Sunday, March 04, 2018

Thoughts on Ericsson's Peak: Good self-help advice (practice deliberately!), framed in terms of brain science

We read Anders Ericsson’s Peak: Secrets from the New Science of Expertise with great interest. On one hand, this is a practical self-help book that provides concrete ways to get better at things. On the other hand, it burnishes a lot of these suggestions with good scientific vignettes. In terms of the useful advice the book is all about the importance of practice -- but not simply the mindless, repetitive practice implied by the famous 10,000 hour rule -- but rather what the author calls, ‘deliberate practice’ where you push yourself to the limit of your ability, get feedback and attempt to form better mental representations of the task at hand. This method applies to everything, from memorizing chess positions to swinging a golf club.

The book provides lots of good case stories on this, from the famous digit-memorizer Steve Faloon to studies on how radiologists with deliberate practice and feedback did much better identifying tumors than those without. It also provides some scientific grounding to the importance of deliberate practice, such as the studies on brain volume and London taxi cab drivers and how those who practice more literally had a physiological change and also how implementing a different time of teaching physics demonstrably improved student performance.

This work dovetails with the idea of self-help through coaching, recently championed by Atul Gwande and others [], and more generally the goal of finding the best way to improve oneself. We think the challenge is two-fold – first, objectively monitoring your performance so that you can best identify which of your weaknesses to target; and second, developing a deliberate practice in which you can receive the feedback needed to improve. Ericsson describes Ben Franklin teaching himself to frame an argument by writing each sentence of a position piece on a separate slip of paper, and then trying to correctly arrange the slips. What is the analog for becoming a better scientist? Can you do it yourself, or do you need a coach?

Altogether, we found it a good read and it will definitely have some pointers that we will keep in mind and continue to practice.

Review written with Daniel Spakowicz (
Peak: Secrets from the New Science of Expertise
by Anders Ericsson


Saturday, March 03, 2018

Meet Your Art Twin - My experience with GoogleArts selfie matching app

Meet Your Art Twin: #GoogleArts new "#selfie" matching app claims I look a bit like portraits of Jan van de Poll, Frank Jay St. John & Harold Ickes. Not sure I agree, but I find it quite amusing nonetheless!

* More detail on the matching images

- 67% match
Jan van de Poll (1597-1678). Burgemeester van Amsterdam Rijksmuseum

- 66% match
Frank Jay St. John - Thomas Eakins — Google Arts & Culture

- 60% match
Harold L. Ickes 1874-1952

* NY Times article

Meet Your Art Twin: A 400-Year-Old With an Oily Complexion

“Long before the Google Arts and Culture app, which became the most downloaded mobile app over the weekend, art aficionados, dabblers, narcissists and soul searchers pondering a cosmic connection to distant humans have been searching for their art twins, a long-gone, sometimes fictional or unknown doppelgänger encased in oil, sculpture or ceramics.”

* Some of the matching images

- GoogleArts Selfie Matches to MBG

- More GoogleArts stuff (private link)

- Mail subject for lots of the images
Does this artwork look like me? Try with your own #selfie at #GoogleArts

- Match Photo (536x916)

Thursday, November 09, 2017

Letter to the Editor RE "Reading This While Walking? It Could Cost You" -- More thoughts on bike & pedestrian safety

I enjoyed the recent articles in the Tuesday, October 24 issue of the
Times that dealt with street safety. On the one hand, they seemed to
admonish both pedestrians and cyclists to be safer: cyclists, by
wearing helmets to protect themselves and pedestrians, by not using
smartphones while crossing the street. While these are reasonable,
they detract from the real villain: cars and trucks. Fundamentally,
streets should be made safer for pedestrians and bicyclists, by
enforcing driving regulations (and also by protecting pedestrians from
rule-breaking bikers.) Obviously, it's not an equal interaction
between a pedestrian and a car, and I think the laws should reflect
this fundamental, asymmetry.
Buckle Up a Helmet to Save a Life
Personal Health
OCT. 23, 2017
Reading This While Walking? In Honolulu, It Could Cost You
OCT. 23, 2017

Saturday, September 16, 2017

Thoughts on Stewart's Calculating the Cosmos: Astronomy as a motivator for math (eg mechanics, chaos, relativity)

I found Ian Stewart’s Calculating the Cosmos to be an interesting work that gives one a sense of how mathematics has been used in astronomy over the ages. The book provides a fairly traditional survey of all the astronomical elements -- the planets, the sun, asteroids, and comets -- and then continues to describe galaxies and the creation of the universe. The book highlights how key bits of mathematics are important for understanding these elements; for instance, angular momentum is crucial in creating disc- and spiral-shaped planetary systems and galaxies, and many-body dynamics is essential to understanding the motion of planets and how Lagrangian points around them confer special status.

The book explains why a lot of important mathematical concepts such as Gaussian curvature and Riemannian manifolds manifest themselves in general relativity. A lot of the things we take for granted just do not hold true in the world of black holes. I particularly liked the description of how chaos arises from small perturbations in initial conditions and how this manifests itself in terms of the overall motions of planetary bodies and the dynamics of systems. Furthermore, it is notable how astronomers used the deviation from what they expected as a way of finding Neptune and eventually Pluto, although the latter in a sense was a mistaken find.

Overall, this was an interesting read and put a lot of mathematics into perspective. I enjoyed it very much and would highly recommend it.
Calculating the Cosmos: How Mathematics Unveils the Universe
by Ian Stewart (Author)


Tuesday, September 05, 2017

Letters commenting on the safety of bikers (& drivers & pedestrians), given the recent increase in urban cycling

Letter #1

I read the recent Times article on the upswing in cycle commuting with
great interest. Certainly, from the perspective of a technocratic
transportation planner, this is a great thing. However, it cannot
relieve the terror one feels at night when a lightless bike zooms by,
narrowingly missing a collision, or when one has to wrench the
steering wheel and swerve almost into another car, to avoid a cyclist
going the wrong way down a street. There must be a way to fix this.
Perhaps give licenses to cycle commuters? Or maybe technology can
help: Trackers on bicycles or cameras monitoring sidewalks to flag bad

Letter #2

I read with great interest the recent Times Op. Ed. about a
65-year-old pedestrian killed by a cyclist. I sympathize as I am often
terrified walking down sidewalks by careening cycles, coming out of
nowhere. Also, there's the scary situation of the driver who swerves,
nearly getting into an accident, to avoid a cyclist going the wrong
way. A fundamental issue underlying cyclists' unsafe behavior is that
it's often ambiguous -- from their perspective -- whether they are
motorists or pedestrians. It's often better for the cyclist to ride on
safer sidewalks than on more perilous roads or to ride through a red
light to keep momentum. What can be done? A cyclist needs to be given
clear rules, and they need to be enforced. If a bike becomes a loose
canon, perhaps it is possible to take it away before damage is done.

Unpublished letter in response to:

Wheels of Misfortune -

More New Yorkers Opting for Life in the Bike Lane - The New York Times

Sunday, August 27, 2017

Thoughts on Dataism by Steve Lohr - Vignettes on Data Science, entertaining specifics but are they representative?

I read Steve Lohr's book Data-ism with keen interest. The author is a noted reporter for The Times that writes on technology. I was looking forward to the book and found it an interesting read. However, it's a bit patchy in certain places and tends to focus specifically on one point or illustration. I'm not sure it gives a balanced view of the emerging field of data science. That said, I particularly liked the way it talked about how data science is revolutionizing farming -- how the putting together of many sensors enables one to get an overview of the field much more so than one could get normally.

Data-ism: The Revolution Transforming Decision Making, Consumer
Behavior, and Almost Everything Else
by Steve Lohr (Author)


Sunday, August 20, 2017

Thoughts on Levitin's Weaponized Lies & A Field Guide to Lies - Good reads on how statistics can deceive

With great interest, I read Daniel Levitin's recent books "Weaponized Lies: How to Think Critically in the Post-Truth Era" and "A Field Guide to Lies: Critical Thinking in the Information Age." Both of these books give a good primer to deal with statistics, going over such things in detail such as Bayesian analysis in terms of a 2x2 table. Levitin also talks a bit about how people sort the truth by simply making false graphs with unlabeled axes or how there are more subtle aspects such as subtly changing the denominator (eg when talking about individuals versus families). Altogether I found these extremely useful reads. One thing that was a little troublesome to me was that they were fairly similar books. I think it does make sense for the author to repurpose his text into more current times where we talk about fake news and so forth but I still found the overlap between the two books disconcerting.
A Field Guide to Lies: Critical Thinking in the Information Age Audio CD – Audiobook, CD, Unabridged
by Daniel J. Levitin
Weaponized Lies: How to Think Critically in the Post-Truth Era Reprint Edition
by Daniel J. Levitin (Author)