Theory of Everything recently posted an addendum to last year’s Enchanting by Numbers. Both episodes are very worthwhile, and both include the same segment talking about how misunderstood Facebook algorithms are to most of its users.
Be sure to listen to the interview with Suw Charman-Anderson, founder of Ada Lovelace Day, which is today! That part starts at 14:30 in part 1.
Margaret Hamilton oversaw the guidance software on the Apollo program. Thanks to sophisticated error-handling in that code, her engineering efforts prevented an abort of the Apollo 11 landing.
Three minutes before the Lunar lander reached the Moon’s surface, several computer alarms were triggered. The computer was overloaded with incoming data, because the rendezvous radar system (not necessary for landing) updated an involuntary counter in the computer, which stole cycles from the computer. Due to its robust architecture, the computer was able to keep running; the Apollo onboard flight software was developed using an asynchronous executive so that higher priority jobs (important for landing) could interrupt lower priority jobs.
This was on a 2 MHz machine with 1 MB of RAM. Even the term software engineering was coined by Hamilton, who also helped develop “concepts of asynchronous software, priority scheduling, end-to-end testing, and human-in-the-loop decision capability.”
A safe mantra to keep in mind with software is “all software has bugs,” which so often means “don’t expect too much.” Margaret Hamilton was instrumental in creating processes to ensure that software can systematically accommodate surprises and continue functioning as expected.
A fan of the space program, Kipp Teague, has uploaded a huge trove of Apollo mission photo scans onto Flickr. They’re organized into albums, but rather overwhelming as a collection, unless you want to just page through thousands of space photos (which, I mean, yeah why not?).
Digg has posted a smaller “best of” selection if you want to see just the highlights.
Does your lifestyle prevent you from qualifying for insurance discounts?
Do you lack sufficient time for exercise or have limited access to sports facilities?
Maybe you just want to keep your personal data private without having to pay higher insurance premiums for the privilege?
Unfit Bits provides solutions. At Unfit Bits, we are investigating DIY fitness spoofing techniques to allow you to create walking datasets without actually having to share your personal data. These techniques help produce personal data to qualify you for insurance rewards even if you can’t afford a high exercise lifestyle.
Photogrammar is an impressive database from Yale “organizing, searching, and visualizing the 170,000 photographs from 1935 to 1945 created by the United States Farm Security Administration and Office of War Information (FSA-OWI).”
No known dialect of Ant employs any verbal person except the third person singular and plural and the first person plural. In this text, only the root forms of the verbs are used; so there is no way to decide whether the passage was intended to be an autobiography or a manifesto.
Long are the tunnels. Longer is the untunneled. No tunnel reaches the end of the untunneled. The untunneled goes on farther than we can go in ten days [i.e., forever]. Praise!
The mark translated “Praise!” is half of the customary salutation “Praise the Queen!” or “Long live the Queen!” or “Huzza for the Queen!”—but the word/mark signifying “Queen” has been omitted.
NETWORKS + NEW TOWNS is an extended site study of Jonathan, Minnesota and related areas. The suburban neighborhood of Jonathan was one of the first “totally planned communities” in the Midwest, born during the short-lived “New Town” movement of the late 1960’s. It grew up during an era characterized by great faith in the power of urban planning and the transformative potential of communications technology. This work uses Jonathan as a microcosm to understand the ways that we augment the earth with matter and data in an ongoing pursuit of better living.
Mat Honan wrote about the experience of riding in Google’s cute self-driving cars.
The first time I rode in a fully autonomous car, what really impressed me was when the car saw something that I could not. As I rode down a residential street in Mountain View, the car slowed, for no apparent reason. Yet in the front seat, a laptop showed everything the car could “see.” And up ahead, there was a man, in the street, standing behind a double-parked vehicle. He was concealed from my eyes, but the car detected him. And it slowed down, anticipating that he might step out unexpectedly.
It anticipated this because each and every one of Google robot cars has experienced the totality of everything all its siblings have experienced. Google’s cars have driven a total of 1.2 million miles on the roads. We tend to think of this as combined experience — an aggregate number. But what it really means, effectively, is that every single car has driven that distance, has experienced it. This is a machine that learns. And in addition to that on road time, the cars log, Google said yesterday, 3 million miles every day running scenarios.
This car is a better driver than me, or you, or any of us.
From Michael Luo’s 2011 series on guns and public safety, this gets at the heart of the political limitations on doing research into the health risks of guns in America.
C.D.C. financing for research on gun violence has not stopped completely, but it is now mostly limited to work in which firearms are only a component.
The centers also ask researchers it finances to give it a heads-up anytime they are publishing studies that have anything to do with firearms. The agency, in turn, relays this information to the N.R.A. as a courtesy, said Thomas Skinner, a spokesman for the centers.
Invariably, researchers said, whenever their work touches upon firearms, the C.D.C. becomes squeamish. In the end, they said, it is often simply easier to avoid the topic if they want to continue to be in the agency’s good graces.
Emphasis added. I’m curious if these circumstances have changed much in the past 4 years.