What I'm reading 4/18-4/25


Against Blacklists - Andrea Shepard

  1. "Instead, we must fight for the establishment of norms which allow peaceful coexistence and free association – and thus, against the authoritarian, universalizing tendencies within every cultural-ideological tribe – and for the possibility of pluralistic technical spaces with norms that protect the purpose they serve for technologists regardless of tribal affiliation, and resist being transformed into a no man's land in between red and blue tribe trenches."

System Overload - James Surowiecki

  1. "This is the heart of our problem: infrastructure policy has become a matter of lurching from crisis to crisis, solving problems after the fact rather than preventing them from happening."

  2. "Today, the Metro is in such a state that fixing it may require shutting whole lines for months at a time. It's yet again an example for the nation, but now it's an example of how underinvestment and political dysfunction have left America with infrastructure that's failing and often downright dangerous."

  3. "As the economist Larry Summers has pointed out, once you adjust for depreciation, the U.S. makes no net investment in public infrastructure."

False Plagiarism Accusation Against Shaun King Shows Dangers of Online Mob Journalism - Glenn Greenwald

  1. "They evidently had no compunction about publicly accusing someone of a reputation-destroying transgression without first contacting the accused and, worse, without making any effort to gather the elementary facts."

A Protocol for Dying - Pieter Hintjens

  1. "In time it became the basis, and then the goal of my work: to go to strange places and meet new people."

  2. "I am so grateful not to have died suddenly. I'm so grateful I won't lose my mind."

What convolutional neural networks look at when they see nudity - Ryan Compton

  1. "However, since deep learning researchers don't specify exactly how the network should behave on a given input, a new problem arises: How can one understand what the convolutional networks are activating on?"

  2. "This highlights a fundamental difference between training rule-based classifiers and modern A.I. research. Rather than redesigning features by hand, we redesign our training data until the discovered features are improved."

Introduction to Scala Macros - Sturm Mabie

  1. "In Scala, as in Lisp, macros are not merely transformations that operate on a set of strings, but rather, are functions that map from a given AST to another AST."

Implementers, Solvers, and Finders - Randall Koutnik

  1. "How could this be? A group of people, granted the ability to do what they love for great pay & perks, all wanting to move on?"

  2. "If we're to escape this situation where fantastic fanatic programmers can't see themselves programming in three years, we need to couple our words to real world meaning."