Archive for January 25, 2016

Monday, January 25, 2016

Marvin Minsky, RIP

Glenn Rifkin (via Dan Bricklin, Hacker News, Slashdot):

Well before the advent of the microprocessor and the supercomputer, Professor Minsky, a revered computer science educator at M.I.T., laid the foundation for the field of artificial intelligence by demonstrating the possibilities of imparting common-sense reasoning to computers.

“Marvin was one of the very few people in computing whose visions and perspectives liberated the computer from being a glorified adding machine to start to realize its destiny as one of the most powerful amplifiers for human endeavors in history,” said Alan Kay, a computer scientist and a friend and colleague of Professor Minsky’s.


Professor Minsky’s scientific accomplishments spanned a variety of disciplines. He designed and built some of the first visual scanners and mechanical hands with tactile sensors, advances that influenced modern robotics. In 1951 he built the first randomly wired neural network learning machine, which he called Snarc. And in 1956, while at Harvard, he invented and built the first confocal scanning microscope, an optical instrument with superior resolution and image quality still in wide use in the biological sciences.

Philip Greenspun:

Marvin questioned many of the assumptions around Academia. He would show up to deliver a formal talk with a stack of notes and pick from them more or less at random. I wish that I could say that the results were amazing due to his dazzling intellect, but unfortunately the lack of an organized outline made these talks less than satisfying. Marvin was at his best in small groups or working one-on-one with others at MIT. What I remember most about him was his genial skepticism. If two people were arguing, rather than take one side, Marvin could show that both were operating from an assumption about the world that perhaps wasn’t necessary or true. He was a bit of a modern-day Socrates.

Update (2016-01-28): Paul Whitmore Sas:

Another defining characteristic of his mind was its utterly stochastic trajectory. No one, and certainly not Marvin, could discipline his curiosity cum hubris. SoM class lectures rode roughshod over a thousand topics, and left many stacks unpopped. The book, which exemplifies a horde of daemons tackling random splinters of cognition, was rumored to have been assembled from shards collected together with Hillis’ help.

Update (2016-01-29): Adam C. Engst:

I’ll remember him as a TidBITS reader and for his openness and kindness. Although plans to meet up at Macworld Boston never came to fruition, I exchanged email with him back in 1992 about Apple’s Casper voice recognition technology (see “Casper Speaks,” 9 March 1992), which evolved into a system called PlainTalk (see “Bossing Your Mac with PlainTalk,” 28 August 2000).

Stephen Wolfram (via Hacker News):

Being no expert on such things (and without the web to look anything up!), I called Marvin to ask what to do. What ensued was a long discussion about the possibility of developing microrobots that could chase mealybugs away. Fascinating though it was, at the end of it I still had to ask, “But what should I actually do about Margaret’s plants?” Marvin replied, “Oh, I guess you’d better talk to my wife.”


Marvin immediately launched into talking about how programming languages are the only ones that people are expected to learn to write before they can read. He said he’d been trying to convince Seymour Papert that the best way to teach programming was to start by showing people good code.

Update (2016-01-30): Scott Aaronson:

For example, in his and Papert’s 1968 book Perceptrons—notorious for “killing neural net research for a decade,” because of its mis- or over-interpreted theorems about the representational limitations of single-layer neural nets—the way Minsky and Papert proved those theorems was by translating questions about computation into questions about the existence or nonexistence of low-degree polynomials with various properties, and then answering the latter questions using MATH. Their “polynomial method” is now a mainstay of quantum algorithms research (having been brought to the subject by Beals et al.), and in particular, has been a mainstay of my own career.

Update (2016-03-17): Steven Levy:

Minsky quotes in fortune cookies.

7 Scandalous Weird Old Things About the C Preprocessor

Robert Elder (via Peter Steinberger):

Bjarne Stroustrup even points out that the standard isn’t clear about what should happen in function macro recursion[1]. With reference to a specific example he says “The question is whether the use of NIL in the last line of this sequence qualifies for non-replacement under the cited text. If it does, the result will be NIL(42). If it does not, the result will be simply 42.”. In 2004, a decision was made to leave the standard in its ambiguous state: “The committee’s decision was that no realistic programs “in the wild” would venture into this area, and trying to reduce the uncertainties is not worth the risk of changing conformance status of implementations or programs.”


Many languages are first tokenized, and then the list of tokens doesn’t change throughout further processing of the program. In the C preprocessor, new tokens can be created at run time! This makes it impossible to build a parse tree ahead of time because you don’t know what tokens would be included in the final tree.


In general, when evaluating a function macro body, you need to consider both the pre-expanded version of the arguments, and the untouched tokens that were passed for that argument. This behaviour is unlike how C arguments and functions are evaluated, because in C you can always replace an argument that’s described by an expression with the result of that expression and have the same meaning (ignoring any side effects).

He also mentions handling of whitespace, which is confusing and also buggy in Clang.

Amazon’s Customer Service Backdoor

Eric Springer (Hacker News, Slashdot):

The attacker gave Amazon my fake details from a whois query, and got my real address and phone number in exchange. Now they had enough to bounce around a few services, even convincing my bank to issue them a new copy of my Credit Card.


I again contact Amazon to reiterate how important it is that they keep my account secure, and not give out my details to anyone with a name and address. They promise they’re putting a note on my account, and it’ll never happen again. And I will be contacted by a specialist (never happened, again)

Anywhere But Medium

Dave Winer:

Medium is on its way to becoming the consensus platform for writing on the web. if you’re not sure you’re going to be blogging regularly, the default place to put your writing is Medium, rather than starting a blog on Tumblr or, for example. I guess the thought is that it’s wasteful to start a blog if you’re not sure you’re going to post that often.


People also post to Medium to get more flow. But at what cost? Which pieces get flow? Ones that are critical of Medium? I doubt it. Or offend the politics of the founder? I don't know.


I also want to preserve the ability of developers to innovate in this area. If Medium sews up this media type, if they own it for all practical purposes, as Google owned RSS (until they dropped it), then you can’t move until they do. And companies with monopolies have no incentive to move forward, and therefore rarely do.

John Gruber:

The comparison to Google Reader is perfect. Google Reader was both (a) the most popular thing that ever happened to RSS, and (b) the worst thing that ever happened to RSS.

The Story Behind F.lux

Matthew Braga (via Hacker News):

“We really thought six months after that, that we were done. We had done this app that was pretty baked. And we didn’t really think it was that hard,” Michael recalled. “Now we have this joke that, everyone thinks this problem is easy until they spend a year or two on it, and then they think it’s the hardest problem they’ve ever worked on.”

Developing f.lux, it turned out, was more complex than either Lorna or Michael initially thought. And in the seven years since they started working on the app full time, Michael and Lorna have realized what they’re really on is a mission to change the way that we sleep.


Michael and Lorna are quick to point out that that brightness also plays a role, as does distance from the source of light. The pair even operate a companion site called f.luxometer that measures the light intensity of various popular devices, and how this factors into your phase shift—the amount of time, say, a MacBook Air might shift your circadian timing if used before bed.

I think I’ve determined that all my Lightroom hangs occur when Flux re-enables itself after being suspended for an hour.