Algorithme et bonheur

IMG_0349-0.JPG
Une avis assez intéressant sur l’expérience de 2012 de Facebook qui a manipulé les murs de ses utilisateurs afin de modifier leur émotions.

A noter que l’auteur fait parti de Microsoft research New-England ce qui peux relativiser certains propos.

Cette approche plus psychologique que commerciale est intéressante car elle met en avant une autre facette sur le pourquoi de cette étude : Facebook nous préfère heureux car nous reviendrons alors sur Facebook.
Mettre en avant des contenus plus légers afin de fidéliser ses utilisateurs.

Facebook retouche activement le contenu que nous voyons. La majorité d’entre nous se concentre sur l’aspect commercial lié à cette pratique, mais la tâche principale des algorithmes est surtout de mettre en avant les contenus qu’ils pensent que nous avons envie de voir. Facebook détermine ainsi, par l’algorithmie, quelles publications d’amis vous verrez, non à des fins seulement commerciales, mais parce qu’ils veulent que vous reveniez sur le site jour après jour, que vous vous y sentiez bien, que vous n’y soyez jamais accablé. Leurs algorithmes servent quotidiennement à manipuler vos émotions. Quels critères participent à cette démarche ? Nous n’en savons rien.

Nous sommes manipulés. Nous le savons mais ce que nous trouvons désagréable, c’est que l’on nous le dise. C’est encore plus effrayant ainsi.

Pour le meilleur ou pour le pire, les gens s’imaginent que Facebook leur est offert par un dictateur bienveillant, que le site est là pour leur permettre de mieux communiquer avec les autres. Dans un certain sens, c’est vrai. Mais Facebook est aussi une entreprise. Et une entreprise publique, en l’espèce. Elle doit trouver le moyen de devenir plus rentable à chaque trimestre. Cela signifie qu’elle conçoit ses algorithmes non seulement pour vous adresser des publicités ciblées, mais aussi pour vous convaincre de revenir encore et encore. Les gens ont une idée abstraite de la façon dont cela fonctionne, mais ils ne le savent pas exactement, et ne veulent pas vraiment savoir. Ils veulent juste que le hot-dog soit bon. Que ce soit sous le sceau de la recherche ou des opérations, les gens ne veulent pas penser qu’ils sont manipulés. Alors, quand ils découvrent ce quoi est fait ce Soleil vert, ils sont outrés. Cette étude n’est donc pas vraiment ce qui est en jeu : ce qui est en jeu, c’est la dynamique sous-jacente de la façon dont Facebook gère ses activités, exploite son réseau, et prend des décisions qui n’ont rien à voir avec la façon dont ses utilisateurs veulent que Facebook fonctionne. Il ne s’agit pas de recherche : c’est une question de pouvoir.

Via www.rslnmag.fr

L’original en anglais sur http://socialmediacollective.org/

Drogue visuelle

Prenez de la drogue. Regardez cette vidéo.

gopro-on-wheel-of-car

Ryan Fox affixed a GoPro to the wheel of his car just to see what the footage would look like. Turns out it’s pretty trippy and slightly hypnotizing!  

 
Lo vi en TwistedSifter http://ift.tt/1n59yBT

Prendre de la hauteur.

Le monde est plus beau de là-haut.
earth-timelapse-from-iss-footage 

Using publicly available photographs taken by astronauts onboard the International Space Station, Bruce W. Berry edited this incredible timelapse sequence. All footage was color graded, denoised, deflickered, slowed down and stabilized and then complied and converted to 1080 HD at 24 frames/sec by Berry. Music: “Manhatta” composed & performed by The Cinematic Orchestra

Images…
Lo vi en TwistedSifter http://ift.tt/1r4P6JI

En images : des vieilles tapisseries transformées en jeux vidéo

Dans la famille “art versus pop culture”, après les citations hip-hop sur les peintures et les estampes japonaises version Game of Thrones, voici les tapisseries de Mamie remixées façon jeux vidéo. C’est Gauvain, un artiste français dont le blog a été repéré par Fluctuat, qui s’amuse à mélanger des broderies vieillottes avec des images de […]

The post En images : des vieilles tapisseries transformées en jeux vidéo appeared first on Konbini.

Lo vi en Konbini http://ift.tt/1l46gPd

20140727-231141-83501526.jpg

Le devin

Comme Astérix dans le tome de ses aventures du même nom, l’avait en son temps compris, il suffit de tout prédire pour que cela se réalise.
La version moderne est un peu plus complexe mais très simple.

Un complot horrible entache la dernière coupe du monde de foot : un « hacktiviste » avait publié 6 mois auparavant les vainqueurs des matchs de la coupe du monde sur twitter. Complot, match truqué.

Non il a juste publié toutes les variantes puis supprimé ses tweets inexacts au fur et à mesure du déroulement de la coupe.

A la fin le résultat est bluffant : des résultats justes parfaitement anti datés.
Lo vi en Medium http://ift.tt/1oEMn2y

Le catalogue vraiment virtuel.

>20140727-230936-83376488.jpg
75 % du catalogue IKEA est en images de synthèse

Avez-vous déjà feuilleté un catalogue IKEA en vous demandant comment faisait le géant suédois pour avoir autant de décors différents dans lesquels mettre en valeur ses produits ? S’il a pendant longtemps reconstitué des intérieurs de maisons et d’appartements dans ses studios photos, le fabricant de meubles en kit a entamé une véritable révolution interne à partir du milieu des années 2000, en jetant son dévolu sur les reconstitutions photo-réalistes en images de synthèse.


Mobility on demand

Pourquoi posséder quand on peut tout partager.
C’est cette maxime que met en place la ville de Helsinski pour ses transports : habitez la ville, payez au forfait ou à la consommation et vous accédez librement aux vélos, bus, tram, parking, et voiture de la ville. Le tout coordonné par votre téléphone.

Ten years from now, transportation in Helsinki may operate very differently from the current system.

The service will be run by transportation operators, through which the regular citizen can buy all they want with a click. This does not only entail public transportation within the city, but also carpool, taxi, a train ticket to Tampere or parking fees in the city centre.

Few want to own their own car in future, when everything can be shared. If one wishes to travel from Puotila to Pukinmäki, the « route planner » of 2025 will provide information on where to change the city bike for a car due to impending rain, in addition to information on the fastest connection.

The City of Helsinki believes in the model so strongly that it plans to test it at the turn of the year with a few major employers in Vallila. Employers are being persuaded to join in by building a platform that enables employees to buy transportation services with their own funds.

via kottke.org http://www.helsinkitimes.fi/finland/

Sexisme sci fi

Voilà une forme étrange et pointue de sexisme : la notion de vaisseau mère que l’on doit détruire pour vaincre.
Presque toujours féminin, toujours méchante. Toujours détruite par Tom Cruise.

Image: Edge of Tomorrow promotional image.

Spoiler alert. Spoiler alert for the new Tom Cruise flick Edge of Tomorrow, spoiler alert for the slightly older Tom Cruise flick Oblivion, and spoiler alert—for good measure, while we’re at it—for a lot of Tom Cruise flicks, plus a healthy percentage of science fiction movies from the last fifty years. Spoiler alert for a trope so insidious, so completely folded into the habitual narratives of our blockbusters, that it has become all but invisible.  

This is the spoiler: at the end of the movie, if he is going to save the world, our hero has to find the mothership, and he has to blow it up. He has to go inside the mothership—penetrate it—through a small aperture. Once there, he won’t have much time to plant his weapon, and it’s likely he’ll barely make it out before a bomb explodes deep within the mothership, collapsing it from within, shooting our hero back into the world on the shockwave of a single, orgasmic shudder of fire and light. 

This meta-spoiler should conjure a montage of cinematic climaxes, from the offing of the Death Star to the menacing shadow of Independence Day’s mighty mothership, wide enough to block out the sun, because it is a storytelling ritual, an incantation performed ad nauseam in the dark collective multiplex of consciousness. 

Independence Day’s Mothership. Screenshot.

The expression “mothership” dates to pre-science fictional times. In the 19th-century whaling trade, a large ship would serve as the central hub from which a group of small, fast cutters, designed to chase and kill whales, would operate. This ship, the mother, was reserved for the processing and storage of whale meat. Such a ship is now called a “factory ship.”

But the variant used almost exclusively today, in science fiction and in the popular imagination, has a very different lineage. It dates back to the first generation of UFO hysteria, which was marked by a spell of sightings in 1947 referred to as the “Great Flying Saucer Wave.” The saucers of said wave were described by those who saw them in manifold ways: as discs skipping across the sky like flat stones on a pond, as platters in birdlike formation, as flying pie-plates. But it was an anonymous woman in Palmdale, California, quoted in the press describing a "mother saucer (with a) bunch of little saucers playing around it" who inadvertently defined the idiom. Mothership. 

The “Mothership” has certainly mutated over the years. The contemporary usage—the trope that should be familiar to any viewers of celluloid SF—is light-years from the Palmdale woman’s vision of a mama surrounded by playful flying saucer babies. Instead, it’s drawn some meaning from the original whaling motherships, as well as imported structural cues from colonies of bees and ants. Now the Mothership is part queen bee, part battery pack. She controls all the power, meting out orders to the drones and fueling them with whatever mysterious alien lifeblood is the look du jour: telepathic commands, the master code, the central brain. Kill the mother, kill the hive. 

On the most elemental level, a Mothership is an easy out for filmmakers, a mechanism by which the scrappy human race can vanquish insurmountable spaceborne invaders using existing technology. David and Goliath stories always hinge on the giant’s central weakness; as he can be beaned in the forehead with a rock, so can an alien fleet be downed instantly with a straight shot to the core. When a mother is killed, in life as in science fiction, it destabilizes the system—robs it of a slow-moving port, mind, home, and family. It also prevents the enemy from reproducing itself, the strategic advantage of the female. 

Killing the mothership is standard operating procedure in alien invasion films: in Independence Day, the mothership must be hacked to deactivate the shields of all the defensive fighters. In Oblivion, Tom Cruise must enter and neutralize the Tet, an alien artificial intelligence with a soothing maternal voice that goes by the name of Sally.

In Edge of Tomorrow (again, spoilers) it’s more of the same: an “Omega,” the primary source of the alien power, characterized almost immediately as a “she,” must be entered and destroyed from within. 

Why not fatherships? In a largely patriarchal world, and in a genre traditionally dominated by male authors and consumers, why is the predominant image of alien power feminized? Well, obviously, aliens have always stood for the “Other.” They are placeholders for the inscrutable, for all those things about women (and for other races, nations, gender assignations, ideologies, too) which are frightening to the readership. Women are a secret power. It makes sense, then, that the summit of Otherness, its military-industrial peak, would be that symbolic apex of womanhood: the Mother. Killing her destroys the alien’s capacity to reproduce—the one thing men alone can’t do.

Science fiction was born from a generation of pulp publishing and popular mechanics magazines marketed to young men, filled with young men’s fantasies of hapless damsels and big steely rockets. Although the great feminist science fiction authors of the mid-1970s—Joanna Russ, Octavia Butler, Ursula K. LeGuin, James Tiptree, and the critic Donna Haraway, whose “Cyborg Manifesto” is a foundational document for yours truly—did much to rearrange this boy’s club, creating matriarchal laboratories in the cosmos, it remains largely subversive to worldbuild outside of the patriarchy.

Which is why we usually take the mothership for granted. What is alien is Other, and what is Other is certainly not male—these are transparent truths, as elemental to science fiction cinema as three-act structure, symptoms of the assumptions we make about ourselves. Again and again, Tom Cruise blows up a womb and saves the world. 

Some science fiction films employ a computational variant on the mothership idea. Instead of a massive craft, a cosmic aircraft carrier hanging in the void, the mothership is a central intelligence system: like the Tet, Elysium’s core computer, the Terminator franchise’s indomitable Skynet, or the “Brain Bug” from Starship Troopers, the most vaginal alien life-form in the entire universe. If it has a voice, like Siri or Cortana, it’s female. “I’m in!” pronounces the hacker who bests her.

In this context, the mothership is the central node of a network. The effect of downing her is like knocking the cloud offline: all the agents, every connected device or entity, is simultaneously robbed of its brain. A million dead drones fall from the sky, clunk-clunk. Which is to say, if we have a mothership among us today, it’s the Internet. 

Unlike the motherships and maternal networks of science fiction, being imploded at every turn by myriad Tom Cruises—those atavistic male heroes—we wouldn’t immediately die if it took a hit, but we’d certainly be weakened, sent wheeling out into the void with nothing left to conquer. Something to consider as we wade through the briar of net neutrality, as the powers that be try their hardest to infiltrate the mothership and plant bombs within her. No spoilers. 

Lo vi en Motherboard RSS Feed http://ift.tt/1lcjOfw

Qu’importe le flacon

Pourvu qu’on ai l’ivresse (sonore)
Voilà une façon originale de presser des pirates.

The history of pirated music extends far deeper than people just not wanting to pay money for something. The Soviet Union, circa the 1940s and ’50s, was a place with zero tolerance for "ideologically foreign" music, according to a 2011 piece in Der Spiegel (via FACT). American jazz, blues, and rock and roll were musics that could get their listener thrown out of school or, worse, sent to prison or a labor camp.

Nonetheless, demand persisted, particularly among Soviet hipsters known as stilyagi. The actual machine for putting music onto records, the "recording apparatus," was easy enough to build, but the record material itself, celluloid, was another matter. "An unexpected path that led to where no one would have guessed the solution: in the archives of the Leningrad hospitals," Der Spiegel recounts (translation by Google). "There outsourced thousands of old x-rays, for which nobody use. Images of broken hands and feet, ribs, skull caves and hip bones." Radiographic film isn’t quite the same as celluloid or its successors, and was prone to irregularities and bending/warping, the result being "wild antics" in playback.

"Rock on the bones" or "skeleton of my grandmother" became keywords for the pirated material, which might be purchased from a dealer in some dark corner of a Moscow subway station. The underground trend went on for some 15 years, only to be subsumed by the introduction of cassette tapes, a device that, relatively speaking, would seem to be invented with pirates in mind.

Image: Wiki

Lo vi en Motherboard RSS Feed http://ift.tt/1iXfJHv

Message in a (plastic) bottle.

Il y a des tonnes de plastiques dans les océans mais… pas assez!
Ou en tout cas pas assez selon notre production. Oú disparaît donc ce plastique? En micro billes absorbées par le plancton

Image: Kevin Krejci/Flickr

There’s some 40,000 tons of plastic floating on the surfaces of the world’s oceans, which is leaving researchers wondering: Where the hell is the rest?

That number is nothing to scoff at, of course, but it’s many orders of magnitude lower than the estimated amount of plastic that has been going into the oceans since at least the mid-1970s. Plastic in the ocean isn’t simply disappearing, but it has to be going somewhere. And that’s the scary thing.

In a paper published today in Proceedings of the National Academy of Sciences, Andres Cozar of Spain’s University of Cadiz and an international team of colleagues report that the "quantity of plastic floating in the ocean and its final destination are still unknown."

"A conservative first-order estimate of the floating plastic released into the open ocean from the 1970s (10^6 tons) is 100-fold larger than our estimate of the current load of plastic stored in the ocean," Cozar wrote. "Large loads of plastic fragments with sizes from microns to some millimeters are unaccounted for in the surface loads. The pathway and ultimate fate of the missing plastic are as yet unknown."

Cozar has a couple theories, which we’ll get to in a minute. Chief among them, however, is the idea that fish are eating microplastics (mistaking them for plankton or accidentally eating them along with plankton, which are increasingly calling plastic home) and pooping them out. The feces is then dense enough to sink to the bottom of the ocean, and that’s where all the plastic is.

Gross, yeah, and probably not good news if we want to have any shot of cleaning this stuff up.

The results of Cozar’s survey. Image: PNAS

Just because we don’t know where a lot of this stuff goes doesn’t mean that there isn’t an incredible amount of micro plastics floating on Earth’s oceans. Plastic generally doesn’t sink under normal circumstances, and 88 percent of the more than 3,000 samples from around the world that his team took had micro plastics in it.

As you might expect, roughly 35 percent of the total amount of micro plastics are located in the North Pacific Ocean, home of the gyre that many have begun referring to as a floating island of plastic. There are also substantial gyres in the North and South Atlantic Ocean, and the Southern Indian Ocean. 

Back in the 1970s, the National Academy of Sciences estimated that roughly 45,000 tons of plastic made its way into the oceans each year, and that was before the annual production of plastic increased fivefold—in 2010, the world made 265 million tons of plastic, for instance.

That brings us to the crux of the study, and the question that’s probably on your mind—where is it? Cozar has four theories, none of them particularly good news for ocean health. 

Shore deposition: Basically, plastics somehow make it out of gyres in the middle of the ocean and make it back onto shore somewhere. This is very unlikely to happen, for pretty obvious reasons, namely that it generally defies the laws of physics. Gyres are essentially very large, circular tides. Absent many large storms, the plastic trapped in the middle of them isn’t making its way back to shore. Cozar wrote, "A selective washing ashore of the millimeter-sized fragments trapped in central areas of the open ocean is unlikely."  

Nanofragmentation: This is the idea that micro plastics have become "nano plastics" that are very, very difficult to detect. Plastic naturally breaks into tinier pieces, and the sun has something to do with that, but Cozar says there’s no reason to suggest that "solar-induced fragmentation" has increased since the 1980s, when several studies were done on the phenomenon.

For the plastics to be broken down further, there’s likely some sort of bacteria or plankton that has evolved to do it, or that does so naturally. There is some research to back that up. "Recent scanning electron micrographs of the surface of micro plastic particles showed indications that oceanic bacterial populations may be contributing to their degradation, potentially intervening in the fragmentation dynamics," Cozar wrote. 

Biofouling: We’ve seen animals make homes out of plastic, we’ve seen reefs that incorporate plastic—that’s biofouling. Cozar suggests that plankton and other small organisms may be accumulating on the plastic, making the plastics able to sink, probably very slowly because seawater density gets higher with depth.

This is another potentially sound possibility, were it not for the fact that, in field tests, plastic makes a very poor home for much of anything. "Field experiments have shown that biofouled plastic debris undergoes a rapid defouling when submerged, causing the plastic to return to the surface," reads the report.

Ingestion: This is the most likely scenario, Cozart suggests. It’s not a pretty one. Microplastics can end up being roughly the same size as zooplankton, an incredibly important part of the oceanic food chain. Previous studies have found that fish that eat plankton often have plastic in their stomachs, so it’s not a farfetched idea. The idea here is that fish eat the plastic, poop it out, and it sinks to the bottom of the ocean.

"Gut content of [plankton-eating] fish is evacuated as long viscous feces that assume spheroid shapes while sinking at high velocities," he wrote. "Hence, micro plastic fragments could also reach the bottom via defecation, a proposition that requires further quantitative testing."

The overall answer, of course, is probably some combination of the four of these scenarios. We’re going to have to figure it out if we want an outside shot of ever cleaning up the oceans. Maybe we can use what we find to fuel 3D printers.

Lo vi en Motherboard RSS Feed http://ift.tt/1z342sP

Hacker le cerveau pour influer sur les décisions.

Le rêve de tout marketer, chaîne de télé ou dictateur (rayer la mention inutile) va se réaliser : des scientifiques ont piraté le cerveaux de macaques (nos plus proches cousins) et leur ont fait changer d’avis lors de prises de décision.
Pas encore un risque immédiat, la procédure passant par une opération chirurgicale mais déjà le scientifique de s’inquiéter du danger de cette procédure.

A rhesus macaque. Image: Peter Nijenhuis/Flickr

While your brain should still be safe from hackers for some time yet, a new study, in which macaques had their choices controlled by electrical impulses, adds to a growing body of work that suggests brains can be manipulated with a surprising degree of precision.

Using electrodes implanted in the ventral tegmental area (VTA), a region deep in the brain associated with the reward circuitry of the brain, researchers were able to fundamentally influence the decision making of macaques. The work was published today in Current Biology.

The study, conducted by a joint team from KU Leuven in Belgium, Massachusetts General Hospital, and Harvard Medical School, consisted first of an A/B test in which macaques were shown a pair of images, and their preference for one or the other recorded. Some monkeys might prefer a picture of a ball, others a star, but in any case, the research team was able to glean a baseline preference for each individual.

Then came the big test: Could electrical microstimulation affect the results? Indeed, by applying small, regular electrical impulses to the VTA, the team "was capable of selectively reinforcing and motivating behavior during operant and Pavlovian conditioning paradigms." In other words, after flipping the switch, macaques that preferred image A picked image B, and vice versa.

I asked Wim Vanduffel, a co-author of the report, if the results suggest that electronically-controlled decision making is possible, to which he emailed, "Certainly so!"

The team used fMRI imaging to guide implantation of the electrodes.

"The data show that the preference of the monkeys changes quite dramatically," he wrote. "The effect was slightly larger in the first compared to the third animal, likely because of the positioning of the electrodes (but that is speculation)."

Of course, it’s more complex than simply plugging a wire into a monkey’s brain and controlling what it does. For one, the team, led by John Arsenault, still had to convince the macaques to actually go along with the game, which meant adding in an additional juice reward; the animals didn’t suddenly turn into robots.

"Indeed, we could not have the monkey ‘work’ for microstimulation alone," Vanduffel wrote. "In other words, to reinforce the operant behavior we used an additional reinforcer (juice)."

And lest you fear that your brain could be taken over by someone on the street, the effect of electrical stimulation on decision making requires the extremely precise placement of electrodes deep into the brain—something not likely to happen without you knowing.

"The targeting itself is probably ‘ever lasting’: once the electrodes are in the right locations (and if there are no complications), they probably will work for long periods of time," Vanduffel wrote. "The critical issue is to reach a sufficiently large number of VTA neurons. So positioning is very important—and not trivial."

Postoperative image confirming the planting of electroding in the VTA, deep in the brain.

I particularly like the way Vanduffel put it in a release, saying, "Of course, there is also a potential danger here: The method could be used maliciously to manipulate a person’s brain remotely without his knowledge. But as yet, there is no reason to worry. Non-invasive, high-precision methods for stimulating deep brain centers are not yet available."

Regardless, it’s more evidence that stimulating the brain is indeed possible. Previous work has focused on optical stimulation, including particularly futuristic research into controlling mice brains with lasers. Deep brain stimulation is already used in humans, but not with the direct decision-making effects of the most recent study. 

Perhaps the closest to the brain hacking concept—albeit for good—is DARPA’s interest in developing brain stimulation therapies for treating brain trauma, including PTSD. So while we’re a ways off from worrying about people hijacking our brains or an offshoot from the Matrix scenario, it’s becoming increasingly apparent that the brain’s electrical processes can be manipulated with a fine degree of precision—and, one day, we may even be able to copy them.

"We’ve known for a long time that the ‘working of the brain’ can be reduced to the electrical activity of neurons," Vanduffel said. "If one were able to ‘mimick’ this artificially, one would have a working brain. Obviously, this is purely hypothetical since it seems impossible to mimick ~100 billion neurons, and many more connections between them."

Regardless, that we can tap into the electrical systems of the brain means a future where we can regulate our brains with more control than we can imagine now. It’s an area with huge potential for treatment of neurological disorders, as electrical implants could perhaps act with more precision than pharmacological treatmeants—just as long as they don’t get hacked.

Lo vi en Motherboard RSS Feed http://ift.tt/1mLco1P

Crackopolis

Un impressionnant reportage en 15 épisodes sur le crack à Paris. Une claque surtout quand on a vécu dans les quartiers cités.

Une plongée lucide sans voyeurisme sur la vie d’un addict.

A écouter absolument.

ARTE Radio – crackopolis.