Datasplaining Podcast: Recent Episodes
Amar Natt, Alex Rinaudo, Steve Lee, Chase Covello
Where law, technology, economics, and data intersect with data. When you throw together an economist, an attorney/computer scientist, a decision scientist, and a securities litigation expert, what you end up with is a lively debate about how to correctly use the data available. Datasplaining, if you will. Together we explore questions about law, ethics, technology, and, of course, data from the perspectives of our specialties to try and figure out how it all comes together.
Data cleaning is the largest part of any data analysis project: we discuss our thoughts on all the decisions we make about data and the implications of those decisions.
What are the filtering and manipulation decisions we make when cleaning data that have larger effects on final datasets? Who handles that data cleaning and how much latitude do they have to make decisions? How does a manager understand what was done to the data without reading the code? Beyond that, how do the huge masses of data we deal with change the landscape for modeling and finding signals in the noise?
Download MP3 | Subscribe: RSS
Links:
Fired by Bot at Amazon: ‘It’s You Against the Machine’ [Bloomberg]
We discuss our thoughts on why bad science and social science are propagated.
Our meandering discussion centers on the human behavior component of academic research and touches on deliberately misleading research vs accidental mistakes, why it’s so hard to catch errors in academic work, the nature of the scientific method, and the human need for factual information that makes science communication with the public such a challenge.
Download MP3 | Subscribe: RSS
Links:
For research citations, no replication is no problem [Ars Technica]
Why Do Scientists Lie? [Liam Kofi Bright]
The 60-Year-Old Scientific Screwup That Helped Covid Kill [Wired]
We discuss our thoughts on the Oracle v Google decision on the use of APIs.
We wander through such thrilling topics as what APIs are, how you can tell programming students are cheating, how to protect intellectual property in software, and what would have happened if the decision had gone the other way?
Download MP3 | Subscribe: RSS
Links:
How the Supreme Court saved the software industry from API copyrights [Ars Technica]
We discuss what AI actually is and how it relates to Machine Learning, how to effectively regulate emerging technology, and why we should care about regulation for AI and Machine Learning.
Download MP3 | Subscribe: RSS
Links:
Regulation of Algorithmic Tools in the United States [SSRN]
We discuss the power of app stores to control information and whether that warrants regulation, the data apps collect and privacy issues surrounding that data, how that data is used in applications that couldn’t have been foreseen by users, and bias in the designs of algorithms.
Download MP3 | Subscribe: RSS
Links:
New York City proposes regulating algorithms used in hiring [Ars Technica]
Job Screening Service Halts Facial Analysis of Applicants [Wired]