Computers are actually pretty simple. We’re sitting here on a bench in this café. Let’s assume that you understood only the most rudimentary of directions and you asked how to find the rest room. I would have to describe it to you in very specific and precise instructions. I might say, “Scoot sideways two meters off the bench. Stand erect. Lift left foot. Bend left knee until it is horizontal. Extend left foot and shift weight 300 centimeters forward…” and on and on. If you could interpret all those instructions 100 times faster than any other person in this café, you would appear to be a magician: You could run over and grab a milk shake and bring it back and set it on the table and snap your fingers, and I’d think you made the milk shake appear, because it was so fast relative to my perception. That’s exactly what a computer does. It takes these very, very simple-minded instructions — “Go fetch a number, add it to this number, put the result there, perceive if it’s greater than this other number” — but executes them at a rate of, let’s say, 1,000,000 per second. At 1,000,000 per second, the results appear to be magic.
Steve Jobs famously directed Apple’s designers to precisely imitate the napa leather seats in his Gulfstream jet to adorn the iCal app. The result was a deceivingly supple and needle stitched interface that, nevertheless, felt like the Gorilla Glass it was. The iOS interface for the iPhone was famous for these expertly crafted, “lickable” icons that were rendered with artificial shadows and depth. Skeuomorphism became the term of art for this habit of borrowing design cues from the physical world to make their digital counterparts attractive and recognizable. Ready for change, designers later embraced “flat design”, a style ostensibly more suited to its medium. But the effort to mirror objects in the world in our artistic creations reaches back into prehistory, from cave paintings and rock sculptures to the latest 3D rendering engine or animatronic sexbot. Whatever the medium, as our tools and skills increase, we are able to make ever more convincing replicas of the genuine article. Today, because of these superficial similarities, we are often fooled into thinking that the artificial intelligence programmed into many of our creations isn’t artificial after all. This misapprehension is the result of being taken in by a magic trick, by a kind of skeuomorphism that Robert J. Marks calls, “seductive optics”.
Skeuomorphism, as wonky as it sounds, is a simple concept. It’s the idea that new designs retain ornamental elements of past iterations no longer necessary to the current objects’ functions. You see this often in Apple’s software: The Notes app is presented as a yellow legal pad, while Game Center is modeled after a Las Vegas-style casino table, with lacquered wood and green-felt cloth.Austin Carr, “A Former iPhone UI Designer Defends Apple’s Fake-Leather Design Philosophy” at Fast Company
Early humans sometimes depicted the drama of the hunt on cave walls by scraping or smearing charcoal and pigment. Others assembled or chipped at rocks to make human effigies. Some of these artifacts are more sophisticated than others, showing movement and depth. Still, no one would think these simple caricatures or the rocks themselves were living, conscious beings. These are plainly soulless, lifeless rocks with no thoughts, hopes, or sorrows.
By the late Renaissance, artists had greatly improved their brushes, pigments, and media as well as their understanding of perspective, foreshortening, underpainting, and shadows. Their artworks captured complex scenes and often the mood of their subjects. We may not know the secret thought behind Mona Lisa’s smirk, but Leonardo Da Vinci’s expert portrayal does make us wonder. Van Eyck’s perfectly painted portraits, with wrinkles and warts and all, make us feel all the more like we can know something of the subject, or that we can peer into Girl with the Pearl Earring’s eyes and share her longings.
Oranges are just one item from an almost infinite list of things seen by Van Eyck with amazing acuity … the textures of fur, flesh, wood, stone and ceramics, the exact pattern of body hair, follicle by follicle, on Adam’s naked body, the precise hue and shape of a wart on a clergyman’s cheek, and so on.“How Jan van Eyck revolutionised painting” by Martin Gayford at The Spectaotr (February 8, 2020)
With increasingly formalized training in academies, artists mastered realism. The subjects of French Academic Realism are so true to life that it is hard not to empathize with them, to storm the Bastille with them, or even to lust after them. Seductive optics indeed.
Today, a subgenera of painters and sculptors called hyperrealists take it even farther, reveling in their ability to capture the most difficult of subjects in a way that is indistinguishable from a photo or living person.
I marvel at this remarkable artistry. I can watch these artists perform their magic for hours on the web. Because each of us has worked with play dough and paints, we know the basics of the process behind these pieces, and so we also know that these images and forms have no more inner life than the oil or clay that comprise them.
Though they may be visually indistinguishable from a man or woman, passing the art equivalent of a Turing Test, we know that these portraits are not in the least the same kind of thing as what they portray. Paint does not suffer. Stone does not breathe.
Dear God! How beauty varies in nature and art. In a woman the flesh must be like marble; in a statue the marble must be like flesh.Victor Hugo, Victor Hugo’s Intellectual Autobiography: Being the Last of the Unpublished Works and Embodying the Author’s Ideas on Literature, Philosophy and Religion “Hyperrealistic Sculptures Blur the Line Between Clay and Flesh”
In truth, Alexa and Sophia are no more conscious or intelligent than Mona Lisa. But the microchip, machine language, and the algorithms lie beyond the veil in darkness, and we are more easily seduced.
Computer animation has followed a similar trajectory toward realism. Early animations were as crude as rock carvings, limited as they were by relatively weak CPU’s, as yet undeveloped rendering engines, and a lack of digital artists. The monochromatic Space Invaders was endlessly playable, as were the whirling, 256 bit dervishes of Centipede and the bubble-gumless, three dimensional spaces of Duke Nukem. But the graphics were a million pixels from virtual reality.
Now decades and countless iterations in, with enormous human capital and ingenuity invested into motion graphics and GPUs, the optics are seductive indeed. Head phones on and eyes glued to the screen, it’s easy to get lost in the sprawling world and epic narrative of Mass Effect, or to feel that you are attacking the Ludendorff Bridge at Remagen with an M1 Garand rifle, your rag-tag squad assisting.
With each iteration, game developers make strides in simulating the unique recoil of each gun model, the aural pattern of explosions near versus far, the game physics of spilt blood and bent grass underfoot, and the fraying edges of a soldier’s ribbon bars, specific to rank and nation.
Bravo! It’s amazing. It’s an adrenaline rush. And it’s all artifice. These animations are becoming ever more realistic, but not in the least bit more real.
In animation … everything is fabricated. … Every moment. Every scratch and look. Everything is a discussion. There are no gifts. There are no happy accidents … It was useful having the live action performance reference. We’d pull them up in front of the animators and say, look at what’s happening here. Little twitches underneath the eyelid, little things that are involuntary became very important.Gore Verbinski, “Rango Behind the Scenes: Breaking the Rules” via Margaret X at YouTube.
In addition to cosmetic advances, motion capture technology has come far. Painstaking attention to plotting our distinctive expressions and gaits has greatly increased the appearance of natural movement. Ever since Andy Serkis so convincingly gave life to Gollum, recordings of humans in motion have supplied the scaffolding for countless characters. One of the joys as a viewer is having that dawning recognition of a familiar actor who has been reincarnated as Aladdin’s Genie, Puss in Boots, Shrek the ogre, or Mator the tow truck. The idiosyncrasies and ticks of the human originals are prized source material for the animators. So, Kristen Bell’s signature phrase — “Wait, what?” — and her habit of biting her lip is mirrored in Anna. Dwayne “The Rock” Johnson’s brow wave personalizes Maui. Although these beloved characters are singing and dancing and joking along, like Mona Lisa, the realism is superficial, artificial. is of the same type as Another vector of advancement has been .
Whatever emotions we bring to the game ourselves, our digital allies and enemies breathe no breaths, make no sacrifices, feel no lonely deaths. Because the textures that skin them and algorithms that animate them are more sophisticated, it is tempting to think that the ducking and evading Juvies, Pouncers, and Swarmaks of Gears of War are more intelligent than Pac Man’s nemeses: Blinky, Pinky, Inky and Clyde.
One might think that whereas the ghosts that haunted Pac Man had, say, .001% of human intelligence, the villainous grunts of a modern first-person shooter have .01%. The truth is, neither has any. Neither has thoughts, plans, or memory. Just as these avatars feel no more, in terms of advancing from artificial intelligence to real intelligence, we have progressed from 0% to 0%.
NPC (Non-Playable Character) AI
It is a category error to think that as we make Just as Mona Lisa’s emotional life is no more complicated than a rock carving, and just as Apple’s redesigned icon had no more leather. Artificial Intelligence is not like a baby growing in understanding,. It is not a rudimentary version of a person. It is a canvas receving more layers, or flywheel getting more gears.
In recent years, we have witnessed a dramatic leap forward in mimicking the primary tool humans use to communicate: speech. Increasingly, our internet connected devices can recognize speech, translate the intent of the speaker, respond using recorded and synthesized voices, and integrate with services that order a pizza or change the thermostat. When we were pushing buttons to control our devices, we were more likely to think of them as behaving mechanistically, as being levers and pulleys at bottom. As dramatized by Joaquin Phoenix and Scarlett Johansson in Her, this new speech-based user interface has introduced a whole new arena of seduction. This aural interface can create a powerful illusion.
Each of the components is key: speech recognition to receive input, Natural Language Processing to translate verbal requests into commands and third party API’s to execute them, and speech synthesis to respond.
Speech recognition is the pledge of this trick: I hear you. Initially, “Audrey” (1952) could only recognize numbers zero through nine, and at that, for best results, only from a particular speaker. It would be decades before speech recognition could pick out the keywords in a human sentence spoken clearly by most anyone. Using machine learning on enormous data sets, speech technology can now generalize over many nuances and varieties of human speech. Fast talkers, low talkers, high talkers, the many accents that enrich human speak, and yada, yada, yada. The ability to receive naturally spoken sentences and go from speech-to-text is Act One.
Before we anthropomorphized these programs as Alexa, Siri, and Cortana, the earliest iterations sounded far from human.
To Joshua (WOPR), the dissonant screeching of a dial-up modem to establish a connection is a perfectly fine “Hello!” But to us, its invitation — “Shall we play a game?” — sounded obviously robotic and impersonal. Early versions of text-to-speech software sounded equally robotic. The words, pieced together individually, lacked the lilting, cadence, emphasis, and rhythm of human speech.
So far, success in humanizing speech technology has been mostly achieved by recording vast libraries of spoken words and phrases. Some melodious few enjoy full-time jobs giving voice to our assistants, recording sentences and word pairings day after day. and seen the process of achieving convincing verbal assistants is . Voice packs featuring celebrity voices became popular on navigation devices in the aughts and are now making their way onto the voice assistants from Amazon, Apple, and Google. For a small fee, you can interact with a disembodied version of the once inimitable Samuel L. Jackson. No doubt the distinctive voices of others will follow.
There will be synthesized versions of gravelly voices, smoker’s hacks with ever having inhaled, nasal tones, deep baritones, fast talking, low talking, Amazon has added functions to its Speech Synthesis Markup Language (SSML) to enable Alexa to whisper, emphasize a word, or understand local slang.
But the limits of these systems are easily discovered. The most basic commands are often mistranslated.
Sense of humor. Preprogrammed jokes.
Talking to your Amazon Alexa is cool, but sometimes her responses can be robotic. (Just because she’s an AI doesn’t mean she has to sound like an AI, right?) Amazon hopes to change that by giving developers the ability to hone Alexa’s responses with “a wider range of natural expression.”Christina Bonnington, “Alexa’s responses are about to get more human” at The Daily Dot
If you ask Alexa what her preferred pronoun is, she replies, “I am female in character.”
This is but one instance of skeuomorphism, a term which refers to the fashioning of artifacts in a form which is more appropriate to another medium. Archeologists often use it to explain the existence of otherwise inexplicable objects. Some of those might be the fake copper rivets on jeans, long made obsolete by modern stitching. Radiator grilles on electric cars. Plastic hair combs dyed to look like tortoiseshell. A camera phone which emits a shutter-click sound. A modern sports car which must meet noise regulations, but pipes in racy sounds through its speakers to console its driver.https://www.bhlingual.com/who-misses-skeuomorphism-blog
Thinking that a computer can learn, see, play, and think is like thinking your computer’s “Desktop” could be chopped up to be used as kindling, that documents in the “Recycle Bin” will be reused as coffee cups, that Mac’s “hello” is a warm greeting from its CPU, and that it has a certain self-awareness when it introduces itself on the local-area-networking as iMac-208c.
Putting it All Together
It is natural to men to judge of things less known, by some similitude they observe, or think they observe, between them and things more familiar or better known. In many cases, we have no better way of judging. And where the things compared have really a great similitude in their nature, when there is reason to think that they are subject to the same laws, there may be a considerable degree of probability in conclusions drawn from analogy. — Reid
There is no subject in which men have always been so prone to form their notions by analogies of this kind, as in what relates to the mind. We form an early acquaintance with material things by means of our senses, and are bred up in a constant familiarity with them.
We are misled by our metaphors and our analogies.
“Skeuomorphic” became one of the nastiest things you could say about a design. A swear word, almost. A hilarious dig at the designers who relied on real world follies to communicate abstract concepts and the silly users who enjoyed them.Michael Flarup, “Bring Back Skeuomorphic Design“