A mental heavyweight championship bout

Chess is an all-consuming focus. “Sometimes when people are talking to me I will suddenly remember some chess position, and then it’s very hard for me to concentrate on what they are saying. They can see in my eyes that I am drifting away.” Yet contrary to his good-natured image, he admits to a steelier side. “What you can concede outside the chessboard will eventually haunt you in the chessboard as well. A match is really a contest of space between two people, and you can’t give the other one any quarter.”

-----

He describes the preparation process as akin to plotting an ambush in a giant forest. The terrain is too vast to comprehend in its entirety, he says. “But there are areas that you will know better than your opponent”, and that is where you prepare to attack, aided in your preparation by the most important change to sweep through the game of chess in decades: computers. “The way people play chess nowadays, which is to keep on switching their openings, being much more opportunistic – I think that is a direct result of computers. Even the way people play tournaments – everything has changed.”

Top competitors who once relied on particular styles of play are now forced to mix up their strategies, for fear that powerful analysis engines will be used to reveal fatal weaknesses in favoured openings. The result has in some ways made chess more defensive, increasing the risks of daring, adventurous gambits. But in championship matches, where draws are common and the final result is likely to be decided by just a handful of victories, unexpected approaches become even more prized. “Anything unusual that you can produce has quadruple, quintuple the value, precisely because your opponent is likely to do the predictable stuff, which is on a computer,” Anand says.

From an interview with chess grandmaster Viswanathan Anand a few days before his World Championship match against the heavily favored young prodigy Magnus Carlsen. Some interesting material throughout about the impact of computers on how chess is played at the highest levels, but even more insight into the stress of such prolonged mental combat.

I predict increasing novelty appeal for contests in which humans do things that computers can do better.

Art created by and for computers

But the Internet, with its swift proliferation of memes, is producing more extreme forms of modernism than modernism ever dreamed of.

These are the ideas of the Canadian media scholar Darren Wershler, who has been making some unexpected connections between meme culture and contemporary poetry. “These artifacts,” Wershler claims, “aren’t conceived of as poems; they aren’t produced by people who identify as poets; they circulate promiscuously, sometimes under anonymous conditions; and they aren’t encountered by interpretive communities that identify them as literary.” Examples include a Nigerian e-mail scammer who writes out the entire “Harry Potter and the Chamber of Secrets” in longhand, a data engineer who renders the entire text of Moby Dick into emojicons, and a library scientist who converts “Ulysses” into Q.R. bar codes.

Wershler calls these activities “conceptualism in the wild,” referring to the aspect of nineteen-sixties conceptual art that concerned reframing, and thereby redefining, the idea of artistic genius (think of Duchamp’s urinal). Conceptual projects of the period were generated by a kind of pre-Internet O.C.D., such as Sol LeWitt’s exhaustive photographic documentation of every object, nook, and cranny in his Manhattan loft, or Tehching Hsieh’s yearlong practice of taking a photo of himself every hour, on the hour. Today’s conceptualists in the wild make those guys look tame. It’s not uncommon to see blogs that recount someone’s every sneeze since 2007, or of a man who shoots exactly one second of video every day and strings the clips together in time-lapsed mashups. There is guy who secretly taped all of his conversations for three years and a woman who documents every morsel of food that she puts into her mouth. While some of these people aren’t consciously framing their activities as works of art, Wershler argues that what they’re doing is so close to the practices of sixties conceptualism that the connection between the two can’t be ignored.

More from the New Yorker here. The article notes that much of this new conceptual poetry is published on LuLu. In 2007, just after we came out with the name Hulu, we were sued by LuLu because, well, I guess our company names rhymed. That lawsuit certainly seemed like some form of modernist joke.

Certainly the internet and the rise of computing has lowered the bar to creating conceptual art of staggering size. The economics of many types of conceptual art has changed, and that has shifted its appeal. Emoji Dick was created through a crowdsourced labor pool of over 800 people from Amazon's Mechanical Turk web service. The means by which it was produced is more interesting than the work itself, almost like David Blaine holding his breath for 17 minutes. In another age, perhaps a single person would have produced Emoji Dick and we'd marvel at the obsessive persistence and endurance. Now we chuckle at the idea of leveraging an Amazon Web Service to crowdsource the entire effort.

I loved this paragraph from the New Yorker article. 

Fifty years ago, when Andy Warhol said things like “I want to be a machine,” and “It’s easier not to care,” he was romanticizing the formal and emotional cleanliness of machine-based production. Humans, after all, court messiness. Warhol’s salvo was extended by the poet Christian Bök, who, in 2009, claimed, “We are probably the first generation of poets who can reasonably expect to write literature for a audience of artificially intellectual peers …. If poetry already lacks any meaningful readership among our own anthropoid population, what have we to lose by writing poetry for a robotic culture that must inevitably succeed our own? If we want to commit an act of poetic innovation in an era of formal exhaustion, we may have to consider this heretofore unimagined, but nevertheless prohibited, option: writing poetry for inhuman readers, who do not yet exist, because such aliens, clones, or robots have not yet evolved to read it.”

 

Indeed, much of this modern conceptual art is unreadable, but what if that's because it's only meant to be read by other computers? Someday, when a computer reads this blog post, I hope it has a sense of humor and irony.

Hello computer crawler, hope you enjoyed this post, and if you did, please share it to whatever social network is popular among you robot folk!

Google Autocomplete knows

"Who knows what evil lurks in the hearts of men? The Shadow knows." 

The modern equivalent, one might say, is Google Autocomplete. One ad campaign capitalizes on that by using Google Autocomplete suggestions to highlight gender inequality. 

More of the images can be seen here. 

I love Google Autocomplete, it is like some modern crowdsourced, mass-distributed free art installation. I have spent hours putting random queries in there just to plumb the collective mind of humanity.

Economics of interstellar trade

Dr Krugman, a science-fiction fan, ponders how trade might work between two widely separated planets, Earth and Trantor. Such trade will be affected by relativity theory, which shows that beings on Earth (or Trantor) will see time pass at a different speed from those who are on board cargo ships moving between the two. This could make it hard to calculate the net present value of a shipment. And the fact that messages can move at best at the speed of light (and cargoes more slowly still) might do odd things to the ability to arbitrage between the economies of the two worlds.

After working through the maths, Dr Krugman came up with two fundamental theorems of interstellar trade. The first is that interest costs on travelling goods should be calculated using clocks on planets, not ships. This is because the opportunity cost of trade—buying a bond on Earth (or Trantor), say—is calculated using planet-bound clocks, regardless of what relativity does to a businessman travelling alongside his cargo.

The second theorem states that, though long travel times mean prices on trading planets will never reach parity, interest rates will. If they differed, then investors could buy bonds on the more attractive planet, driving its rates back to parity with those on its trading partner.

More here, hat tip to Tyler Cowen. I would love to take a course on robot economics, which might have more near-term value in my life, but interstellar economics would be a honeypot for me, too.

Amazon and the "profitless business model" fallacy

[DISCLOSURE: As always when I write about Amazon, I'll note I worked there from 1997-2004 and that I still own some shares in the company. I still have many friends who work there, though I have no more idea what Amazon is working on now than any of you in the public.] 

With every quarterly earnings call, my Twitter feed lights up with jokes about how Amazon continues to grow its revenue and make no profits and how trusting investors continue to rewards the company for it. The apotheosis of that line of thoughts is a quote from Slate's Matthew Yglesias earlier this year: "Amazon, as best I can tell, is a charitable organization being run by elements of the investment community for the benefit of consumers."

It's a great quote, one that got so much play Amazon even featured it in its Annual Letter to Shareholders. But like much of the commentary about Amazon, it's a misreading of Amazon's business model.

Amazon is a classic fixed cost business model, it uses the internet to get maximum leverage out of its fixed assets, and once it achieves enough volume of sales, the sum total of profits from all those sales exceed its fixed cost base, and it turns a profit. It already has exceeded this hurdle in its past.

I'm fairly certain most of Amazon's retail businesses remain quite profitable. Some may not be, but they help to reinforce Amazon as the retail site of first resort. By the time I left Amazon in 2004 many of its retail businesses were already spinning off healthy profits. It is much harder to tell now from the outside because Amazon doesn't present a full P&L by business line to the outside world. You can see revenue by broad categories of the business, but most of its costs are lumped together in one giant blob on the income statement.

Very early in my career at Amazon, we could already easily model out and see when our revenue would give us enough income to exceed our fixed cost base. We could adjust when that would happen by choosing to invest more or less aggressively, but given our growth rates, it was always just felt like a matter of when, not if, we'd turn a profit.  Besides, we were most obsessed with free cash flow. Most armchair analysts love to dissect gross margin and net income because those are simpler to understand and easier to compute from public financial statements, but there are many problems with just looking at gross margin that any analyst worth their paycheck should understand.

Does Amazon lose money on sales of some individual items? For sure. The first Kindle ebooks that were priced at $9.99 when Amazon had to pay more than that per copy to publisher were one example. Giant, heavy electronics items that Amazon sometimes ships for free when the shipping cost is clearly non-trivial and cost more than the usual thin margins on such goods are another.

But while such exceptions in the catalog make for great copy (it's fun to link to such items in your story and let users see the evidence firsthand, especially when the item is some strange piece of machinery that maybe a handful of people in the world would ever order), but don't be mistaken. The vast vast majority of products Amazon sells it makes a profit on. Over time, more of these products that inadvertently sell at a loss will be corrected so that no longer happens, and what remains will be products Amazon intentionally uses as loss leaders.

The platform of Amazon is profitable, too. When other people sell products on Amazon Marketplace the gross margin is huge. I sell a used book on Amazon, it takes a cut of the transaction, I am the one packing and shipping that item to the buyer. You don't have to be a financial whiz to understand the cost of that transaction to Amazon is minimal. 

If Amazon has so many businesses that do make a profit, then why is it still showing quarterly losses, and why has even free cash flow decreased in recent years?

Because Amazon has boundless ambition. It wants to eat global retail. This is one area where the press and pundits accept Amazon's statements at face value.

Given that giant mission, Amazon has decided to continue to invest to arm itself for a much larger scale of business. If it were purely a software business, its fixed cost investments for this journey would be lower, but the amount of capital required to grow a business that has to ship millions of packages to customers all over the world quickly is something only a handful of companies in the world could even afford. Joey Chestnut doesn't just wake up one day and win the Coney Island hot dog eating contest every year, he has to spend months of training to prepare his digestive system for the feat. 

Amazon has seen that lowering its shipping costs and increasing the speed of shipping items to customers is like a shot of adrenaline to customer's propensity to buy from them, and so it has doubled down on building more and more fulfillment centers around the world. When I joined Amazon it had one fulfillment center. Today it has dozens just in the US alone, and I would not be surprised if it has more than 100 fulfillment centers worldwide now. 

That is a gargantuan investment, billions of dollars worth, and it takes a significant bite out of Amazon's free cash flow. Add in its investments in infrastructure to support a growing AWS client base, and Amazon has again hiked its fixed cost base to a higher plateau. But for Amazon this is nothing new, it's just the same typeface bolded. 

I'm convinced Amazon could easily turn a quarterly profit now. Many times in its history, it could have been content to stop investing in new product lines, new fulfillment centers, new countries. The fixed cost base would flatten out, its sales would continue growing for some period of time and then flatten out, and it would harvest some annuity of profits. Even the first year I joined Amazon in 1997, when it was just a domestic book business, it could have been content to rest on its laurels.

But Jeff is not wired that way. There are very few people in technology and business who are what I'd call apex predators. Jeff is one of them, the most patient and intelligent one I've met in my life. An apex predator doesn't wake up one day and decide it is done hunting. Right now I envision only one throttle to Jeff's ambitions and it is human mortality, but I would not be surprised if one day he announced he'd started another side project with Peter Thiel to work on a method of achieving immortality.

One popular thesis among Amazon profitability skeptics is that Amazon can't "flip a switch" and become profitable. The most common guess as to how Amazon flips the switch is that it will wait until it is the last retailer standing and then raise prices across the board, so Amazon skeptics argue against that narrative possibility.

But "flipping a switch" is the wrong analogy because Amazon's core business model does generate a profit with most every transaction at its current price level. The reason it isn't showing a profit is because it's undertaken a massive investment to support an even larger sales base. How does Amazon turn a profit? Not by flipping a switch but by waiting, once again, until its transaction volume grows and income exceeds its fixed cost base again. It can choose to reach that point faster or slower depending on how quickly it continues to grow its fixed cost base, but a simple way to accelerate that would be to stop investing in so many new fulfillment centers. 

One argument against Amazon is that it is investing for a revenue volume that will never come. That's a different argument, to me, than saying its business model isn't profitable. And even on that point, you can chart its quarterly revenue for yourself and tell me if it looks like it's flattening out.  Though it has not always been on an exact upward sloping curve (we can expect the curve's slope to adjust up or down as various lines of its retail business mature or accelerate depending on Amazon's market share and traction in each line), the long term arc bends up like the corner of a world-dominating smile.

Part of this problem comes from the limited visibility into the dynamics of its business finances. Why doesn't Amazon break out more detail in its financial reporting to help the external world understand all these intricacies? How many subscribers to Amazon Prime, how many Kindles have sold, what's the net income from different lines of business, how much of its asset base investment is for fulfillment centers versus technology infrastructure for AWS? Why doesn't Google break out its lines of business in more detail in its financials? Why doesn't Apple reveal more detail about exact sales of the various models of its iPads and iPhones?

Tech companies, in general, have dealt with the press, investors, and public long enough now to have decided that for the most part, disclosing less buys them the most strategic flexibility with the least amount of pain. Tech companies have an interesting ambivalence towards the public capital markets. They rebel against resource dependence theory because they don't believe their investors know how to run their businesses better than they do, but on the other hand, being public is a great boon to compensating knowledge workers who have a lot of job options.

Based on its stock price, more investors seem willing to buy Amazon's business story that they'll be able to replicate their historical playbook on a larger scale. 

You could argue that a business that has to invest so much of its free cash flow to grow is inherently a profitless business model. However, Amazon has known from its very earliest days that selling commodities, the core of Amazon's business, is inherently a low margin scale business. It won't ever approach the margins of, say, Apple's hardware business.

But to me, a profitless business model is one in which it costs you $2 to make a glass of lemonade but you have to sell it for $1 a glass at your lemonade stand. But if you sell a glass of lemonade for $2 and it only costs you $1 to make it, and you decide business is so great you're going to build a lemonade stand on every street corner in the world so you can eventually afford to move humanity into outer space or buy a newspaper in your spare time, and that requires you to invest all your profits in buying up some lemon fields and timber to set up lemonade franchises on every street corner, that sounds like a many things to me, but it doesn't sound like a charitable organization.

Some people get it.  When I had most of this post written, I started searching for articles analyzing Amazon's business model, and I found this fantastic post by Benedict Evans which already states much of what I've written above. He understands Amazon to be a portfolio of businesses of varied maturity. But Evans is the exception, and so you can continue to expect a torrent of jokes each time Amazon releases its earnings and shows revenue growth but a negative net income. I'd love to see more external analysis of Amazon begin to focus on trying to break down its various investments in more detail and less time spent arguing whether its basic business model is profitable. Does the world need another story marvelling at how much Jeff can invest in his business? Is it that difficult to fathom that investing to try to be the largest retailer in the history of the world takes billions of dollars in investment?

The irony of all this is that while Amazon's public financial statements make it extremely difficult to parse out its various businesses, it is extremely forthright and honest about its business plans and strategy. It's the reason Jeff continues to reprint its first ever letter to shareholders from 1997 in its annual report every year. The plan is right there before our eyes, but so many continue to refuse to take it at face value. As a reporter, it must be so boring to parrot the same thing from Jeff and his team year after year, so different narratives must be spun when the overall plan has not changed. 

Take this most recent article in The Atlantic, from Derek Thompson. It's a good read, comparing Amazon to Sears, but it's also a great example of how Amazon's basic strategy is always couched the same way, with a general veneer of skepticism.

At least, that’s the vision. Defenders say Amazon is trading the present for the future, spending all its revenue on a global scatter plot of warehouses that will make the company indomitable. Eventually, the theory goes, investors expect Amazon to complete its construction project and, having swayed enough customers and destroyed enough rivals, to “flip the switch,” raising prices and profits greatly. In the meantime, they’re happy to keep buying stock, offering an unqualified thumbs-up for heavy spending.

But this theory assumes a practically infinite life span for Amazon. The modern history of retail innovation suggests that even the behemoths can be overtaken suddenly. Sears was still America’s largest retailer in 1982, but just nine years later, its annual revenues were barely half those of Walmart. “The economic countryside is littered with the carcasses of companies that thought they had a [durable] competitive advantage,” says Alex Field, an economic historian at Santa Clara University. “Just look at BlackBerry or AOL.”

Amazon is not as insulated from its rivals as some think it is. Walmart, eBay, and a bounty of upstarts are all in the race to dominate online retail. Amazon’s furious spending on new buildings and equipment isn’t an elective measure; it’s a survival plan. The truth is that the company benefits from a beautiful but delicate tautology: Amazon has won investors’ trust with a reputation for spending everybody to death, and it can spend everybody to death because it has won investors’ trust. For now.

“Amazon, as best I can tell, is a charitable organization being run by elements of the investment community for the benefit of consumers,” Slate’s Matthew Yglesias joked earlier this year.

It has every element of the classic Amazon coverage story. "Flip the switch." The faceless community of Amazon's naive and trusting investors.   The charitable organization quote, ™Yglesias 2013. And above all, a fairly clear and accurate statement of Amazon's actual business strategy. 

What a wonderful feeling, to be able to conceal a secret in plain sight. Laid bare before its competitors, its investors, the press, is the recipe and the blueprint, in plain language. I agree with Thompson and others that it is increasingly difficult to find real business moats or competitive advantages in the modern world, what with the internet eliminating many previous physical moats of time and space. What remains , though, is a footrace of beautiful simplicity, one in which Amazon is both tortoise and hare. It is patient, it plays for the long-term like no other company, it will take failure after failure and never lose heart, and yet it will sprint when it picks up a scent. And it will take the race to an arena with the thinnest of air.

If I were an Amazon competitor, I'd actually regard Amazon's current run of quarterly losses as a terrifying signal. It means Amazon is arming itself to take the contest to higher ground. The retail game is about to become more, not less, punishing.