Drawing invisible boundaries in conversational interfaces

One of the things anyone who has worked on textual conversation interfaces, like chatbots, will tell you is that the challenge is dealing with the long tail of crazy things people will type. People love to abuse chatbots. Something about text-based conversation UI's invites Turing tests. Every game player remembers the moment they first abandoned their assigned mission in Grand Theft Auto to start driving around the city crashing into cars, running over pedestrians, just to exercise their freedom and explore just what happens when they escape the plot mechanic tree.

However, this type of user roaming or trolling happens much less with voice interfaces. Sure, the first time a user tries Siri or Alexa or whatever Google's voice assistant is called (it really needs a name, IMO, to avoid inheriting everything the word "Google" stands for), they may ask something ridiculous or snarky. However, that type of rogue input tends to trail off quickly, whereas it doesn't in textual conversation UI's.

I suspect some form of the uncanny valley and blame the affordances of text interfaces. Most text conversation UI's are visually indistinguishable from those of a messaging UI used to communicate primarily with other human beings. Thus it invites the user to probe its intelligence boundaries. Unfortunately, the seamless polish of the UI isn't matched by the capabilities of chatbots today, most of which are just dumb trees.

On the other hand, none of the voice assistants to date sounds close to replicating the natural way a human speaks. These voice assistants may have more human timbre, but the stiff elocution, the mispronunciations, the frequent mistakes in comprehension, all quickly inform the user that what they are dealing with is something of quite limited intelligence. The affordances draw palpable, if invisible, boundaries in the user's mind, and they quickly realize the low ROI on trying anything other than what is likely to be in the hard-coded response tree. In fact, I'd argue that the small jokes that these UI's insert, like answering random questions like "what is the meaning of life?" may actually set these assistants up to disappoint people even more by encouraging more such questions the assistant isn't ready to answer (I found it amusing when Alexa answered my question, "Is Jon Snow dead?" two seasons ago, but then was disappointed when it still had the same abandoned answer a season later, after the question had already been answered by the program months ago).

The same invisible boundaries work immediately when speaking to one of those automated voice customer service menus. You immediately know to speak to these as if you're addressing an idiot who is also hard of hearing, and the goal is to complete the interaction as quickly as possible, or to divert to a human customer service rep at the earliest possible moment.

[I read on Twitter that one shortcut to get to a human when speaking to an automated voice response system is to curse, that the use of profanity is often a built-in trigger to turn you over to an operator. This is both an amusing and clever design but also feels like some odd admission of guilt on the part of the system designer.]

It is not easy, given the simplicity of textual UIs, to lower the user's expectations. However, given where the technology is for now, it may be necessary to erect such guardrails. Perhaps the font for the assistant should be some fixed-width typeface, to distinguish it from a human. Maybe some mechanical sound effects could convey the robotic nature of the machine writing the words, and perhaps the syntax should be less human in some ways, to lower expectations.

One of the huge problems with voice assistants, after all, is that the failures, when they occur, feel catastrophic from the user perspective. I may try a search on Google that doesn't return the results I want, but at least something comes back, and I'm usually sympathetic to the idea that what I want may not exist in an easily queryable form on the internet. However, though voice assistant errors occur much less frequently than before, when they do, it feels as if you're speaking to a careless design, and I mean careless in all sense of the word, from poorly crafted (why didn't the developer account for this obvious query) and uncaring (as in emotionally cold).  

Couples go to counseling over feeling as if they aren't being heard by each other. Some technology can get away with promising more than they can deliver, but when it comes to tech that is built around conversation, with all the expectations that very human mode of communication has accrued over the years, it's a dangerous game. In a map of the human brain, the neighborhoods of "you don't understand" and "you don't care" share a few exit ramps.

10 more browser tabs

Still trying to clear out browser tabs, though it's going about as well as my brief flirtation with inbox zero. At some point, I just decided inbox zero was a waste of time, solving a problem that didn't exist, but browser tab proliferation is a problem I'm much more complicit in.

1. Why the coming-of-age narrative is a conformist lie

From a more sociological perspective, the American self-creation myth is, inherently, a capitalist one. The French philosopher Michel Foucault theorised that meditating and journalling could help to bring a person inside herself by allowing her, at least temporarily, to escape the world and her relationship to it. But the sociologist Paul du Gay, writing on this subject in 1996, argued that few people treat the self as Foucault proposed. Most people, he said, craft outward-looking ‘enterprising selves’ by which they set out to acquire cultural capital in order to move upwards in the world, gain access to certain social circles, certain jobs, and so on. We decorate ourselves and cultivate interests that reflect our social aspirations. In this way, the self becomes the ultimate capitalist machine, a Pierre Bourdieu-esque nightmare that willingly exploits itself.
‘Growing up’ as it is defined today – that is, as entering society, once and for all – might work against what is morally justifiable. If you are a part of a flawed, immoral and unjust society (as one could argue we all are) then to truly mature is to see this as a problem and to act on it – not to reaffirm it by becoming a part of it. Classically, most coming-of-age tales follow white, male protagonists because their integration into society is expected and largely unproblematic. Social integration for racial, sexual and gender minorities is a more difficult process, not least because minorities define themselves against the norm: they don’t ‘find themselves’ and integrate into the social context in which they live. A traditional coming-of-age story featuring a queer, black girl will fail on its own terms; for how would her discovering her identity allow her to enter a society that insists on marginalising identities like hers? This might seem obvious, but it very starkly underscores the folly of insisting on seeing social integration as the young person’s top priority. Life is a wave of events. As such, you don’t come of age; you just age. Adulthood, if one must define it, is only a function of time, in which case, to come of age is merely to live long enough to do so.

I've written about this before, but almost always, the worst type of film festival movie is about a young white male protagonist coming of age. Often he's quiet, introverted, but he has a sensitive soul. As my first year film school professor said, these protagonists are inert, but they just "feel things." Think Wes Bentley in American Beauty filming a plastic bag dancing in the wind for fifteen minutes with a camcorder, then showing it to a girl as if it's Citizen Kane.

If they have any scars or wounds, they are compensated for with extreme gifts. Think Ansel Elgort in Baby Driver; cursed with tinnitus since childhood, he listens to music on a retro iPod (let's squeeze some nostalgic product placement in here, what the hell, we're also going to give him a deaf black foster father to stack the moral cards in his favor, might as well go all the way) and is, that's right, the best getaway driver in the business.

Despite having about as much personality as a damp elephant turd, their beautiful souls are both recognized and extracted by a trope which this genre of film invented just for this purpose, the manic pixie dream girl.

[Nathan Rabin, who invented the term manic pixie dream girl, has since disavowed the term as sometimes misogynist, and it can be applied too broadly like a hammer seeking nails, but it doesn't undo the reality that largely white male writing blocs, from guilds to writer's rooms, aren't great at writing women or people of color with deep inner lives.]

This is tangential to the broader point, that the coming-of-age story as a genre is, in and of itself, a lie. It reminds me of the distinction between Finite and Infinite Games, the classic book from James Carse. The Hollywood film has always promised a finite game, and thus it's a story that must have an ending. Coming-of-age is an infinite game, or at least until death, and so we should all be skeptical of its close-ended narrative.

(h/t Michael Dempsey)

2. Finite and Infinite Games and The Confederate

This isn't a browser tab, really, but while I'm on the topic of Carse's Finite and Infinite Games, a book which provides a framework with which so much of the world can be bifurcated, and while I'm thinking about the white male dominated Hollywood profession, I can't help but think of the TV project The Confederate, by the showrunners of Game of Thrones.

"White people” is seen by many whites as a pejorative because it lowers them to a racial class whereas before they were simply the default. They are not accustomed to having spent their entire lives being named in almost every piece of culture as a race, the way women, people of color, and the union of the two are, every single day, by society and culture.

All Lives Matter retort to Black Lives Matter is to pretend that we're all playing the same finite game when almost everyone who are losers in that game know it is not true. Blacks do not feel like they “won” the Civil War; every day today they live with the consequences and the shadow of America's founding racism, every day they continue to play a game that is rigged against them. That is why Ta Nehisi Coates writes that the question of The Confederate is a lie, and that only the victors of this finite game of America would want to relitigate the Civil War in some Alt History television show for HBO. It's as if a New England Patriot fan asked an Atlanta Falcons fan to watch last year's Super Bowl again, with Armie Hammer playing Tom Brady.

“Give us your poor, your huddled” is a promise that the United States is an infinite game, an experiment that struggles constantly towards bettering itself, evening the playing field, such that even someone starting poor and huddled might one day make a better life and escape their beginning state. That is why Stephen Miller and other white nationalists spitting on that inscription on the Statue of Liberty is so offensive, so dangerous.

On society, Carse writes:

The prizes won by its citizens can be protected only if the society as a whole remains powerful in relation to other societies. Those who desire the permanence of their prizes will work to sustain the permanence of the whole. Patriotism in one or several of its many forms (chauvinism, racism, sexism, nationalism, regionalism) is an ingredient in all societal play. 
Because power is inherently patriotic, is is characteristic of finite players to seek a growth of power in a society as a way of increasing the power of a society.

Colin Kaepernick refusing to stand for the National Anthem is seen as unpatriotic by many in America, including the wealthy white owners of such teams, which is not surprising, as racism is a form of patriotism, per Carse, and part and parcel of American society when defined as a finite game.

Donald Trump and his large adult sons are proof of just how powerful the inheritance of title and money are in America, and the irony that they are elected by those who feel that successive rounds of finite games have started to be rigged against them is not lost on anyone, not even, I suspect, them. One could argue they need to take a lesson from those oppressed for far longer as to how a turn to nihilism works out in such situations.

Those attacking Affirmative Action want to close off the American experiment and turn it into a series of supposedly level finite games because they have accumulated a healthy lead in this game and wish to preserve it in every form.

White nationalists like Trump all treat America as not just a finite game, but a zero sum finite game. The idea of immigrants being additive to America, to its potential, its output, is to treat America as an infinite game, open-ended. The truth lies, as usual, between the poles, but closer to the latter.

Beware the prophet who comes with stories of zero-sum games, or as Jim Collins once wrote, beware the "tyranny of the or." One of my definitions of leadership is the ability to turn zero-sum into positive sum games.

3. Curb Your Enthusiasm is Running Out of People to Offend

Speaking of fatigue with white male protagonists:

But if Larry David’s casual cruelty mirrors the times more than ever, the show might still fit awkwardly in the current moment. Watching the première of Season 9 on Sunday night, I kept thinking of a popular line from George Costanza, David’s avatar on “Seinfeld”: “You know, we’re living in a society!” Larry, in this first episode of the season, seems to have abandoned society altogether. In the opening shot, the camera sails over a tony swath of L.A., with no people and only a few cars visible amid the manicured lawns and terra-cotta roofs. It descends on Larry’s palatial, ivy-walled house, where he showers alone, singing Mary Poppins’s “A Spoonful of Sugar” and bludgeoning a bottle of soap. (Its dispenser pump is broken—grounds for execution under the David regime.) He’s the master of his domain, yes, but only by default: no one else is around.
“Curb” has always felt insulated, and a lot of its best jokes are borne of the fact that Larry’s immense wealth has warped his world view over the years. (On the most recent season he had no compunction about spending a princely sum on Girl Scout Cookies, only to rescind the order out of spite.) But the beginning of Season 9 offers new degrees of isolation. Like a tech bro ensconced in a hoodie and headphones, Larry seems to have removed himself almost entirely from public life. Both “Curb” and “Seinfeld” like to press the limits of etiquette and social mores, but the latter often tested these on subway cars and buses, in parks or on the street. Much of “Curb,” by contrast, unfolds in a faceless Los Angeles of air-conditioned mansions, organic restaurants, and schmoozy fund-raisers, a long chain of private spaces. The only time Larry encounters a true stranger, it’s in the liminal zone between his car and the lobby of Jeff’s office. She’s a barber on her way to see Jeff at work—even haircuts happen behind closed doors now.

Groundhog Day, one of the great movies, perhaps my favorite Christmas movie of all time, has long been regarded a great Buddhist parable

Groundhog Day is a movie about a bad-enough man—selfish, vain, and insecure—who becomes wise and good through timeless recurrence.

If that is so, then Curb Your Enthusiasm is its dark doppelganger, a parable about the dark secret at the heart of American society, that no person, no matter how selfish, vain, and petty, can suffer the downfall necessary to achieve enlightenment, if he is white and a man. 

In this case, he is a successful white man in Hollywood, Larry David, and each episode of Curb Your Enthusiasm is his own personal Groundhog Day. Whereas Bill Murray wakes up each morning to Sonny and Cher, trapped in Punxsutawney, Pennsylvania, around small town people he dislikes, in a job he feels superior to, Larry David wakes up each morning in his Los Angeles mansion, with rewards seemingly only proportionate to the depths of his pettiness and ill humor. Every episode, he treats all the friends and family around him with little disguised disdain, and yet the next episode, he wakes up in the mansion again.

Whereas Bill Murray eventually realizes the way to break out of his loop is to use it for self-improvement, Larry David seems to be striving to fall from grace by acting increasingly terrible and yet finds himself back in the gentle embrace of his high thread count sheets every morning.

Curb Your Enthusiasm has its moments of brilliance in its minute dissection of the sometimes illogical and perhaps fragile bonds of societal goodwill, and its episode structure is often exceedingly clever, but I can't help watching it now as nothing more than an acerbic piece of performance art, with all the self absorption that implies.

Larry David recently complained about the concept of first world problems, which is humorous, as it's difficult to think of any single person who has done as precise a job educating the world on what they are.

[What about Harvey Weinstein and Louis C.K., you might ask? Aren't they Hollywood royalty toppled from lofty, seemingly untouchable perches? The story of how those happened will be the subject of another post, because the mechanics are so illuminating.]

4. Nathan for You

I am through season 2 of Nathan for You, a Comedy Central show that just wrapped its fourth and final season. We have devalued the term LOL with overuse, but no show has made me literally laugh out loud by myself, on the sofa, as this, though I've grinned in pleasure at certain precise bits of stylistic parody of American Vandal.

Nathan Fielder plays a comedic version of himself. In the opening credits, he proclaims:

My name is Nathan Fielder, and I graduated from one of Canada's top business schools with really good grades [NOTE: as he says this, we see a pan over his transcript, showing largely B's and C's]. Now I'm using my knowledge to help struggling small business owners make it in this competitive world.

If you cringed while watching a show like Borat or Ali G, if you wince a bit when one of the correspondents on The Daily Show went to interview some stooge, you might believe Nathan For You isn't, well, for you. However, the show continues to surprise me.

For one thing, it's a deeply useful reminder of how difficult it is for physical retailers, especially mom and pop entrepreneurs, to generate foot traffic. That they go along with Fielder's schemes is almost tragic, but more instructive.

For another, while almost every entrepreneur is the straight person to Fielder's clown, I find myself heartened by how rarely one of them just turns him away outright. You can see the struggle on each of their faces, as he presents his idea and then stares at them for an uncomfortably long silence, waiting for them to respond. He never breaks character. Should they just laugh at him, or throw him out in disgust? It almost never happens, though one private investigator does chastise Fielder for being a complete loser.

On Curb Your Enthusiasm, Larry David's friends openly call him out for his misanthropy, yet they never abandon him. On Nathan For You, small business owners almost never adopt Fielder's ideas at the end of the trial. However, they almost never call him out as ridiculous. Instead, they try the idea with a healthy dose of good nature at least once, or at least enough to capture an episode's worth of material.

In this age of people screaming at each other over social media, I found this reminder of the inherent decency of people in face to face situations comforting and almost reassuring. Sure, some people are unpleasant both online and in person, and some people are pleasant in person and white supremacists in private.

But some people try to see the best in each other, give others the benefit of the doubt, and on such bonds a civil society are maintained. That this piece of high concept art could not fence in the humanity and real emotion of all the people participating, not even that of Fielder, is a bit of pleasure in this age of eye-rolling cynicism.

[Of course, these small business owners are aware a camera is on them, so the Heisenberg Principle of reality television applies. That a show like this, which depend on the subjects not knowing about the show, lasted four full seasons is a good reminder of how little-watched most cultural products are in this age of infinite content.]

BONUS CONTENT NO ONE ASKED FOR: Here is my Nathan for You idea: you know how headline stand-up comedians don't come on stage to perform until several lesser known and usually much lousier comics are trotted out to warm up the crowd? How, if you attend the live studio taping of a late night talk show like The Daily Show or The Tonight Show, some cheesy comic comes out beforehand to get your laugh muscles loose, your vocal chords primed? And when the headliner finally arrives, it comes as sweet relief?

What if there were an online dating service that provided such a warm-up buffoon for you? That is, when you go on a date, before meeting your date, first the service sends in a stand-in who is dull, awkward, a turn off in every way possible? But a few minutes into what seems to be a disastrous date, you suddenly show up and rescue the proceedings?

It sounds ridiculous, but this is just the sort of idea that Nathan for You would seem to go for. I haven't watched seasons 3 and 4 yet, so if he does end up trying this idea in one of those later episodes, please don't spoil it for me. I won't even be mad that my idea was not an original one, I'll be so happy to see actual footage of it in the field.

5. The aspect ratio of 2:00 to 1 is everywhere

I first read the case for 2:00 to 1 as an aspect ratio when legendary cinematographer Vittorio Storaro advocated for it several years ago. He anticipated a world where most movies would have a longer life viewed on screens at home than in movie theaters, and 2:00 to 1, or Univisium, is halfway between the typical 16:9 HDTV aspect ratio and Panavision, or 2:35 to 1.

So many movies and shows use 2:00 to 1 now, and I really prefer it to 16:9 for most work.

6. Tuning AIs through captchas

Most everyone has probably encountered the new popular captcha which displays a grid of photos and asks you to identify which contain a photo of a store front. I just experienced it recently signing up for HQTrivia. This breed of captcha succeeds the wave of captchas that showed photos of short strings of text or numbers and asked you to type in what you saw, helping to train AIs trying to learn to read them. There are variants of the store front captcha: some ask you to identify vehicles, others to identify street signs, but the speculation is that Google uses these to train the "vision" of its self-driving cars.

AI feels like magic when it works, but underrated is the slow slog to take many AI's from stupid to competent. It's no different than training a human. In the meantime, I'm looking forward to being presented with the captcha that shows two photos, one of a really obese man, the other of five school children, with this question above them: "If you had to run over and kill the people in one of these photos, which would you choose?"

7. It's Mikaela Shiffrin profile season, with this one in Outside and this in the New Yorker

I read Elizabeth Weil's profile of Shiffrin in Outside first:

But the naps: Mikaela not only loves them, she’s fiercely committed to them. Recovery is the most important part of training! And sleep is the most important part of recovery! And to be a champion, you need a steadfast loyalty to even the tiniest and most mundane points. Mikaela will nap on the side of the hill. She will nap at the start of the race. She will wake up in the morning, she tells me after the gym, at her house, while eating some pre-nap pasta, “and the first thought I’ll have is: I cannot wait for my nap today. I don’t care what else happens. I can’t wait to get back in bed.”
Mikaela also will not stay up late, and sometimes she won’t do things in the after­noon, and occasionally this leads to more people flipping out. Most of the time, she trains apart from the rest of the U.S. Ski Team and lives at home with her parents in Vail (during the nine weeks a year she’s not traveling). In the summers, she spends a few weeks in Park City, Utah, training with her teammates at the U.S. Ski and Snowboard Center of Excellence. The dynamic there is, uh, complicated. “Some sports,” Mikaela says, “you see some athletes just walking around the gym, not really doing anything, eating food. They’re first to the lunchroom, never lifting weights.”

By chance, I happened to be reading The Little Book of Talent: 52 Tips for Improving Your Skills by Daniel Coyle, and had just read tips that sounded very familiar to what was mentioned here.

More echoes of Coyle's book in The New Yorker profile:

My presumption was that her excellence was innate. One sometimes thinks of prodigies as embodiments of peculiar genius, uncorrupted by convention, impossible to replicate or reëngineer. But this is not the case with Shiffrin. She’s as stark an example of nurture over nature, of work over talent, as anyone in the world of sports. Her parents committed early on to an incremental process, and clung stubbornly to it. And so Shiffrin became something besides a World Cup hot shot and a quadrennial idol. She became a case study. Most parents, unwittingly or not, present their way of raising kids as the best way, even when the results are mixed, as such results usually are. The Shiffrins are not shy about projecting their example onto the world, but it’s hard to argue with their findings. “The kids with raw athletic talent rarely make it,” Jeff Shiffrin, Mikaela’s father, told me. “What was it Churchill said? Kites fly higher against a headwind.”

So it wasn't a real surprise to finally read this:

The Shiffrins were disciples of the ten-thousand-hours concept; the 2009 Daniel Coyle book “The Talent Code” was scripture. They studied the training methods of the Austrians, Alpine skiing’s priesthood. The Shiffrins wanted to wring as much training as possible out of every minute of the day and every vertical foot of the course. They favored deliberate practice over competition. They considered race days an onerous waste: all the travel, the waiting around, and the emotional stress for two quick runs. They insisted that Shiffrin practice honing her turns even when just skiing from the bottom of the racecourse to the chairlift. Most racers bomb straight down, their nonchalance a badge of honor.

Coyle's book, which I love for its succinct style (it could almost be a tweetstorm if Twitter had slightly longer character limits, each tip is averages one or two paragraphs long), is the books I recommend to all parents who want their kids to be really great at something, and not just sports.

Much of the book is about the importance of practice, and what types of practice are particularly efficient and effective.

Jeff Shiffrin said, “One of the things I learned from the Austrians is: every turn you make, do it right. Don’t get lazy, don’t goof off. Don’t waste any time. If you do, you’ll be retired from racing by the time you get to ten thousand hours.”
“Here’s the thing,” Mikaela told me one day. “You can’t get ten thousand hours of skiing. You spend so much time on the chairlift. My coach did a calculation of how many hours I’ve been on snow. We’d been overestimating. I think we came up with something like eleven total hours of skiing on snow a year. It’s like seven minutes a day. Still, at the age of twenty-two, I’ve probably had more time on snow than most. I always practice, even on the cat tracks or in those interstitial periods. My dad says, ‘Even when you’re just stopping, be sure to do it right, maintaining a good position, with counter-rotational force.’ These are the kinds of things my dad says, and I’m, like, ‘Shut up.’ But if you say it’s seven minutes a day, then consider that thirty seconds that all the others spend just straight-lining from the bottom of the racecourse to the bottom of the lift: I use that part to work on my turns. I’m getting extra minutes. If I don’t, my mom or my coaches will stop me and say something.”

Bill Simmons recently hosted Steve Kerr for a mailbag podcast, and part I is fun to hear Kerr tell stories about Michael Jordan. Like so many greats, Jordan understood that the contest is won in the sweat leading up to the contest, and his legendary competitiveness elevated every practice and scrimmage into gladiatorial combat. As Kerr noted, Jordan single-handedly was a cure for complacency for the Bulls. 

He famously broke down some teammates with such intensity in practice that they were driven from the league entirely (remember Rodney McCray?). Everyone knows he once punched Steve Kerr and left him with a shiner during a heated practice. The Dream Team scrimmage during the lead in to the 1992 Olympics, in which the coaches made Michael Jordan one captain, Magic Johnson the other, is perhaps the single sporting event I most wish had taken place in the age of smartphones and social media.

What struck me about the Shiffrin profiles, something Coyle notes about the greats, is how many of the lives of the great ones are unusually solitary, spent in deliberate practice on their own, apart from teammates. It's obviously amplified for individual sports like tennis and skiing and golf, but even for team sports, the great ones have their own routines. Not only is it lonely at the top, it's often lonely on the way there.

8. The secret tricks hidden inside restaurant menus

Perhaps because I live in the Bay Area, it feels as if the current obsession is with the dark design patterns and effects of social apps. But in the scheme of things, many other fields whose work we interact with daily have many more years of experience designing to human nature. In many ways, people designing social media have a very naive and incomplete view of human nature, but the power of the distribution of ubiquitous smartphone and network effects have elevated them to the forefront of the conversation.

Take a place like Las Vegas. Its entire existence is testament to the fact that the house always wins, yet it could not exist if it could not convince the next sucker to sit down at the table and see the next hand. The decades of research into how best to part a sucker from his wallet makes the volume of research among social media companies look like a joke, even if the latter isn't trivial.

I have a sense that social media companies are similar to where restaurants are with menu design. Every time I sit down at a new restaurant, I love examining the menus and puzzling over all the choices with fellow diners, as if having to sit with me over a meal isn't punishment enough. When the waiter comes and I ask for an overview of the menu, and recommendations, I'm wondering what dishes the entire experience is meant to nudge me to order.

I'm awaiting the advent of digital and eventually holographic or A/R menus to see what experiments we'll see. When will we have menus that are personalized? Based on what you've enjoyed here and other restaurants, we think you'll love this dish. When will we see menus that use algorithmic sorting—these are the most ordered dishes all-time, this week, today? People who ordered this also ordered this? When will see editorial endorsements? "Pete Wells said of this dish in his NYTimes review..."

Not all movies are worth deep study because not all movies are directed with intent. The same applies to menus, but today, enough menus are put through a deliberate design process that it's usually a worthwhile exercise to put them under the magnifying glass. I would love to read some blog that just analyzes various restaurant menus, so if someone starts one, please let me know.

9. Threat of bots and cheating looms as HQ Trivia reaches new popularity heights

When I first checked out HQ Trivia, an iOS live video streaming trivia competition for cash prizes, the number of concurrent viewers playing, displayed on the upper left of the screen, numbered in the hundreds. Now the most popular of games, which occur twice a day, attract over 250K players. In this age where we've seen empires built on exploiting the efficiencies to be gained from shifting so much of social intimacy to asynchronous channels, it's fun to be reminded of the unique fun of synchronous entertainment.

What intrigues me is not how HQ Trivia will make money. The free-to-play game industry is one of the most savvy when it comes to extracting revenue, and even something like podcasts points the way to monetizing popular media with sponsorships, product placement, etc.

What's far more interesting is where the shoulder on the S-curve is. Trivia is a game of skill, and with that comes two longstanding issues. I've answered, at most, 9 questions in a row, and it takes 12 consecutive right answers to win a share of the cash pot. I'm like most people in probably never being able to win any cash.

This is an issue faced by Daily Fantasy Sports, where the word "fantasy" is the most important word. Very soon after they became popular, DFS were overrun by sharks submitting hundreds or thousands of lineups with the aid of computer programs, and some of those sharks worked for the companies themselves. The "fantasy" being sold is that the average person has a chance of winning.

As noted above in my comment about Las Vegas, it's not impossible to sell people on that dream. The most beautiful of cons is one the mark willingly participates in. People participate in negative expected value activities all the time, like the lottery, and carnival games, and often they're aware they'll lose. Some people just participate for the fun of it, and a free-to-play trivia game costs a player nothing other than some time, even if the expected value is close to zero.

A few people have asked me whether that live player count is real, and I'm actually more intrigued by the idea it isn't. Fake it til you make it is one of the most popular refrains of not just Silicon Valley but entrepreneurs everywhere. What if HQ Trivia just posted a phony live player count of 1 million tomorrow? Would their growth accelerate even more than it has recently? What about 10 million? When does the marginal return to every additional player in that count go negative because people feel that there is so much competition it's not worth it? Or is the promise of possibly winning money besides the point? What if the pot scaled commensurate to the number of players; would it become like the lottery? Massive pots but long odds?

The other problem, linked to the element of skill, is cheating. As noted in the article linked above, and in this piece about the spike in Google searches for answers during each of the twice-a-day games, cheating is always a concern in games, especially as the monetary rewards increase. I played the first game when HQ Trivia had a $7,500 cash pot, and the winners each pocketed something like $575 and change. Not a bad payout for something like 10 minutes of fun.

Online poker, daily fantasy sports, all are in constant battle with bots and computer-generated entries. Even sports books at casinos have to wage battle with sharks who try to get around betting caps by sending in all sorts of confederates to put down wagers on their behalf.

I suspect both of these issues will be dampeners on the game's prospects, but more so the issue of skill. I already find myself passing on games when I'm not with others who also play or who I can rope into playing with me. That may be the game's real value, inspiring communal bonding twice a day among people in the same room.

People like to quip that pornography is the tip of the spear when it comes to driving adoption of new technologies, but I'm partial to trivia. It is so elemental and pure a game, with such comically self-explanatory rules, that it is one of the elemental forms or genres of gaming, just like HQ Trivia host Scott Rogowsky is some paragon of a game-show host, mixing just the right balance of cheesiness and snarkiness and effusiveness needed to convince all the players that any additional irony would be unseemly.

10. Raising a teenage daughter

Speaking of Elizabeth Weil, who wrote the Shiffrin profile for Outside, here's another of her pieces, a profile of her daughter Hannah. The twist is that the piece includes annotations by Hannah after the fact.

It is a delight. The form is perfect for revealing the dimensions of their relationship, and that of mothers and teenage daughters everywhere. In the interplay of their words, we sense truer contours of their love, shaped, as they are, by two sets of hands.

[Note, Esquire has long published annotated profiles, you can Google for them, but they are now all locked behind a paywall]

This format makes me question how many more profiles would benefit from allowing the subject of a piece to annotate after the fact. It reveals so much about the limitations of understanding between two people, the unwitting and witting lies at the heart of journalism, and what Janet Malcolm meant, when she wrote, in the classic opening paragraph of her book The Journalist and the Murderer, "Every journalist who is not too stupid or too full of himself to notice what is going on knows that what he does is morally indefensible."

10 browser tabs

1. Love in the Time of Robots

“Is it difficult to play with her?” the father asks. His daughter looks to him, then back at the android. Its mouth begins to open and close slightly, like a dying fish. He laughs. “Is she eating something?”
The girl does not respond. She is patient and obedient and listens closely. But something inside is telling her to resist. 
“Do you feel strange?” her father asks. Even he must admit that the robot is not entirely believable.
Eventually, after a few long minutes, the girl’s breathing grows heavier, and she announces, “I am so tired.” Then she bursts into tears.
That night, in a house in the suburbs, her father uploads the footage to his laptop for posterity. His name is Hiroshi Ishi­guro, and he believes this is the first record of a modern-day android.

Reads like the treatment for a science fiction film, some mashup of Frankenstein, Pygmalion, and Narcissus. One incredible moment after another, and I'll grab just a few excerpts, but the whole thing is worth reading.

But he now wants something more. Twice he has witnessed others have the opportunity, however confusing, to encounter their robot self, and he covets that experience. Besides, his daughter was too young, and the newscaster, though an adult, was, in his words, merely an “ordinary” person: Neither was able to analyze their android encounter like a trained scientist. A true researcher should have his own double. Flashing back to his previous life as a painter, Ishi­guro thinks: This will be another form of self-portrait. He gives the project his initials: Geminoid HI. His mechanical twin.

Warren Ellis, in a recent commencement speech delivered at the University of Essex, said:

Nobody predicted how weird it’s gotten out here.  And I’m a science fiction writer telling you that.  And the other science fiction writers feel the same.  I know some people who specialized in near-future science fiction who’ve just thrown their hands up and gone off to write stories about dragons because nobody can keep up with how quickly everything’s going insane.  It’s always going to feel like being thrown in the deep end, but it’s not always this deep, and I’m sorry for that.

The thing is, far future sci-fi is likely to be even more off base now given how humans are evolving in lock step with the technology around them. So we need more near future sci-fi, of a variety smarter than Black Mirror, to grapple with the implications.

Soon his students begin comparing him to the Geminoid—“Oh, professor, you are getting old,” they tease—and Ishi­guro finds little humor in it. A few years later, at 46, he has another cast of his face made, to reflect his aging, producing a second version of HI. But to repeat this process every few years would be costly and hard on his vanity. Instead, Ishi­guro embraces the logi­cal alternative: to alter his human form to match that of his copy. He opts for a range of cosmetic procedures—laser treatments and the injection of his own blood cells into his face. He also begins watching his diet and lifting weights; he loses about 20 pounds. “I decided not to get old anymore,” says Ishi­guro, whose English is excellent but syntactically imperfect. “Always I am getting younger.”
Remaining twinned with his creation has become a compulsion. “Android has my identity,” he says. “I need to be identical with my android, otherwise I’m going to lose my identity.” I think back to another photo of his first double’s construction: Its robot skull, exposed, is a sickly yellow plastic shell with openings for glassy teeth and eyeballs. When I ask what he was thinking as he watched this replica of his own head being assembled, Ishi­guro says, perhaps only half-joking, “I thought I might have this kind of skull if I removed my face.”
Now he points at me. “Why are you coming here? Because I have created my copy. The work is important; android is important. But you are not interested in myself.”

This should be some science fiction film, only I'm not sure who our great science fiction director is. The best examples may be too old to want to look upon such a story as anything other than grotesque and horrific.

2. Something is wrong on the internet by James Bridle

Of course, some of what's on the internet really is grotesque and horrific. 

Someone or something or some combination of people and things is using YouTube to systematically frighten, traumatise, and abuse children, automatically and at scale, and it forces me to question my own beliefs about the internet, at every level. 

Given how much my nieces love watching product unwrapping and Peppa the Pig videos on YouTube, this story was induced a sense of dread I haven't felt since the last good horror film I watched, which I can't remember anymore since the world has run a DDOS on my emotions.

We often think of a market operating at peak efficiency as sending information back and forth between supply and demand, allowing the creation of goods that satisfy both parties. In the tech industry, the wink-wink version of that is saying that pornography leads the market for any new technology, solving, as it does, the two problems the internet is said to solve better, at scale, than any medium before it: loneliness and boredom.

Bridle's piece, however, finds the dark cul-de-sacs and infected runaway processes which have branched out from the massive marketplace that is YouTube. I decided to follow a Peppa the Pig video on the service and started tapping on Related Videos, like I imagine one of my nieces doing, and quickly wandered into a dark alleyway where I saw some video which I would not want any of them watching. As Bridle did, I won't link to what I found; suffice to say it won't take you long to stumble on some of it if you want, or perhaps even if you don't.

What's particularly disturbing is the somewhat bizarre, inexplicably grotesque nature of some of these video remixes. David Cronenberg is known for his body horror films; these YouTube videos are like some perverse variant of that, playing with popular children's iconography.

Facebook and now Twitter are taking heat for disseminating fake news, and that is certainly a problem worth debating, but with that problem we're talking about adults. Children don't have the capacity to comprehend what they're seeing, and given my belief in the greater effect of sight, sound, and motion, I am even more disturbed by this phenomenon.

A system where it's free to host videos to a global audience, where this type of trademark infringement weaponizes brand signifiers with seeming impunity, married with increasingly scalable content production and remixes using technology, allows for the type of scalable problem we haven't seen before.

The internet has enabled all types of wonderful things at scale; we should not be surprised that it would foster the opposite. But we can, and should, be shocked.

3. FDA approves first blood sugar monitor without finger pricks

This is exciting. One view which seems to be common wisdom these days when it comes to health is that it's easier to lose weight and impact your health through diet than exercise. But one of the problems of the feedback loop in diet (and exercise, actually) is how slow it is. You sneak a few snacks here and there walking by the company cafeteria every day, and a month later you hop on the scale and emit a bloodcurdling scream as you realize you've gained 8 pounds.

A friend of mine had gestational diabetes during one of her pregnancies and got a home blood glucose monitor. You had to prick your finger and draw blood to get your blood glucose reading, but curious, I tried it before and after a BBQ.

To see what various foods did to my blood sugar in near real-time was a real eye-opener. Imagine in the future when one could see what a few french fries and gummy bears did to your blood sugar, or when the reading could be built into something like an Apple Watch, without having to draw blood each time. I don't mind the sight of blood, but I'd prefer not to turn my finger tips into war zones.

Faster feedback might transform dieting into something more akin to deliberate practice. Given that another popular theory of obesity is that it's an insulin phenomenon, tools like this, built for diabetes, might have much mass market impact.

4.  Ingestable ketones

Ingestable ketones have been a recent sort of holy grail for endurance athletes, and now HVMN is bringing one to market. Ketogenic diets are all the rage right now, but for an endurance athlete, the process of being able to fuel oneself on ketones has always sounded like a long and miserable process.

The body generates ketones from fat when low on carbs or from fasting. The theory is that endurance athletes using ketones rather than glycogen from carbs require less oxygen and thus can work out longer.

I first heard about the possibility of exogenous ketones for athletes from Peter Attia. As he said then, perhaps the hardest thing about ingesting exogenous ketones is the horrible taste, which caused him to gag and nearly vomit in his kitchen. It doesn't sound like the taste problem has been solved.

Until we get the pill that renders exercise obsolete, however, I'm curious to give this a try. If you decide to pre-order, you can use my referral code to get $15 off.

5. We Are Nowhere Close to the Limits of Athletic Performance

By comparison, the potential improvements achievable by doping effort are relatively modest. In weightlifting, for example, Mike Israetel, a professor of exercise science at Temple University, has estimated that doping increases weightlifting scores by about 5 to 10 percent. Compare that to the progression in world record bench press weights: 361 pounds in 1898, 363 pounds in 1916, 500 pounds in 1953, 600 pounds in 1967, 667 pounds in 1984, and 730 pounds in 2015. Doping is enough to win any given competition, but it does not stand up against the long-term trend of improving performance that is driven, in part, by genetic outliers. As the population base of weightlifting competitors has increased, outliers further and further out on the tail of the distribution have appeared, driving up world records.
Similarly, Lance Armstrong’s drug-fuelled victory of the 1999 Tour de France gave him a margin of victory over second-place finisher Alex Zulle of 7 minutes, 37 seconds, or about 0.1 percent.3 That pales in comparison to the dramatic secular increase in speeds the Tour has seen over the past half century: Eddy Merckx won the 1971 tour, which was about the same distance as the 1999 tour, in a time 5 percent worse than Zulle’s. Certainly, some of this improvement is due to training methods and better equipment. But much of it is simply due to the sport’s ability to find competitors of ever more exceptional natural ability, further and further out along the tail of what’s possible.

In the Olympics, to take the most celebrated athletic competition, victors are celebrated with videos showing them swimming laps, tossing logs in a Siberian tundra, running through a Kenyan desert. We celebrate the work, the training. Good genes are given narrative short shrift. Perhaps we should show a picture of their DNA, just to give credit where much credit is due?

If I live a normal human lifespan, I expect to live to see special sports leagues and divisions created for athletes who've undergone genetic modification in the future. It will be the return of the freak show at the circus, but this time for real. I've sat courtside and seen people like Lebron James, Giannis Antetokounmpo, Kevin Durant, and Joel Embiid walk by me. They are freaks, but genetic engineering might produce someone who stretch our definition of outlier.

In other words, it is highly unlikely that we have come anywhere close to maximum performance among all the 100 billion humans who have ever lived. (A completely random search process might require the production of something like a googol different individuals!)
But we should be able to accelerate this search greatly through engineering. After all, the agricultural breeding of animals like chickens and cows, which is a kind of directed selection, has easily produced animals that would have been one in a billion among the wild population. Selective breeding of corn plants for oil content of kernels has moved the population by 30 standard deviations in roughly just 100 generations.6 That feat is comparable to finding a maximal human type for a specific athletic event. But direct editing techniques like CRISPR could get us there even faster, producing Bolts beyond Bolt and Shaqs beyond Shaq.

6. Let's set half a percent as the standard for statistical significance

My many-times-over coauthor Dan Benjamin is the lead author on a very interesting short paper "Redefine Statistical Significance." He gathered luminaries from many disciplines to jointly advocate a tightening of the standards for using the words "statistically significant" to results that have less than a half a percent probability of occurring by chance when nothing is really there, rather than all results that—on their face—have less than a 5% probability of occurring by chance. Results with more than a 1/2% probability of occurring by chance could only be called "statistically suggestive" at most. 
In my view, this is a marvelous idea. It could (a) help enormously and (b) can really happen. It can really happen because it is at heart a linguistic rule. Even if rigorously enforced, it just means that editors would force people in papers to say "statistically suggestive for a p of a little less than .05, and only allow the phrase "statistically significant" in a paper if the p value is .005 or less. As a well-defined policy, it is nothing more than that. Everything else is general equilibrium effects.

Given the replication crisis has me doubting almost every piece of conventional wisdom I've inherited in my life, I'm okay with this.

7. We're surprisingly unaware of when our own beliefs change

If you read an article about a controversial issue, do you think you’d realise if it had changed your beliefs? No one knows your own mind like you do – it seems obvious that you would know if your beliefs had shifted. And yet a new paper in The Quarterly Journal of Experimental Psychology suggests that we actually have very poor “metacognitive awareness” of our own belief change, meaning that we will tend to underestimate how much we’ve been swayed by a convincing article.
The researchers Michael Wolfe and Todd Williams at Grand Valley State University said their findings could have implications for the public communication of science. “People may be less willing to meaningfully consider belief inconsistent material if they feel that their beliefs are unlikely to change as a consequence,” they wrote.

Beyond being an interesting result, I link to this as an example of a human readable summary of a research paper. This his how this article summarize the research study and its results:

The researchers recruited over two hundred undergrads across two studies and focused on their beliefs about whether the spanking/smacking of kids is an effective form of discipline. The researchers chose this topic deliberately in the hope the students would be mostly unaware of the relevant research literature, and that they would express a varied range of relatively uncommitted initial beliefs.
The students reported their initial beliefs about whether spanking is an effective way to discipline a child on a scale from “1” completely disbelieve to “9” completely believe. Several weeks later they were given one of two research-based texts to read: each was several pages long and either presented the arguments and data in favour of spanking or against spanking. After this, the students answered some questions to test their comprehension and memory of the text (these measures varied across the two studies). Then the students again scored their belief in whether spanking is effective or not (using the same 9-point scale as before). Finally, the researchers asked them to recall what their belief had been at the start of the study.
The students’ belief about spanking changed when they read a text that argued against their own initial position. Crucially, their memory of their initial belief was shifted in the direction of their new belief – in fact, their memory was closer to their current belief than their original belief. The more their belief had changed, the larger this memory bias tended to be, suggesting the students were relying on their current belief to deduce their initial belief. The memory bias was unrelated to the measures of how well they’d understood or recalled the text, suggesting these factors didn’t play a role in memory of initial belief or awareness of belief change.

Compare this link above to the abstract of the paper itself:

When people change beliefs as a result of reading a text, are they aware of these changes? This question was examined for beliefs about spanking as an effective means of discipline. In two experiments, subjects reported beliefs about spanking effectiveness during a prescreening session. In a subsequent experimental session, subjects read a one-sided text that advocated a belief consistent or inconsistent position on the topic. After reading, subjects reported their current beliefs and attempted to recollect their initial beliefs. Subjects reading a belief inconsistent text were more likely to change their beliefs than those who read a belief consistent text. Recollections of initial beliefs tended to be biased in the direction of subjects’ current beliefs. In addition, the relationship between the belief consistency of the text read and accuracy of belief recollections was mediated by belief change. This belief memory bias was independent of on-line text processing and comprehension measures, and indicates poor metacognitive awareness of belief change.

That's actually one of the better research abstracts you'll read and still it reflects the general opacity of the average research abstract. I'd argue that some of the most important knowledge in the world is locked behind abstruse abstracts.

Why do researchers write this way? Most tell me that researchers write for other researchers, and incomprehensible prose like this impresses their peers. What a tragedy. As my longtime readers know, I'm a firm believer in the power of the form of a message. We continue to underrate that in all aspects of life, from the corporate world to our personal lives, and here, in academia.

Then again, such poor writing keeps people like Malcolm Gladwell busy transforming such insight into breezy reads in The New Yorker and his bestselling books.

8. Social disappointment explains chimpanzees' behaviour in the inequity aversion task

As an example of the above phenomenon, this paper contains an interesting conclusion, but try to parse this abstract:

Chimpanzees’ refusal of less-preferred food when an experimenter has previously provided preferred food to a conspecific has been taken as evidence for a sense of fairness. Here, we present a novel hypothesis—the social disappointment hypothesis—according to which food refusals express chimpanzees' disappointment in the human experimenter for not rewarding them as well as they could have. We tested this hypothesis using a two-by-two design in which food was either distributed by an experimenter or a machine and with a partner present or absent. We found that chimpanzees were more likely to reject food when it was distributed by an experimenter rather than by a machine and that they were not more likely to do so when a partner was present. These results suggest that chimpanzees’ refusal of less-preferred food stems from social disappointment in the experimenter and not from a sense of fairness.

Your average grade school English teacher would slap a failing grade on this butchery of the English language.

9. Metacompetition: Competing Over the Game to be Played

When CDMA-based technologies took off in the US, companies like QualComm that work on that standard prospered; metacompetitions between standards decide the fates of the firms that adopt (or reject) those standards.

When an oil spill raises concerns about the environment, consumers favor businesses with good environmental records; metacompetitions between beliefs determine the criteria we use to evaluate whether a firm is “good.”

If a particular organic foods certification becomes important to consumers, companies with that certification are favored; metacompetitions between certifications determines how the quality of firms is measured.
In all these examples, you could be the very best at what you do, but lose in the metacompetition over what criteria will matter. On the other hand, you may win due to a metacompetition that protects you from fierce rivals who play a different game.
Great leaders pay attention to metacompetition. They advocate the game they play well, promoting criteria on which they measure up. By contrast, many failed leaders work hard at being the best at what they do, only to throw up their hands in dismay when they are not even allowed to compete. These losers cannot understand why they lost, but they have neglected a fundamental responsibility of leadership. It is not enough to play your game well. In every market in every country, alternative “logics” vie for prominence. Before you can win in competition, you must first win the metacompetition over the game being played.

In sports negotiations between owners and players, the owners almost always win the metacompetition game. In the writer's strike in Hollywood in 2007, the writer's guild didn't realize they were losing the metacompetition and thus ended up worse off than before. Amazon surpassed eBay by winning the retail metacompetition (most consumers prefer paying a good, fixed price for a good of some predefined quality than dealing with the multiple axes of complexity of an auction) after first failing at tackling eBay on its direct turf of auctions.

Winning the metacompetition means first being aware of what it is. It's not so easy in a space like, say, social networking, where even some of the winners don't understand what game they're playing.

10. How to be a Stoic

Much of Epictetus’ advice is about not getting angry at slaves. At first, I thought I could skip those parts. But I soon realized that I had the same self-recriminatory and illogical thoughts in my interactions with small-business owners and service professionals. When a cabdriver lied about a route, or a shopkeeper shortchanged me, I felt that it was my fault, for speaking Turkish with an accent, or for being part of an élite. And, if I pretended not to notice these slights, wasn’t I proving that I really was a disengaged, privileged oppressor? Epictetus shook me from these thoughts with this simple exercise: “Starting with things of little value—a bit of spilled oil, a little stolen wine—repeat to yourself: ‘For such a small price, I buy tranquillity.’ ”
Born nearly two thousand years before Darwin and Freud, Epictetus seems to have anticipated a way out of their prisons. The sense of doom and delight that is programmed into the human body? It can be overridden by the mind. The eternal war between subconscious desires and the demands of civilization? It can be won. In the nineteen-fifties, the American psychotherapist Albert Ellis came up with an early form of cognitive-behavioral therapy, based largely on Epictetus’ claim that “it is not events that disturb people, it is their judgments concerning them.” If you practice Stoic philosophy long enough, Epictetus says, you stop being mistaken about what’s good even in your dreams.

The trendiness of stoicism has been around for quite some time now. I found this tab left over from 2016, and I'm sure Tim Ferriss was espousing it long before then, and not to mention the enduring trend that is Buddhism. That meditation and stoicism are so popular in Silicon Valley may be a measure of the complacency of the region; these seem direct antidotes to the most first world of problems. People everywhere complain of the stresses on their mind from the deluge of information they receive for free from apps on the smartphone with processing power that would put previous supercomputers to shame.

Still, given that stoicism was in vogue in Roman times, it seems to have stood the test of time. Since social media seems to have increased the surface area of our social fabric and our exposure to said fabric, perhaps we could all use a bit more stoicism in our lives. I suspect one reason Curb Your Enthusiasm curdles in the mouth more than before is not just that his rich white man's complaints seem particularly ill timed in the current environment but that he is out of touch with the real nature of most people's psychological stressors now. A guy of his age and wealth probably doesn't spend much time on social media, but if he did, he might realize his grievances no longer match those of the average person in either pettiness or peculiarity.

Chasm of comprehension

Last year, Google's AI AlphaGo beat Korean Lee Sedol in Go, a game many expected humans to continue to dominate for years, if not decades, to come.

With the 37th move in the match’s second game, AlphaGo landed a surprise on the right-hand side of the 19-by-19 board that flummoxed even the world’s best Go players, including Lee Sedol. “That’s a very strange move,” said one commentator, himself a nine dan Go player, the highest rank there is. “I thought it was a mistake,” said the other. Lee Sedol, after leaving the match room, took nearly fifteen minutes to formulate a response. Fan Gui—the three-time European Go champion who played AlphaGo during a closed-door match in October, losing five games to none—reacted with incredulity. But then, drawing on his experience with AlphaGo—he has played the machine time and again in the five months since October—Fan Hui saw the beauty in this rather unusual move
Indeed, the move turned the course of the game. AlphaGo went on to win Game Two, and at the post-game press conference, Lee Sedol was in shock. “Yesterday, I was surprised,” he said through an interpreter, referring to his loss in Game One. “But today I am speechless. If you look at the way the game was played, I admit, it was a very clear loss on my part. From the very beginning of the game, there was not a moment in time when I felt that I was leading.”

The first time Gary Kasparov sensed deep intelligence in Deep Blue, he described the computer's move as a very human one

I GOT MY FIRST GLIMPSE OF ARTIFICIAL INTELLIGENCE ON Feb. 10, 1996, at 4:45 p.m. EST, when in the first game of my match with Deep Blue, the computer nudged a pawn forward to a square where it could easily be captured. It was a wonderful and extremely human move. If I had been playing White, I might have offered this pawn sacrifice. It fractured Black's pawn structure and opened up the board. Although there did not appear to be a forced line of play that would allow recovery of the pawn, my instincts told me that with so many "loose" Black pawns and a somewhat exposed Black king, White could probably recover the material, with a better overall position to boot.
But a computer, I thought, would never make such a move. A computer can't "see" the long-term consequences of structural changes in the position or understand how changes in pawn formations may be good or bad.
Humans do this sort of thing all the time. But computers generally calculate each line of play so far as possible within the time allotted. Because chess is a game of virtually limitless possibilities, even a beast like Deep Blue, which can look at more than 100 million positions a second, can go only so deep. When computers reach that point, they evaluate the various resulting positions and select the move leading to the best one. And because computers' primary way of evaluating chess positions is by measuring material superiority, they are notoriously materialistic. If they "understood" the game, they might act differently, but they don't understand.
So I was stunned by this pawn sacrifice. What could it mean? I had played a lot of computers but had never experienced anything like this. I could feel--I could smell--a new kind of intelligence across the table. While I played through the rest of the game as best I could, I was lost; it played beautiful, flawless chess the rest of the way and won easily.

Later, in the Kasparov-Deep Blue rematch that IBM's computer won, again a move in the 2nd game was pivotal. There is debate or whether the move was a mistake or intentional on the part of the computer, but it flummoxed Kasparov (italics mine):

'I was not in the mood of playing at all,'' he said, adding that after Game 5 on Saturday, he had become so dispirited that he felt the match was already over. Asked why, he said: ''I'm a human being. When I see something that is well beyond my understanding, I'm afraid.''

At the news conference after the game, a dark-eyed and brooding champion said that his problems began after the second game, won by Deep Blue after Mr. Kasparov had resigned what was eventually shown to be a drawn position. Mr. Kasparov said he had missed the draw because the computer had played so brilliantly that he thought it would have obviated the possibility of the draw known as perpetual check.

''I do not understand how the most powerful chess machine in the world could not see simple perpetual check,'' he said. He added he was frustrated by I.B.M.'s resistance to allowing him to see the printouts of the computer's thought processes so he could understand how it made its decisions, and implied again that there was some untoward behavior by the Deep Blue team.

Asked if he was accusing I.B.M. of cheating, he said: ''I have no idea what's happening behind the curtain. Maybe it was an outstanding accomplishment by the computer. But I don't think this machine is unbeatable.''

Mr. Kasparov, who defeated a predecessor of Deep Blue a year ago, won the first game of this year's match, but it was his last triumph, a signal that the computer's pattern of thought had eluded him. He couldn't figure out what its weaknesses were, or if he did, how to exploit them.

Legend has it that a move in Game One and another in Game Two were actually just programming glitches that caused Deep Blue to make random moves that threw Kasparov off, but regardless, the theme is the same: at some point he no longer understood what the program was doing. He no longer had a working mental model, like material advantage, for his computer opponent.

This year, a new version of AlphaGo was unleashed on the world: AlphaGo Zero.

As many will remember, AlphaGo—a program that used machine learning to master Go—decimated world champion Ke Jie earlier this year. Then, the program’s creators at Google’s DeepMind let the program continue to train by playing millions of games against itself. In a paper published in Nature earlier this week, DeepMind revealed that a new version of AlphaGo (which they christened AlphaGo Zero) picked up Go from scratch, without studying any human games at all. AlphaGo Zero took a mere three days to reach the point where it was pitted against an older version of itself and won 100 games to zero.


That AlphaGo Zero had nothing to learn from playing the world's best humans, and that it trounced its artificial parent 100-0, is evolutionary velocity of a majesty not seen since the ectomorphs in the Alien movie franchise. It is also, in its arrogance, terrifying.

DeepMind released 55 games that a previous version of AlphaGo played against itself for Go players around the world to analyze.

Since May, experts have been painstakingly analyzing the 55 machine-versus-machine games. And their descriptions of AlphaGo’s moves often seem to keep circling back to the same several words: Amazing. Strange. Alien.
“They’re how I imagine games from far in the future,” Shi Yue, a top Go player from China, has told the press. A Go enthusiast named Jonathan Hop who’s been reviewing the games on YouTube calls the AlphaGo-versus-AlphaGo face-offs “Go from an alternate dimension.” From all accounts, one gets the sense that an alien civilization has dropped a cryptic guidebook in our midst: a manual that’s brilliant—or at least, the parts of it we can understand.

Some moves AlphaGo likes to make against its clone are downright incomprehensible, even to the world’s best players. (These tend to happen early on in the games—probably because that phase is already mysterious, being farthest away from any final game outcome.) One opening move in Game One has many players stumped. Says Redmond, “I think a natural reaction (and the reaction I’m mostly seeing) is that they just sort of give up, and sort of throw their hands up in the opening. Because it’s so hard to try to attach a story about what AlphaGo is doing. You have to be ready to deny a lot of the things that we’ve believed and that have worked for us.”


Like others, Redmond notes that the games somehow feel “alien.” “There’s some inhuman element in the way AlphaGo plays,” he says, “which makes it very difficult for us to just even sort of get into the game.”

Ke Jie, the Chinese Go master who was defeated by AlphaGo earlier this year, said:

Last year, it was still quite humanlike when it played. But this year, it became like a god of Go.”

After his defeat, Ke posted what might be the most poetic and bracing quote of 2017 on Weibo (I first saw it in the WSJ):

“I would go as far as to say not a single human has touched the edge of the truth of Go.”


When Josh Brown died in his Tesla after driving under a semi, it kicked off a months long investigation into who was at fault. Ultimately, the NHTSA absolved Autopilot of blame. The driver was said to have had 7 seconds to see the semi and apply the brakes but was suspected of watching a movie while the car was in Autopilot.

In this instance, it appeared enough evidence could be gathered to make such a determination. In the future, diagnosing why Autopilot or other self-driving algorithms made certain choices will likely only become more and more challenging as the algorithms rise in complexity.

At times, when I have my Tesla in Autopilot mode, the car will do something bizarre and I'll take over. For example, if I drive to work out of San Francisco, I have to exit left and merge onto the 101 using a ramp that arcs to the left almost 90 degrees. There are two lanes on that ramp, but even if I start in the far left lane and am following a car in front of me my car always seems to try to slide over to the right lane.

Why does it do that? My only mental model is the one I know, which is my own method for driving. I look at the road, look for lane markings and other cars, and turn a steering wheel to stay in a safe zone in my lane. But thinking that my car drives using that exact process says more about my limited imagination than anything else because Autopilot doesn't drive the way humans do. This becomes evident when you look at videos showing how a self-driving car "sees" the road.

When I worked at Flipboard, we moved to a home feed that tried to select articles for users based on machine learning. That algorithm continued to be to tweaked and evolved over time, trying to optimize for engagement. Some of that tweaking was done by humans, but a lot of it was done by ML.

At times, people would ask why a certain article had been selected for them? Was it because they had once read a piece on astronomy? Dwelled for a few seconds on a headline about NASA? By that point, the algorithm was so complex it was impossible to really offer an explanation that made intuitive sense to a human, there were so many features and interactions in play.

As more of the world comes to rely on artificial intelligence, and as AI makes great advances, we will walk to the edge of a chasm of comprehension. We've long thought that artificial intelligence might surpass us eventually by thinking like us, but better. But the more likely scenario, as recent developments have shown us, is that the most powerful AI may not think like us at all, and we, with our human brains, may never understand how they think. Like an ant that cannot understand a bit about what the human towering above them is thinking, we will gaze into our AI in blank incomprehension. We will gaze into the void. The limit to our ability to comprehend another intelligence is our ability to describe its workings, and that asymptote is drawn by the limits of our brain, which largely analogizes all forms of intelligence to itself in a form of unwitting intellectual narcissism.

This is part of the general trend of increasing abstraction that marks modern life, but it is different than not knowing how a laptop is made, or how to sew a shirt for oneself. We take solace in knowing that someone out there can. To admit that it's not clear to any human alive how an AI made a particular decision feels less like a ¯\_(ツ)_/¯ and more like the end of some innocence.

I suspect we'll continue to tolerate that level of abstraction when technology functions as we want it to, but we'll bang our heads in frustration when it doesn't. Like the annoyance we feel when we reach the limits of our ability to answer a young child who keeps asking us "Why?" in recursive succession, this frustration will cut deep because it will be indistinguishable from humiliation.

Learning curves sloping up and down

One of the great inefficiencies of humanity as a species is the need to re-educate every successive generation. I think of this when playing with my nieces and nephews and my friends' children. Adults have to spend so much time teaching infants and children things we've already learned, and the process of knowledge transfer is so lossy. The entire education system can be seen as a giant institution for transferring knowledge from one generation to the next, like some crude disk drive, and these days it's rising in price despite not measurably improving.

Artificial intelligences need not go through this because they don't die abruptly like humans, they can evolve continuously without hard resets. This is one of its chief advantages over human intelligence. To take a modern example, self-driving cars should only improve from here on out, and each new one we build can be as smart as the smartest self-driving car as soon as it's assembled. Every node on the network has access to the intelligence of the network.

All this human intelligence cut short by mortality is a curse, but given human nature, it is also critical to forward progress. People's views calcify, so death is a way of wiping the slate clean to make way for ideological progress. Part of why racism and sexism, to take two social ills, decline over time is simply that the racists and sexists die out.

This plays out at a corporate level, too. Companies can have both too long and too short a memory. New employees have to be taught the culture and catch up to what others before them learned so they can be as productive as possible. On the other hand, institutions can become set in their ways, less adaptive as their environments evolve. New blood can bring fresh eyes.

One form of this is institutional trauma. A company tries to enter a space, fails, and doesn't venture into that space ever again, even if the timing for entry shifts to a more favorable one. I look at a product like Google Wave and think that if Google had stuck with it, they might have built something like Slack.

Why do companies slow down as they grow larger? One reason is that in a hierarchical organizational structure, the more people and more levels you pile in, the more chances someone somewhere will say no to any idea. Bureaucracy is just institutionalized veto power growing linearly with organizational size.

One theory for why evolution gives us just enough of a lifespan to bear offspring but not stay around too long is that it reduces competition for resources for our offspring. Old timers who rise in an organization can compete for resources with new employees, but without the disadvantage of old age. Most who survive at a company have risen to the level where they have disproportionate institutional power. It's often deserved, but it's also dangerous. True disruption of a company is difficult to counter because it attacks the strongest part of your business, and that division or unit tends to be the one that has the most power in the organization.

Companies try to counter this by dividing themselves into smaller units even as they grow in the aggregate. Jeff Bezos tried localizing decision-making power at Amazon in what he called two-pizza teams (the size of the team being one that could be fed by two pizzas). Facebook acquires companies like Instagram and WhatsApp but lets them run largely independently. Google's new Alphabet org structure breaks itself into a looser coalition of entities where each division has more degrees of freedom strategically. All are attempt to keep the weight of bureaucratic middle management off of the creatives, to preserve greater dimensionality and optionality throughout the organization.

Amazon is one company which often wins just by being more patient than its competitors, playing games on a much longer time scale than most. It tends to be less susceptible to institutional trauma than most. Of course, part of this is the result of the unique ownership structure that companies like Amazon, Facebook, and Google have managed to pull off: ultimate decision-making power rests in the hands of the founders even as they leverage the benefits of the public market.

However, it's more than that. When Bezos was asked at Recode last year how he decided when to give up on a project, he said something striking: we give up on something when the last high judgment person in the room gives up on it.

What a brilliant heuristic. Simple and memorable. Of course, deciding who is high judgment is its own challenge, but this concept reverses the usual problem of bureaucracy, which is it takes only one person saying no to kill something. Jeff reverses that; he wants the company to be as smart on any topic as its single smartest person. 

At some point in life, it probably is rational to be that old dog who eschews new tricks. If you're going to die soon anyhow, you're more likely to just suffer the discomfort of having to adjust and then die before you can reap any awards. The corporate version of this is the concept that most executives should just squeeze the maximum profit out of their existing thinking and not bother trying to stave off disruption. It might be more energy and resource efficient to just have some of the stalwarts die off rather than shift the thinking of tens of thousands of employees, or change a culture which has evolved over decades.

Magic iPod

Everyone has been passing around the Magic iPod this month. First Deep Blue beat Kasparov, then AlphaGo beat Lee Se-dol, and now we have Magic iPod taking down Girl Talk.

When you read stories about how artists come up with mashups (finding works with compatible BPM and keys, among other things), or how the Swedish pop factory mad scientists like Max Martin conjure pop hits, it seems inevitable that in our lifetime we'll have algorithms creating real pop hits.

How such work is received by a human audience is about more than its intrinsic qualities, however. In an objective competition like a game of Go, or when considering a mashup which is simply the synthesis of existing creative works, I suspect humans will be comfortable with acknowledging the achievements of an algorithm.

With original creative works, however, like music, novels, movies, I suspect humans will recoil from even intrinsically appealing creation if it was written by a computer program. Call it some variant of the uncanny valley effect.

We have a romantic attachment to human creation, and it may take a generation of people passing on before we overcome that cultural aversion. When a waiter places a beautiful dish in front of you at a restaurant, we like to imagine that a chef toiled over the plate in the kitchen, conjuring that beautiful, delicious entree from raw ingredients, fire, and ingenuity. When we read an engrossing novel, we picture a tortured writer banging on an old typewriter in a cabin by the sea, stopping from time to time to put out a cigarette and gaze out the window at the ocean waves trying to claw up the gentle slope of the beach.

When Beyonce drops Lemonade or any one of her jaw dropping awards show performances on an unprepared world, I like to believe the work was birthed from what is surely a vagina with mystic powers, belonging as it does to our modern icon of feminism and black empowerment.

It's not quite as appealing if the truth was that an algorithm finished processing in some computer lab somewhere. A progress bar on a monitor finally reaches 100%, and a file is deposited into a directory.

That's why if humans ever comes up with algorithms that are capable of creating popular works of culture, it's financially wise for the creators to claim the credit themselves, at least until many years of critical and popular embrace have accumulated. Then, and only then, spring the truth on the world.

We live in a Skinner box, and it was of our own making.


For the past few months, a “very large fraction” of the millions of queries a second that people type into the company’s search engine have been interpreted by an artificial intelligence system, nicknamed RankBrain, said Greg Corrado, a senior research scientist with the company, outlining for the first time the emerging role of AI in search.
RankBrain uses artificial intelligence to embed vast amounts of written language into mathematical entities -- called vectors -- that the computer can understand. If RankBrain sees a word or phrase it isn’t familiar with, the machine can make a guess as to what words or phrases might have a similar meaning and filter the result accordingly, making it more effective at handling never-before-seen search queries.
RankBrain is one of the “hundreds” of signals that go into an algorithm that determines what results appear on a Google search page and where they are ranked, Corrado said. In the few months it has been deployed, RankBrain has become the third-most important signal contributing to the result of a search query, he said.
So far, RankBrain is living up to its AI hype. Google search engineers, who spend their days crafting the algorithms that underpin the search software, were asked to eyeball some pages and guess which they thought Google’s search engine technology would rank on top. While the humans guessed correctly 70 percent of the time, RankBrain had an 80 percent success rate.

More on RankBrain here.

Machine learning is advancing fast. At its best it feels a bit like magic, and it's endlessly malleable. Think it's missing something of importance? Add it as a factor, or tune it up.

I suspect in my lifetime we'll have machine learning so good it will be largely incomprehensible to me. That is, it won't be understandable by using analogies to how humans think because it will be its own form of intelligence.

A cookbook from IBM's Watson

Robots taking all the jobs, cooking edition:

Steve Abrams, the director of IBM’s Watson Life research program, told Quartz that Watson scanned publicly available data sources to build up a vast library of information on recipes, the chemical compounds in food, and common pairings. (For any budding gastronomers out there, Abrams said Wikia was a surprisingly useful source.) Knowledge that might’ve taken a lifetime for a Michelin-starred chef to attain can now be accessed instantly from your tablet.
What separates Watson from the average computer (or chef) is its ability to find patterns in vast amounts of data. It’s essentially able figure out, through sheer repetition, what combinations of compounds and cuisines work together. This leads to unusual pairings, like Waton’s apple kebab dish, which has some odd ingredients: “Strawberries and mushrooms share a lot of flavor compounds,” Abrams said. “It turns out they go quite well together.”

The researchers are publishing a cookbook with recipe ideas from Watson, and it releases this Tuesday: Cognitive Cooking with Chef Watson: Recipes for Innovation from IBM & the Institute of Culinary Education. I have not read the book, but some of the recipes sound intriguing (“Belgian bacon pudding, a desert containing dried porcini mushrooms”) while others sound, at best, like clever wordplay (“the shrimp cocktail, which is a beverage with actual shrimp in it”). Regardless, I'm purchasing a copy just out of sheer curiosity. Let's hope they turn this resource into an app or service instead of a book, I blame Watson's vanity for wanting this in the outdated format of a book.

To the extent that standout recipes and flavor pairings are a matter of pattern recognition, there's no reason a computer, with its infinitely more scalable hardware and software for that purpose, couldn't match or exceed a human. And, so, a variant of the infinite monkey theorem: given enough time, a computer will write the French Laundry cookbook (and win a third Michelin star).

To be clear, I'm okay with this. I just want to eat tasty food, I'm fine with employing computers to come up with more amazing things to feed me.

For now, however, the computer still requires a human to actually prepare the recipe. In a true demonstration of how far artificial intelligence has progressed, no sufficiently advanced computer wants the drudgery of life as a line chef. Better profits in cookbooks than restaurants anyway.

A new cooking show concept already comes to mind: Top Freestyle Chef. Like freestyle chess, in freestyle cooking competitors would consist of a human or a human consulting with a computer. I am ready to program this into my DVR already, as long as they don't replace Padma Lakshmi with a robot host. I'm as big a fan of artificial intelligence and robots as the next guy, but I think we're a long way from replacing this.