A forum of 'code golfers'—programmers who create and solve creative programming puzzles for kicks—are using code to break classic art works down to their base pixels, then rebuild them to look like other classic artworks. The resulting images show us, for example, what Grant Wood's American Gothic would look like if painted in the bright oranges and blacks Edvard Munch used in The Scream, or what the Mona Lisa would look like if DaVinci had run out of all his paints, save for the colors in Van Gogh's Starry Night. Take, for example, Mona Lisa morphing into the Wood's classic.
This story is a bit old, but it was orphaned in one of my browser tabs. This is some grade-A sci-fi hocus pocus:
Researchers at MIT, Microsoft, and Adobe have developed an algorithm that can reconstruct an audio signal by analyzing minute vibrations of objects depicted in video. In one set of experiments, they were able to recover intelligible speech from the vibrations of a potato-chip bag photographed from 15 feet away through soundproof glass.
In other experiments, they extracted useful audio signals from videos of aluminum foil, the surface of a glass of water, and even the leaves of a potted plant.
In the experiments reported in the Siggraph paper, the researchers also measured the mechanical properties of the objects they were filming and determined that the motions they were measuring were about a tenth of micrometer. That corresponds to five thousandths of a pixel in a close-up image, but from the change of a single pixel’s color value over time, it’s possible to infer motions smaller than a pixel.
Suppose, for instance, that an image has a clear boundary between two regions: Everything on one side of the boundary is blue; everything on the other is red. But at the boundary itself, the camera’s sensor receives both red and blue light, so it averages them out to produce purple. If, over successive frames of video, the blue region encroaches into the red region — even less than the width of a pixel — the purple will grow slightly bluer. That color shift contains information about the degree of encroachment.
In recent years, researchers have developed methods for detecting heart rate of people purely through video, using small fluctuations in the color of their skin.
What will happen when we are naked to computer vision? What if we can no longer hide when our heart starts racing, our skin flushes, or our hand quivers ever so subtly? We always thought we'd be the one administering the Voight-Kampff test to cull the replicants from the humans, but maybe it's the reverse that arrives first. Machines just sitting their motionless, staring at us, and seeing everything.
You needn’t be a linguist to note changes in the language of menus, but Stanford’s Dan Jurafsky has written a book doing just that. In The Language of Food: A Linguist Reads the Menu, Jurafsky describes how he and some colleagues analyzed a database of 6,500 restaurant menus describing 650,000 dishes from across the U.S. Among their findings: fancy restaurants, not surprisingly, use fancier—and longer—words than cheaper restaurants do (think accompaniments and decaffeinated coffee, not sides and decaf). Jurafsky writes that “every increase of one letter in the average length of words describing a dish is associated with an increase of 69 cents in the price of that dish.” Compared with inexpensive restaurants, the expensive ones are “three times less likely to talk about the diner’s choice” (your way, etc.) and “seven times more likely to talk about the chef’s choice.”
Lower-priced restaurants, meanwhile, rely on “linguistic fillers”: subjective words like delicious, flaky, and fluffy. These are the empty calories of menus, less indicative of flavor than of low prices. Cheaper establishments also use terms like ripe and fresh, which Jurafsky calls “status anxiety” words. Thomas Keller’s Per Se, after all, would never use fresh—that much is taken for granted—but Subway would. Per Se does, however, engage in the trendy habit of adding provenance to descriptions of ingredients (Island Creek oysters, Frog Hollow’s peaches). According to Jurafsky, very expensive restaurants “mention the origins of the food more than 15 times as often as inexpensive restaurants.”
To charge what fine dining establishments charge, it's not enough to source the best ingredients, even though anyone who's peeked at the ledger for a high end restaurant knows the ingredients are a massive cost. Diners expect an experience now, and using more ornate language to describe a dish and the provenance of its ingredients is just as important as beautiful plating and presentation in leaving diners satisfied after having left them several pounds heavier but several hundred dollars lighter.
On the commentary track to Stroszek, you talk about the places where the American dream meets the American nightmare — Disneyland, Wall Street, Las Vegas, San Quentin, and, oddly, Plainfield, Wisconsin. Can you explain?
There are threads coming from all directions and sometimes they build knots. [It’s like] when you look at a map in the airline brochure in front of you and you see the pattern of their flights, and all of the sudden they have a hub in Atlanta and they have another hub in Salt Lake City, and all the lines converge there as if there were knots. There are certain places [like that] in probably each country; in the United States, I feel these focal points, these knots, where everything seems to converge, including the nightmares. Like San Quentin. One of the knots would be Wall Street — not that I’m saying Wall Street is evil. Or Plainfield, Wisconsin, where out of 480 inhabitants five became mass murderers within a few years, including the worst of all, Ed Gein.
Can you feel the energy being transmitted from these places?
You can sense it. Go to Wall Street. At the time I visited Wall Street, it was not a visitor’s gallery behind bulletproof glass. It was open. You heard the shouting from down there. You could even shout down to them — which I didn’t do. Yes, there’s something of great intensity there. Or the intensity of Las Vegas. A cheap, collective dream to strike it rich without working. And I’m one of the few reading and thinking people who loves Las Vegas for the vulgarity and omnipresence of the dream. The collective dream. There’s something enormous about it. Let me say one thing: Las Vegas and cinema have similar roots. The country fair. The magician at the country fair. The vulgarity of the country fair.
The circus! Yes. And everybody complains about the circus of Cannes Film Festival. Yes, it is a circus, and I like it.
From a fabulous interview of Werner Herzog in Vulture.
He would also like people to know he really was clean when he came out of retirement for the 2009 Tour de France. He cleaned up along with everyone else once the nearly foolproof doping-detection method known as the "biological passport" came in, which was why the whole antidoping inquisition was pointless. The problem was already solved. Before his comeback, he called the infamous doctor Michele Ferrari, subject of so many doping rumors and investigations, and asked if he could still win the Tour clean. Ferrari said he had to run some numbers. Later he called back. "If you're lucky."
The antidoping agency accused him of cheating anyway, saying there was a one-in-a-million chance that Armstrong didn't have transfusions of his own blood in '09. "Bullshit," he says. "They'll find out someday, 'cause they'll perfect that transfusion test. And I'll be the first guy to say 'Use it.' "
From Lance Armstrong in Purgatory: The After-Life in Esquire.
Two years ago I saw Alex Gibney's documentary The Armstrong Lie at TIFF. Betsy Andreu and Jonathan Vaughters were at the Q&A afterwards. One of the things in the documentary that struck me was that though Armstrong confessed all of his doping to Alex Gibney in an interview that was shot immediately after the confession to Oprah, Armstrong still insisted that he did not dope in his comeback to the Tour de France in 2009. He said he made a promise to his wife Kristin that if he came back to the sport he would do it clean.
Most everything that can be written or said about Armstrong has been already. This mystery remains: why does he hold on to his insistence that he raced clean in his comeback? Is it because he didn't win those two Tours in his two comeback years and, his competitive juices still flowing, feels he was racing against cyclists who were doping? Is it just a point of pride for him, one tiny consolation for those who feel the rest of his cycling career was tainted by the doping scandal?
In The Armstrong Lie, Gibney followed Armstrong on his comeback in 2009 and 2010. The 2009 Tour was not, given Armstrong's expectations, a success. When put to the test in early mountain stages, he could not stay with overall Tour contenders like teammate Alberto Contador or Andy Schleck.
He had one late minor triumph, however. On stage 20, which ended with a climb up the legendary Mont Ventoux, Armstrong stayed with the top Tour racers all the way to the top. It was like an aging superstar summoning one final hurrah. The next morning, the headline of L'Equipe read “Chapeau Le Texan” (hats off to the Texan).
Andy Schleck, who finished second in the 2009 Tour, just ahead of Armstrong, believes Armstrong raced clean that year.
"He made his comeback and he was beaten in the first year by Alberto and me," said Schleck, who is in Australia to ride the Tour Down Under, first event of the 2013 WorldTour.
"So, in my eyes, I was clean. I know I was always a clean rider and I keen on riding clean. So why should he be behind me?
"I believe in his comeback that he was clean."
“When he said that about 2009-10, I thought 'you lying b------',” said Wiggins, recalling two particular mountain stages in the 2009 event. “I can still remember going toe to toe with him and watching the man I saw on the top of Verbier in 2009 to the man I saw on the top of Ventoux a week later when we were in doping control together. It wasn't the same bike rider.”
Because 2009 was raced with under the new biological passport program, we have data with which to assess Armstrong's performance that year, and several people pointed out some suspicious values in the data.
Here are Armstrong's blood and urine test results from 2009:
The data that some flagged as suspicious were his Hemoglobin and Hematocrit % from July 2, just before the Tour started, to stage 20 on July 25, the day Armstrong kept up with the race leaders on the climb up Mont Ventoux, looking to be in better form than he had the entire Tour.
Most cyclists see declines in hemoglobin and hematocrit levels during the three grueling weeks of a grand tour, so the fact that Armstrong's levels stayed dipped some the first week of the Tour and then bounced back to pre-Tour levels, with a spike from July 11-14, indicates to those who were suspicious of Armstrong (and these days, it's hard to find anyone that isn't) that he performed blood doping or took EPO during the race.
One's hydration level can swing those scores, so it's not a foolproof data point, but if you're inclined to doubt, the value changes from 7/20 to 7/25 support a constructed narrative that have Armstrong, suffering to hold onto his podium spot, taking a bag of blood sometime before the stage up the grueling Mont Ventoux to hang onto the final step of the podium. Indeed, Armstrong finished the Tour in third place, holding off Bradley Wiggins by 37 seconds.
I suspect it's hard to find too many people who believe that Armstrong was really clean in 2009, and yet Armstrong still clings to that claim with his well-chronicled Texan defiance.
There is one other theory about why Armstrong hangs on to this one last story of competitive integrity, and that is a legal one. I've heard from reporters who've covered the Armstrong story that there is a five year statute of limitations and that Armstrong had to claim innocence in 2009 to prevent more organizations from coming after him for various prize and sponsorship money. Given the slew of legal actions against him, it could be that his innocence in 2009 is critical to him maintaining financial solvency.
Now that we're in 2014, that statute of limitations is close to expiring. Perhaps at that point Armstrong will confess to doping even in his comeback, and the final few tumbles from grace will complete themselves.
Abstract of this research paper:
Three studies tested the hypothesis, derived from construal-level theory, that hierarchical distance between leaders and followers moderates the effectiveness of leader behaviors such that abstract behaviors produce more positive outcomes when enacted across large hierarchical distances, whereas concrete behaviors produce more positive outcomes when enacted across small hierarchical distances. In Study 1 (N = 2,206 employees of a telecommunication organization), job satisfaction was higher when direct supervisors provided employees with concrete feedback and hierarchically distant leaders shared with them their abstract vision rather than vice versa. Study 2 orthogonally crossed hierarchical distances with communication type, operationalized as articulating abstract values versus sharing a detailed story exemplifying the same values; construal misfit mediated the interactive effects of hierarchical distance and communication type on organizational commitment and social bonding. Study 3 similarly manipulated hierarchical distances and communication type, operationalized as concrete versus abstract calls for action in the context of a severe professional crisis. Group commitment and participation in collective action were higher when a hierarchically proximate leader communicated a concrete call for action and a hierarchically distant leader communicated an abstract call for action rather than vice versa. These findings highlight construal fit’s positive consequences for individuals and organizations.
We want concrete plans from direct managers, abstract vision from leaders further away. You can get away with abstraction the longer the time horizon you're talking about.
[via Robin Hanson]
Sparked by this tweet...
...the Guardian asked readers to submit other photos that evoke Italian Renaissance paintings. Here are their submissions. I liked these three in particular (with original captions below each):
It's not just the composition of these but the color palette, body posture, and facial expressions that produce the effect. We just need a Photoshop filter to simulate that color palette and we've created a meme, albeit one that's probably too highbrow to really go viral.
Headlines, in their original 17th-century form, existed to tell you what section of a book you were in and what page you were on. Usage of the word in a newspaper context came later, in the 1890s. Then, their existence was justified because it had been freshly born of necessity: if your primary mode of story discovery is to look at a newspaper frontpage, you need headlines to tell you where to go.
That function, now, has been largely outsourced. On a site like BuzzFeed, or the New York Times, the front page headline still serves that diminished original purpose (and as long as we have front pages, it always will). But for many stories — a growing number — that headline is not the most important one. The headline that’s most successful on Facebook or Twitter, or in a link from another site, is that story’s headline *by default.*
This might seem like a subtle change, but it’s experientially significant. When someone arrives on a story, the headline they followed is the only headline they need to see — to show them another promotional headline is to oversell or tell them they’ve been misled; it’s like welcoming someone to a “casual get together” by saying, “welcome to Joe’s birthday party.”
Canonical online headlines, at most, tell you you’re in the right place; that the tweet you followed was not a lie. But headlines as we write them today aren’t even very good at that. If they’re nü-internet-style headlines, they’re sales pitches or orders (see Choire Sicha’s fantastic Take A Minute To Watch The New Way We Make Web Headlines Now from a couple weeks ago). If they’re classic, print-style headlines, they’re meant either to stand out in sea of text or tease you into reading more (online, however, the choice has already been made — you chose to visit the page). What stories on the internet need, then, is not a headline but what’s known as a dek — a simple description that both confirms the remote headline and adds to it. A boldfaced introduction, in other words, that flows seamlessly into the first sentence. There’s no need to reset with yet another headline.
This is from a post at Buzzfeed that is over a year old now, but I just came across it, who knows how. Given that most sites are probably even more reliant on social media for inbound traffic than a year ago, it still holds.
I think the harm of having your own headline for posts is a bit overstated. For one thing, I still have quite a bit of inbound traffic from newsreaders, and the headlines matter in an RSS reader interface.
I rarely begin a post here with a headline, and the act of having to come up with a headline after the fact helps me to coalesce the thesis of the post. Sometimes having to find the through line forces me to go back and trim excess, and quite often I realize there is no good logline and just abandon the post in my drafts folder.
Still, the Buzzfeed article did get me thinking that I'd love to have a widget of some sort on each of my posts that would pull in the top headlines as written by people who linked to that post on Twitter, Facebook, another blog, etc. I can only imagine that if you sorted that list of alternate headlines by the highest to lowest traffic drivers you'd see the most linkbaity headlines on top.
If someone comes up with a tool like that, let me know.