Allen Downey
January 2020
Overview of my professional activities in 2019.
I am continuing work on an undergraduate curriculum in Data Science for use at Olin and other schools.
In September I was appointed as a Visiting Professor of Computer Science at Harvard University, where I was a co-instructor for a graduate data science seminar and collaborated on developing a new data science class which will be offered in Spring 2020.
One of my contributions is a new online data science curriculum, called "Elements of Data Science", and two case studies on:
In the spring I taught Data Science at Olin. As a class project, some students participated in the ASA DataFest. One of the more successful projects looked at the effect of the Opioid Epidemic on life expectancy in the U.S.; I published a summary of the class projects in my blog.
In April I launched an online course with DataCamp, "Exploratory Data Analysis in Python". At this point, almost as many people have taken this course online as I have taught, in person, in my academic career. Since July I have been a member of DataCamp's Instructor Advisory Board.
In July I presented a talk at SciPy, "Generational Changes in Support for Gun Laws: A Case Study in Computational Statistics". and made a presentation for the teen track. I served on the SciPy tutorial selection committee.
In September I completed instructor training with The Carpentries.
In September I attended the New England Isolated Statisticians Meeting (NEISM) at Wellesley College.
In November I presented "The Inspection Paradox is Everywhere" at the PyData NYC Conference and the PyData Boston Meetup
In 2019 I published several blog articles related to Data Science, including:
- A series of articles on political alignment in the U.S.
- An article on characterizing the tail behavior of distributions
- An update of my previous reports on marriage patterns in the U.S.
These articles were viewed more than 30,000 times. In an experiment to reach a broader audience, I wrote several articles on Medium, two of which were picked up by the online publication Towards Data Science:
These articles were viewed more than 35,000 times.
Related to my work on Bayesian statistics for undergrads, in February I presented "The Bayesian Zig Zag: Developing Probabilistic Models Using Grid Methods and MCMC" as a TechTalk for the ACM Learning Center.
At SciPy in July, I led a half-day tutorial called Bayesian Statistics Made Easy.
In Spring 2020 I will teach Data Science at Olin in parallel with DS10 at Harvard, with both classes using the new curriculum my collaborators and I developed.
In January, I collaborated with Jason Moore from UC Davis to lead a workshop, "Eight Ways to Use Computation to Teach Everything Else", at the KEEN National Conference.
In March I presented a version of this workshop at Harvard University. I also taught this workshop in a longer format at Olin's Summer Institute 2019 in June.
During the summer I revised Modeling and Simulation in Python and the accompanying software, which were used in the fall offering of Modeling and Simulation at Olin.
In July I taught a half-day tutorial on Complexity Science at SciPy 2019.
In the fall semester I taught Complexity Science at Olin. Some of the more successful projects are here, here, and here.
I am continuing to revise Physical Modeling in MATLAB, to be published by No Starch Press in 2020. I am working with MathWorks toward developing online materials to support the book.
All of my books are available under free licenses that allow readers to copy, modify, and redistribute them. These licenses make possible many translations, adaptations, and derivative works, and lead to collaborations with co-authors all over the world.
Think Julia, a collaboration with Ben Lauwens at the Royal Military Academy in Belgium, was published by O'Reilly Media in April 2019.
Think Java, 2nd edition, a collaboration with Chris Mayfield at James Madison University, was published by O'Reilly Media in December 2019. A version of Think Java has been adapted as an interactive online book by Trinket.
I continue to develop Think OS and The Little Book of Semaphores, which I used in the Spring 2019 offering of Software Systems.
Several of my books have been licensed by Codio and are in development as interactive online books.
In 2019, new translations of my books have been published in Russian, traditional Chinese, and simplified Chinese. Other books have been licensed for translation into Polish, Greek, and Korean.
My course material, slides from my presentations, and material from my workshops are published on the web. And the supporting code for my classes and books are on GitHub, where anyone can copy, modify, and make corrections and contributions.
Because my books are freely available, they are used in classes all over the world. Since my last report, I have heard from people using them at Northeastern University, Indiana University, San Diego State University, University of Iowa, University of Oregon, San Jose State University, Cal State University Monterey Bay, Loyola University - Maryland, University of New Haven, and Black Hills State University; Lancing College, Dawson College, Union College, Santa Monica College, Principia College, and Berklee College of Music in Boston; NIIT University, Universidade Federal de Alagoas, and Simon Bolivar University in Venezuela; Kearsarge Regional High School, Herricks High School, Atkins School District, and Ball High School in Texas.
In all of my projects, I “release early and release often”, getting feedback from readers and inviting collaboration. Following the principles of Open Science, I publish data and code along with research results, allowing others to reproduce and modify my analysis. And I don't publish in venues that do not make papers and supporting materials freely available.