; Keynesians berate Herbert Hoover for his attempts to balance the budget as the crisis worsened.
. . .
Generally, . . . , Morris is remarkably evenhanded, giving both sides of scholarly debates in deep detail. This is particularly the case in his coverage of the New Deal, where he weighs the practical effects of the dizzying array of policies begun by Roosevelt, from his devaluation of the dollar to the creation of the Civilian Conservation Corps. And Morris explains in accessible prose how economists have used modeling to study the New Deal (he wryly notes that this “is still a work in progress — if only because results are often suspiciously consistent with the political dispositions of the modeler”).
Category: Methodology
Complexity of Drug Discovery Requires More Than A.I.
(p. B1) Every two years, hundreds of scientists enter a global competition. Tackling a biological puzzle they call “the protein folding problem,” they try to predict the three-dimensional shape of proteins in the human body.
. . .
Mohammed AlQuraishi, a biologist who has dedicated his career to this kind of research, flew in early December to Cancun, Mexico, where academics were gathering to discuss the results of the latest contest. As he checked into his hotel, a five-star resort on the Caribbean, he was consumed by melancholy.
The contest, the Critical Assessment of Structure Prediction, was not won by academics. It was won by DeepMind, the artificial intelligence lab owned by Google’s parent company.
. . .
“It is not that machines are going to replace chemists,” said Derek Lowe, a longtime drug discovery researcher and the author of In the Pipeline, a widely read blog dedicated to drug discovery. “It’s that the chemists who use machines will replace those that don’t.”
. . .
(p. 5) Working with two other computer scientists, the DeepMind researcher Rich Evans homed in on protein folding. They found a game that simulated this scientific task. They built a system that learned to play the game on its own, and the results were promising enough for DeepMind to greenlight a full-time research project.
The protein folding problem asks a straightforward question: Can you predict the physical structure of a protein — its shape in three dimensions?
If scientists can predict a protein’s shape, they can better determine how other molecules will “bind” to it — attach to it, physically — and that is one way drugs are developed. A drug binds to particular proteins in your body and changes their behavior.
In the latest contest, DeepMind made these predictions using “neural networks,” complex mathematical systems that can learn tasks by analyzing vast amounts of data. By analyzing thousands of proteins, a neural network can learn to predict the shape of others.
. . .
Mr. Hassabis said DeepMind was committed to solving the protein folding problem. But many experts said that even if it was solved, more work was needed before doctors and patients benefited in any practical way.
“This is a first step,” said David Baker, the director of the Institute for Protein Design at the University of Washington. “There are so many other steps still to go.”
As they work to better understand the proteins in the body, for instance, scientists must also create new proteins that can serve as drug candidates. Dr. Baker now believes that creating proteins is more important to drug discovery than the “folding” methods being explored, and this task, he said, is not as well suited to DeepMind-style A.I.
DeepMind researchers focus on games and contests because they can show a clear improvement in artificial intelligence. But it is not clear how that approach translates to many tasks.
“Because of the complexity of drug discovery, we need a wide variety of tools,” Dr. Alvarez said. “There is no one-size-fits-all answer.”
For the full story, see:
(Note: ellipses added.)
(Note: the online version of the story has the date Feb. 5, 2019, and has the title “Making New Drugs With a Dose of Artificial Intelligence.”)
Boghossian May Be Punished for Exposing the “Faulty Epistemology” of Grievance Studies
(p. A15) A massive academic hoax has taken a surprising twist. Peter Boghossian, an assistant professor of philosophy, faces disciplinary action at Oregon’s Portland State University. The accusations against him raise constitutional questions about federal regulation of academic research. They also implicitly acknowledge that the prank had a serious point.
Mr. Boghossian–along with two confederates, neither of whom has an academic affiliation–set out to expose shoddy scholarship in what they call “grievance studies.” They concocted 20 pseudonymous “academic papers,” complete with fake data, and submitted them to leading peer-reviewed scholarly journals in fields like “queer studies” and “fat studies.” The Journal’s Jillian Melchior discovered the deception last summer and broke the story in October, by which time seven of the phony papers had been accepted for publication and four published.
“It had to be done,” Mr. Boghossian tells me. “We saw what was happening in these fields, and we were horrified at the faulty epistemology that these people were using to credential themselves and teach others.” The effort drew praise from some well-known public intellectuals, including Richard Dawkins, Jordan Peterson and Steven Pinker.
. . .
A hastily formed university committee recommended that Mr. Boghossian be investigated for “research misconduct”–that is, purposely fabricating data. That case would seem to be open and shut, but the investigation has stalled.
More serious are the sanctions against Mr. Boghossian announced Dec. 21 on behalf of Portland State’s Institutional Review Board for conducting research on “human subjects” without submitting his research protocol to the IRB for review as required by the federal National Research Act of 1974. The “human subjects” in question were the editors and peer-reviewers of the duped journals. Portland State ordered Mr. Boghossian to undergo “human subjects research training,” and its letter warns that “further actions may be required,” with no elaboration.
. . .
Philip Hamburger, a law professor at Columbia, argues that the National Research Act and the HHS’s regulations violate the First Amendment, infringing on scholars’ freedom of expression. Mr. Hamburger has likened IRB vetting procedures to the Star Chamber’s licensing of publications that prevailed in 17th-century England–which the Constitution’s drafters were eager not to replicate. “Licensing . . . prohibits generally, and then selectively permits what otherwise is forbidden,” Mr. Hamburger wrote in 2007.
For the full commentary, see:
Charlotte Allen. “A Hoax and Its ‘Human Subjects’; An Institutional Review Board disciplines an academic prankster. But is it constitutional?” The Wall Street Journal (Tuesday, Jan. 29, 2019): A15.
(Note: ellipses between paragraphs, added; ellipsis internal to last paragraph, in original.)
(Note: the online version of the commentary has the date Jan. 28, 2019.)
Origin of False Memories
(p. A19) Memories are subject to serious flaws, given the limitations and imperfections of the biological and psychological processes of recording, retaining and recalling them. Memories aren’t computer files with exacting recall and retrieval functions. They are often disassembled and stored in “packets” in multiple brain locations. People don’t store the fine details of all daily experiences, because of neuron capacity limitations. Even important details can be missed or lost.
Hence the brain must be selective in which memories it stores and must condense them so that many details are left out. Many eyewitnesses and even victims of crimes don’t take note of the facial features of gun-toting assailants or the make and color of getaway cars.
. . .
My colleague Elizabeth Loftus was able to “implant” false memories in a significant subset of laboratory subjects by showing them an official-looking poster of Disney characters, including Mickey Mouse and Bugs Bunny. Many subjects later remembered meeting Bugs Bunny on a childhood trip to Disneyland. Some of them even reported that Bugs had touched them inappropriately.
That was impossible. Bugs Bunny isn’t a Disney character.
For the full commentary, see:
Richard B. McKenzie. “A Stumble Down Memory Lane; Like Kavanaugh’s latest accuser, people often have ‘gaps.’ They don’t always fill them with truth..” The Wall Street Journal (Tuesday, September 25, 2018): A19.
(Note: ellipsis added.)
(Note: the online version of the commentary has the date Sept. 24, 2018.)
The commentary quoted above is partly based on McKenzie’s book:
McKenzie, Richard B. A Brain-Focused Foundation for Economic Science: A Proposed Reconciliation between Neoclassical and Behavioral Economics. Basingstoke, UK: Palgrave Macmillan, 2018..
Ridiculed Nathan Myhrvold Perseveres on Asteroids and Is Vindicated
Nathan Myhrvold has also been ridiculed on his entrepreneurial patent clearinghouse (called Intellectual Ventures), and on his geoengineering solution to global warming.
(p. D1) Thousands of asteroids are passing through Earth’s neighborhood all the time. Although the odds of a direct hit on the planet any time soon are slim, even a small asteroid the size of a house could explode with as much energy as an atomic bomb.
So scientists at NASA are charged with scanning the skies for such dangerous space rocks. If one were on a collision course with our planet, information about how big it is and what it’s made of would be essential for deflecting it, or calculating the destruction if it hits.
For the last couple of years, Nathan P. Myhrvold, a former chief technologist at Microsoft with a physics doctorate from Princeton, has roiled the small, congenial community of asteroid scientists by saying they know less than they think about these near-Earth objects. He argues that a trove of data from NASA they rely on is flawed and unreliable.
. . .
(p. D4) Dr. Myhrvold’s findings pose a challenge to a proposed NASA asteroid-finding mission called Neocam, short for Near-Earth Object Camera, which would likely cost hundreds of millions of dollars. A congressional committee that controls NASA’s purse strings just included $10 million more in a budget bill for the development of Neocam.
. . .
When Dr. Myhrvold made his initial claims, the Neowise scientists made fun of a few errors like an equation that mixed up radius and diameter.
“It is too bad Myhrvold doesn’t have Google’s bug-finding bounty policy,” Dr. Wright told Scientific American. “If he did, I’d be rich.”
Dr. Mainzer also said at the time, “We believe at this point it’s best to allow the process of peer review — the foundation of the scientific process — to move forward.”
. . .
Earlier this year, Icarus published Dr. Myhrvold’s first paper on how reflected sunlight affects measurements of asteroids at the shorter infrared wavelengths measured by WISE. It has now accepted and posted a second paper last month containing Dr. Myhrvold’s criticisms of the NASA asteroid data.
. . .
When the scientists reported their findings, they did not include the estimates produced by their models, which would have given a sense of how good the model is. Instead they included the earlier measurements.
Other astronomers agreed that the Neowise scientists were not clear about what numbers they were reporting.
“They did some kind of dumb things,” said Alan W. Harris, a retired NASA asteroid expert who was one of the reviewers of Dr. Myhrvold’s second paper.
Dr. Myhrvold has accused the Neowise scientists of going into a NASA archive of planetary results, changing some of the copied numbers and deleting others without giving notice.
“They went back and rewrote history,” he said. “What it shows is even this far in, they’re still lying. They haven’t come clean.”
Dr. Harris said he did not see nefarious behavior by the Neowise scientists, but agreed, “That’s still weird.”
. . .
Dr. Myhrvold said NASA and Congress should put planning for the proposed Neocam spacecraft on hold, because it could suffer from the same shortfalls as Neowise. “Why does it get to avoid further scrutiny and just get money directly from Congress?” he asked.
For the full story, see:
Kenneth Chang. “A Collision Over Asteroids.” The New York Times (Tuesday, June 19, 2018): D1 & D4.
(Note: ellipses added.)
(Note: the online version of the story has the date June 14, 2018, and has the title “Asteroids and Adversaries: Challenging What NASA Knows About Space Rocks.”)
The Diversity That Matters Most Is Diversity of Thought
(p. A15) If you want anyone to pay attention to you in meetings, don’t ever preface your opposition to a proposal by saying: “Just to play devil’s advocate . . .” If you disagree with something, just say it and hold your ground until you’re convinced otherwise. There are many such useful ideas in Charlan Nemeth’s “In Defense of Troublemakers,” her study of dissent in life and the workplace. But if this one alone takes hold, it could transform millions of meetings, doing away with all those mushy, consensus-driven hours wasted by people too scared of disagreement or power to speak truth to gibberish. Not only would better decisions get made, but the process of making them would vastly improve.
. . .
In the latter part of her book, Ms. Nemeth explores in more detail how dissent improves the way in which groups think. She is ruthless toward conventional “brainstorming,” which tends toward the uncritical accumulation of bad ideas rather than the argumentative heat that forges better ideas. It’s only through criticism that concepts receive proper scrutiny. “Repeatedly we find that dissent has value, even when it is wrong, even when we don’t like the dissenter, and even when we are not convinced of his position,” she writes. “Dissent . . . enables us to think more independently” and “also stimulates thought that is open, divergent, flexible, and original.”
. . .
Ms. Nemeth’s punchy book also has an invaluable section on diversity in groups. All too often, she writes, in pursuit of diversity we focus on everything but the way people think. We look at a group’s gender, color or experience, and once the palette looks right declare it diverse. But you can have all of that and still have a group that thinks the same and reinforces a wrong-headed consensus.
By contrast, you can have a group that is demographically homogeneous yet violently heterogeneous in the way it thinks. The kind of diversity that leads to well-informed decisions is not necessarily the kind of diversity that gives the appearance of social justice. That will be a hard message for many organizations to swallow. But as with many of the arguments that Ms. Nemeth makes in her book, it is one that she gamely delivers and that all managers interested in the quality and integrity of their decision-making would do well to heed.
For the full review, see:
Philip Delves Broughton. “BOOKSHELF; Rocking The Boat.” The Wall Street Journal (Thursday, May 9, 2018): A15.
(Note: ellipsis internal to a paragraph, in original; ellipses between paragraphs, added.)
(Note: the online version of the review has the date May 10, 2018, and has the title “BOOKSHELF; ‘In Defense of Troublemakers’ Review: Rocking the Boat.”)
The book under review, is:
Nemeth, Charlan. In Defense of Troublemakers: The Power of Dissent in Life and Business. New York: Basic Books, 2018.
Labor-Intensive Tinkering Can Advance Science
(p. A24) When John E. Sulston was 5 years old and growing up in Britain, the son of an Anglican priest, his parents sent him to a private school. There, he discovered, sports were his nemesis.
“I absolutely loathed games,” he said. “I was hopeless.”
When it came to schoolwork, he said, he was “not a books person.”
He had only one consuming interest: science. He liked to tinker, to figure out how things were put together.
. . .
The Nobel he received, shared with two other scientists, recognized the good data he amassed in his work on the tiny transparent roundworm C. elegans in an effort to better understand how organisms develop.
. . .
At the time, it was widely believed that the 558 cells the worm had when it hatched were all it would ever have. But Dr. Sulston noticed that, in fact, the worm kept gaining cells as it developed. And by tracing the patterns of divisions that gave rise to those new cells, he found, surprisingly, that the worm also lost cells in a predictable way. Certain cells were destined to die at a specific time, digesting their own DNA.
Dr. Sulston’s next major project was to trace the fate of every single cell in a worm. It was a task so demanding and labor-intensive that other scientists still shake their heads in amazement that he got it done.
Each day, bending over his microscope for eight or more hours, he would start with a worm embryo and choose one of its cells. He would then watch the cell as it divided and follow each of its progeny cells as, together, they grew and formed the organism. This went on for a total of 18 months.
In the end, he had a complete map of every one of the worm’s 959 cells (not counting sperm and egg cells).
For the full obituary, see:
GINA KOLATA. “John Sulston, 75; Tiny Worm Guided Him to Nobel.” The New York Times (Friday, March 16, 2018): A24.
(Note: ellipses added.)
(Note: the online version of the obituary has the date MARCH 15, 2018, and has the title “John E. Sulston, 75, Dies; Found Clues to Genes in a Worm.”)