Amazon Health

Posted on

It’s pretty rare for the same company to feature in two consecutive Weekly Articles; yesterday’s announcement of a health care initiative involving Amazon, though, is not only incredibly intriguing, it also fits directly into some of the most important themes on Stratechery. I couldn’t resist.

The Announcement

From a joint press release:

Amazon, Berkshire Hathaway and JPMorgan Chase & Co. announced today that they are partnering on ways to address healthcare for their U.S. employees, with the aim of improving employee satisfaction and reducing costs. The three companies, which bring their scale and complementary expertise to this long-term effort, will pursue this objective through an independent company that is free from profit-making incentives and constraints. The initial focus of the new company will be on technology solutions that will provide U.S. employees and their families with simplified, high-quality and transparent healthcare at a reasonable cost.

Tackling the enormous challenges of healthcare and harnessing its full benefits are among the greatest issues facing society today. By bringing together three of the world’s leading organizations into this new and innovative construct, the group hopes to draw on its combined capabilities and resources to take a fresh approach to these critical matters…

The effort announced today is in its early planning stages, with the initial formation of the company jointly spearheaded by Todd Combs, an investment officer of Berkshire Hathaway; Marvelle Sullivan Berchtold, a Managing Director of JPMorgan Chase; and Beth Galetti, a Senior Vice President at Amazon. The longer-term management team, headquarters location and key operational details will be communicated in due course.

I’ve gotten more and more questions from readers about the possibilities of Amazon and health care, even before this announcement. I’ve been surprised, to be honest, but perhaps I shouldn’t be: I was the one who declared on The Bill Simmons Podcast that “Amazon’s goal is to basically take a skim off of all economic activity”, and given that health care was 17.9% of GDP in 2016, well, I guess that means I predicted this!

Amazon Health Marketplace

What is “this”, though? It certainly is tempting to jump immediately to a possible end game predicated on the ideas I have laid out in The Amazon Tax, Amazon’s New Customer, and Amazon Go and the Future:

  • Amazon builds out “interfaces” for its employees (as well as those of Berkshire Hathaway and J.P. Morgan Chase — I’ll just refer to Amazon from here on out), both digital and physical, to access basic healthcare needs; these sit in front of pharmacy benefit managers (PBMs), insurance administrators, wholesale distributors and pharmacies.
  • Amazon starts building out infrastructure for those healthcare suppliers, requiring them to serve Amazon’s employees using a standard interface.

Amazon could then go in one of two directions. First, Amazon could start to backwards integrate into its suppliers’ business; there are hints the company is already exploring pharmaceutical sales, and the Wall Street Journal says the idea was broached. That said, I actually think this is less likely; insurance operates best at more scale, not less: first and foremost, the larger the pool, the more risk can be spread, as well as obvious efficiency gains in administration. More scale also gives more bargaining power over other parts of the healthcare chain. Three companies, large though they may be, aren’t going to be as effective as large insurers, no matter how well-managed they may be.

What would make more sense to me is that, having first built an interface for its employees, and then a standardized infrastructure for its health care suppliers, is that Amazon converts the latter into a marketplace where PBMs, insurance administrators, distributors, and pharmacies have to compete to serve employees. And then, once that marketplace is functioning, Amazon will open the floodgates on the demand side, offering that standard interface to every large employer in America.

Aggregation and Suppliers

This is certainly ambitious enough — basically intermediating U.S. employers and the U.S. healthcare industry — but in fact this only sets the stage for the wholesale disruption of American healthcare. First, Amazon could not only open up its standard interface to other large employers, but small-and-medium sized businesses, and even individuals; in this way the Amazon Health Marketplace could aggregate by far the most demand for healthcare.

Consolidating demand by offering a superior user experience is how aggregators gain power; given the scenario I just sketched out, Aggregation Theory has a prediction about what might happen next:

Once an aggregator has gained some number of end users, suppliers will come onto the aggregator’s platform on the aggregator’s terms, effectively commoditizing and modularizing themselves. Those additional suppliers then make the aggregator more attractive to more users, which in turn draws more suppliers, in a virtuous cycle.

This means that for aggregators, customer acquisition costs decrease over time; marginal customers are attracted to the platform by virtue of the increasing number of suppliers. This further means that aggregators enjoy winner-take-all effects: since the value of an aggregator to end users is continually increasing it is exceedingly difficult for competitors to take away users or win new ones.

The key words there are “commoditize and modularize”, and this is where the option I dismissed above comes into play, but not in the way most think: Amazon doesn’t create an insurance company to compete with other insurance companies (or the other pieces of healthcare infrastructure); rather, Amazon makes it possible — and desirable — for individual health care providers to come onto their platform directly, be that doctors, hospitals, pharmacies, etc.

After all, if Amazon is facilitating the connection to patients, what is the point of having another intermediary? Moreover, by virtue of being the new middleman, Amazon has the unique ability to consolidate patient data in a way that is not only of massive benefit to patients and doctors but also to the application of machine learning.

Of course that leaves the insurance piece, which makes Berkshire Hathaway a useful partner; conveniently, Berkshire Hathaway is not in the health insurance business, but rather the health reinsurance business — that is, they insure the insurers. Or, to put it another way, they don’t provide any of the services that Amazon Health Marketplace might make obsolete, and specialize in the one thing Amazon Health Services would need.

Oh, and this will be really expensive, and take years to get off the ground. It certainly would be helpful to have access to financing and capital markets, which means it would be very helpful to partner with JPMorgan Chase & Company. The skills these three companies bring to bear seems far more relevant than the number of employees (and besides, the company alliance approach to traditional health care has been done).

Is This Happening?

Needless to say, what I just sketched out is extremely ambitious; it is easy to let one’s mind run wild when it comes to a company without a name, a management team, or a location. Moreover, the press release was quite modest in its ambitions; I quoted it above, but here is the relevant piece again:

The three companies, which bring their scale and complementary expertise to this long-term effort, will pursue this objective through an independent company that is free from profit-making incentives and constraints. The initial focus of the new company will be on technology solutions that will provide U.S. employees and their families with simplified, high-quality and transparent healthcare at a reasonable cost.

Ah yes, “technology solutions”. We’ve certainly seen that before, and it hasn’t worked.

That, though, is where the previous line comes in: the scenario that I sketched out above is wildly profitable, to be sure, but only years down the road when demand is fully aggregated and Amazon Health Marketplace is taking a skim off of every transaction; if short-term profit isn’t the goal, long-term goals become much more realistic.

And there it is, in the first sentence: “this long-term effort.” These three companies are clear up-front that this isn’t a one-off effort; there is the commitment to the long-term, and while “technological solutions” seems like a short-term play, I just explained why that is the place the start. Aggregators win with products that are simple, high-quality, and easy to understand — exactly what this press release promised.

Is This Possible?

I’m not a healthcare expert by any means; I know enough to know that the U.S. system is incredibly complex, bedeviled by incentive problems, and tied up in all kinds of messy ways with regulations (mostly justified!).

At the same time, the U.S. healthcare system is inextricably tied up with the post-World War 2 order; indeed, the entire reason employers are so important to the system is because of World War 2 regulations that instituted price controls on wages, incentivizing employers to use benefits as a means of attracting workers (this was further enshrined by making healthcare benefits tax-exempt).

That system, though, is under more duress than ever. I wrote in TV Advertising’s Surprising Strength — and Inevitable Fall:

What should be terrifying to television executives is that all of those pieces that make television advertising the gold mine that it has been are under the exact same threat that TV watching itself is: the threat of the Internet. Start with the top 25 advertisers in the U.S.…

Notice that the vast majority of the industries on on this list are dominated by massive companies that compete on scale and distribution. CPG is the perfect example: building a “house of brands” allows a company like Procter & Gamble to target demographic groups even as they leverage scale to invest in R&D, bring down the cost of products, and most importantly, dominate the distribution channel (i.e. retail shelf space). Said retailers, meanwhile, are huge in their own right, not only so they can match their massive suppliers at the bargaining table but also so they can scale logistics, inventory management, store development, etc. Automobile companies, meanwhile, are not unlike CPG companies: they operate a “house of brands” to serve different demographics while benefitting from scale in production and distribution; the primary difference is that they make money through one large purchase instead of over many smaller purchases over time.

Note [that nearly all] of the companies on this list are threatened by the Internet.

My thesis in that article — repeated in Dollar Shave Club and the Disruption of Everything and The Sports Linchpin — is that the post-World War 2 economic system was deeply intertwined and interdependent, and that the root of everything was control of distribution. The Internet, though, made the distribution of information free, upsetting not just information providers like publishers, but all industries; it follows, then, that to the extent that the current health care system is built on that post-World War 2 order, such is the extent to which it is vulnerable.

That is not to say its collapse is imminent — quite the opposite, in fact. Each seemingly distinct industry, by virtue of being interdependent on others, supports each other. My expectation, then, is not that the Internet methodically disrupts industry after industry in some sort of chronological order, but rather that the entire edifice lasts far longer than technologists think, only to one day collapse far quicker than anyone expected.

The ultimate winners of this shakeout, then, are not only companies that are building businesses predicated on the Internet, but just as importantly, are willing and able to build those businesses with the patience that will be necessary to wait for the old order to collapse, particularly if that collapse happens years or decades after the underlying business models are rotten.

There is no more patient company than Amazon.

Amazon Go and the Future

Posted on

Amazon Go is the story of technology, and so is this tweet:

Yesterday the Amazon Go concept store in Seattle opened to the public, filled with sandwiches, salads, snacks, various groceries, and even beer and wine (Recode has a great set of pictures here). The trick is that you don’t pay, at least in person: a collection of cameras and sensors pair your selection to your Amazon account — registered at the door via smartphone app — which rather redefines the concept of “grab-and-go.”

The economics of Amazon Go define the tech industry; the strategy, though, is uniquely Amazon’s. Most of all, the implications of Amazon Go explain both the challenges and opportunities faced by society broadly by the rise of tech.

The Economics of Tech

This point is foundational to nearly all of the analysis of Stratechery, which is why it’s worth repeating. To understand the economics of tech companies one must understand the difference between fixed and marginal costs, and for this Amazon Go provides a perfect example.

A cashier — and forgive the bloodless language for what is flesh and blood — is a marginal cost. That is, for a convenience store to sell one more item requires some amount of time on the part of a cashier, and that time costs the convenience store operator money. To sell 100 more items requires even more time — costs increase in line with revenue.

Fixed costs, on the other hand, have no relation to revenue. In the case of convenience stores, rent is a fixed cost; 7-11 has to pay its lease whether it serves 100 customers or serves 1,000 in any given month. Certainly the more it serves the better: that means the store is achieving more “leverage” on its fixed costs.

In the case of Amazon Go specifically, all of those cameras and sensors and smartphone-reading gates are fixed costs as well — two types, in fact. The first is the actual cost of buying and installing the equipment; those costs, like rent, are incurred regardless of how much revenue the store ultimately produces.

Far more extensive, though, are the costs of developing the underlying systems that make Amazon Go even possible. These are R&D costs, and they are different enough from fixed costs like rent and equipment that they typically live in another place on the balance sheet entirely.

These different types of costs affect management decision-making at different levels (that is, there is a spectrum from purely marginal costs to purely fixed costs; it all depends on your time frame):

  • If the marginal cost of selling an individual item is more than the marginal revenue gained from selling the item (i.e. it costs more to pay a cashier to sell an item than the gross profit earned from an item) then the item won’t be sold.
  • If the monthly rent for a convenience store exceeds the monthly gross profit from the store, then the store will be closed.
  • If the cost of renovations and equipment (in the case of small businesses, this cost is usually the monthly repayments on a loan) exceeds the net profit ex-financing, then the owner will go bankrupt.

Keep in mind, most businesses start out in the red: it usually takes financing, often in the form of a loan, to buy everything necessary to even open the business in the first place; a company is not truly profitable until that financing is retired. Of course once everything is paid off a business is not entirely in the clear: physical objects like shelves or refrigeration units or lights break and wear out, and need to be replaced; until that happens, though, money can be made by utilizing what has already been paid for.

This, though, is why the activity that is accounted for in R&D is so important to tech company profitability: while digital infrastructure obviously needs to be maintained, by-and-large the investment reaps dividends far longer than the purchase of any physical good. Amazon Go is a perfect example: the massive expense that went into developing the underlying system powering cashier-less purchasing does not need to be spent again; moreover, unlike shelving or refrigerators, the output of that expense can be duplicated infinitely without incurring any additional cost.

This principle undergirds the fantastic profitability of successful tech companies:

  • It was expensive to develop mainframes, but IBM could reuse the expertise to build them and most importantly the software needed to run them; every new mainframe was more profitable than the last.
  • It was expensive to develop Windows, but Microsoft could reuse the software on all computers; every new computer sold was pure profit.
  • It was expensive to build Google, but search can be extended to anyone with an Internet connection; every new user was an opportunity to show more ads.
  • It was expensive to develop iOS, but the software can be used on billions of iPhones, every one of which generates tremendous profit.
  • It was expensive to build Facebook, but the network can scale to two billion people and counting, all of which can be shown ads.

In every case a huge amount of fixed costs up front is overwhelmed by the ongoing ability to make money at scale; to put it another way, tech companies combine fixed costs with marginal revenue opportunities, such that they make more money on additional customers without any corresponding rise in costs.

This is clearly the goal with Amazon Go: to build out such a complex system for a single store would be foolhardy; Amazon expects the technology to be used broadly, unlocking additional revenue opportunities without any corresponding rise in fixed costs — of developing the software, that is; each new store will still require traditional fixed costs like shelving and refrigeration. That, though, is why this idea is so uniquely Amazonian.

The Strategy of Technology

The most important difference between Amazon and most other tech companies is that the latter generally invest exclusively in research and development — that is, to say, in software. And why not? As I just explained software development has the magical properties of value retention and infinite reproduction. Better to let others handle the less profitable and more risky (at least in the short term) marginal complements. To take the three most prominent examples:

  • Microsoft builds the operating system (and eventually, application software) and leaves the building of computers to OEMs
  • Google builds the search engine and leaves the creation of web pages to be searched to the rest of the world
  • Facebook builds the infrastructure of the network, and leaves the creation of content to be shared to its users

All three companies are, at least in terms of their core businesses, pure software companies, which means the economics of their businesses align with the economics of software: massive fixed costs, and effectively zero marginal costs. And while Microsoft’s market, large though it may have been, was limited by the price of a computer, Google and Facebook, by virtue of their advertising model, are super-aggregators capable of scaling to anyone with an Internet connection. All three also benefit (or benefited) from strong network effects, both on the supply and demand side; these network effects, supercharged by the ability to scale for free, are these companies’ moats.

Apple and IBM, on the other hand, are/were vertical integrators, particularly IBM. In the mainframe era the company built everything from components to operating systems to application software and sold it as a package with a long-term service agreement. By doing so all would-be competitors were foreclosed from IBM’s market; eventually, in a(n unsuccessful) bid to escape antitrust pressure, application software was opened up, but that ended up entrenching IBM further by adding on a network effect. Apple isn’t nearly as integrated as IBM was back in the 60s, but it builds both the software and the finished products on which it runs, foreclosing competitors (while gaining economies of scale from sourcing components and two-sided network effects through the App Store); Apple is also happy to partner with telecoms, which have their own network effects.

Amazon is doing both.

In market after market the company is leveraging software to build horizontal businesses that benefit from network effects: in e-commerce, more buyers lead to more suppliers lead to more buyers. In cloud services, more tenants lead to great economies of scale, not just in terms of servers and data centers but in the leverage gained by adding ever more esoteric features that both meet market needs and create lock-in. As I wrote last year the point of buying Whole Foods was to jump start a similar dynamic in groceries.

At the same time Amazon continues to vertically integrate. The company is making more and more products under its own private labels on one hand, and building out its fulfillment network on the other. The company is rapidly moving up the stack in cloud services, offering not just virtual servers but microservices that obviate the need for server management entirely. And in logistics the company has its own airplanes, trucks, and courier services, and has promised drones, with the clear goal of allowing the company to deliver products entirely on its own.

To be both horizontal and vertical is incredibly difficult: horizontal companies often betray their economic model by trying to differentiate their vertical offerings; vertical companies lose their differentiation by trying to reach everyone. That, though, gives a hint as to how Amazon is building out its juggernaut: economic models — that is, the constraint on horizontal companies going vertical — can be overcome if the priority is not short-term profit maximization.

Amazon’s Triple Play

In 2012 Amazon acquired Kiva Systems for $775 million, the then-second largest acquisition in company history.1 Kiva Systems built robots for fulfillment centers, and many analysts were puzzled by the purchase: Kiva Systems already had a plethora of customers, and Amazon was free to buy their robots for a whole lot less than $775 million. Both points argued against a purchase: continuing to sell to other companies removed the only plausible strategic rationale for buying the company instead of simply buying robots, but to stop selling to Kiva Systems’ existing customers would be value-destructive. It’s one thing to pay 8x revenue, as Amazon did; it’s another to cut off that revenue in the process.

In fact, though, that is exactly what Amazon did. The company had no interest in sharing Kiva Systems’ robots with its competitors, leaving a gap in the market. At the same time the company ramped up its fulfillment center build-out, gobbling up all of Kiva Systems’ capacity. In other words, Amazon made the “wrong” move in the short-term for a long-term benefit: more and better fulfillment centers than any of its competitors — and spent billions of dollars doing so.

This willingness to spend is what truly differentiates Amazon, and the payoffs are tremendous. I mentioned telecom companies in passing above: their economic power flows directly from massive amounts of capital spending; said power is limited by a lack of differentiation. Amazon, though, having started with a software-based horizontal model and network-based differentiation, has not only started to build out its vertical stack but has spent massive amounts of money to do so. That spending is painful in the short-term — which is why most software companies avoid it — but it provides a massive moat.

That is why, contra most of the analysis I have seen, I don’t think Amazon will license out the Amazon Go technology. Make no mistake, that is exactly what a company like Google would do (and as I expect them to do with Waymo), and for good reason: the best way to get the greatest possible return on software R&D is to spread it as far and wide as possible, which means licensing. The best way to build a moat, though, is to actually put in the effort to dig it, i.e. spend the money.

To that end, I suspect that in five to ten years the countries Amazon serves will be blanketed with Amazon Go stores, selling mostly Amazon products, augmented by Amazon fulfillment centers. That is the other point many are missing; yes, the Amazon Go store took pains to show that it still had plenty of workers: shelf stockers, ID checkers, and food preparers, etc.

Workers as seen from a sidewalk window as they assemble sandwiches in an Amazon Go store Monday, Jan. 22, 2018, in Seattle. More than a year after it introduced the concept, Amazon opened its artificial intelligence-powered Amazon Go store in downtown Seattle on Monday. The store on the bottom floor of the company’s Seattle headquarters allows shoppers to scan their smartphone with the Amazon Go app at a turnstile, pick out the items they want and leave. (AP Photo/Elaine Thompson)

Unlike cashiers, though, none of these jobs have to actually be present in the store most of the time. It seems obvious that Amazon Go stores of the future will rarely have employees in store at all: there will be a centralized location for food preparation and a dedicated fleet of shelf stockers. That’s the thing about Amazon: the company isn’t afraid of old-world scale. No, sandwich preparation doesn’t scale infinitely, but it does scale, particularly if you are willing to spend.

Marx Versus Bezos

The political dilemma embedded in this analysis is hardly new: Karl Marx was born 200 years ago. Technology like Amazon Go is the ultimate expression of capital: invest massive amounts of money up front in order to reap effectively free returns at scale. What has fundamentally changed, though, is the role of labor: Marx saw a world where capital subjugated labor for its own return; technologies like Amazon Go have increasingly no need for labor at all.

Some, certainly, see this as a problem: what about all the cashiers? What about all the truck drivers? What about all of the other jobs that will be displaced by automation? Well, I would ask, what about the labor of Marx’s day, the factory workers borne of the industrial revolution that he thought should overthrow the bourgeoisie?

In fact, they are all gone, replaced by automation. And, in the meantime, nearly all of humanity has been lifted out of abject poverty. As Nicholas Kristof wrote in the New York Times:

2017 was probably the very best year in the long history of humanity. A smaller share of the world’s people were hungry, impoverished or illiterate than at any time before. A smaller proportion of children died than ever before. The proportion disfigured by leprosy, blinded by diseases like trachoma or suffering from other ailments also fell…

Every day, the number of people around the world living in extreme poverty (less than about $2 a day) goes down by 217,000, according to calculations by Max Roser, an Oxford University economist who runs a website called Our World in Data. Every day, 325,000 more people gain access to electricity. And 300,000 more gain access to clean drinking water.

I don’t seek to minimize real struggles, much less the real displacement that will come from technologies like Amazon Go writ large. For decades technology helped the industrial world work better; more and more, technology is replacing that world completely, and there will be pain. That, though, is precisely why it is worth remembering that the world is not static: to replace humans is, in the long run, to free humans to create entirely new needs and means to satisfy those needs. It’s what we do, and the faith to believe it will happen again will be the best guide in figuring out how.

As for Amazon, the company’s goal to effectively tax all economic activity continues apace. Surely the company is grateful about the attention Facebook is receiving from the public, even as it builds a monopoly with a triple moat. The lines outside Amazon Go, though, are a reminder of exactly why aggregator monopolies are something entirely new: these companies are dominant because people love them. Regulation may be as elusive as Marx’s revolution.

I wrote a follow-up to this article in this Daily Update.

  1. The original version of this article mistakenly said then-largest; Zappos was acquired for $900 million in 2009 ↩

Facebook’s Motivations

Posted on

The trepidation — and inevitable outrage — with which much of the media has greeted Facebook’s latest change to the News Feed algorithm seems rather anticlimactic. Nearly three years ago I wrote in The Facebook Reckoning that any publisher that was not a “destination site” — that is, a site that had a direct connection with readers — had no choice but to go along with Facebook’s Instant Article initiative, even though Facebook could change their mind at any time. A few months later, in Popping the Publishing Bubble, I explained why advertising would coalesce with Google and Facebook; that is indeed what has happened, which is the real problem for publishers. Facebook’s algorithm change simply hastens the inevitable.

The story for media is for all intents and purposes unchanged: success depends on building a direct relationship with readers; monetizing that relationship (likely through subscriptions, but not necessarily); and leveraging Facebook as an acquisition channel for those long-term relationships, not short-term page views. If anything this change will help reader-focused publications: users will be more likely to see links shared by their friends, enhancing the word-of-mouth marketing that is the foundation of reader-centric publications.

What I find far more compelling is the question of Facebook’s motivation. Facebook CEO Mark Zuckerberg wrote on Facebook:

One of our big focus areas for 2018 is making sure the time we all spend on Facebook is time well spent. We built Facebook to help people stay connected and bring us closer together with the people that matter to us. That’s why we’ve always put friends and family at the core of the experience. Research shows that strengthening our relationships improves our well-being and happiness.

We feel a responsibility to make sure our services aren’t just fun to use, but also good for people’s well-being. So we’ve studied this trend carefully by looking at the academic research and doing our own research with leading experts at universities. The research shows that when we use social media to connect with people we care about, it can be good for our well-being. We can feel more connected and less lonely, and that correlates with long term measures of happiness and health. On the other hand, passively reading articles or watching videos — even if they’re entertaining or informative — may not be as good.

Based on this, we’re making a major change to how we build Facebook. I’m changing the goal I give our product teams from focusing on helping you find relevant content to helping you have more meaningful social interactions. We started making changes in this direction last year, but it will take months for this new focus to make its way through all our products. The first changes you’ll see will be in News Feed, where you can expect to see more from your friends, family and groups. As we roll this out, you’ll see less public content like posts from businesses, brands, and media. And the public content you see more will be held to the same standard — it should encourage meaningful interactions between people…

Now, I want to be clear: by making these changes, I expect the time people spend on Facebook and some measures of engagement will go down. But I also expect the time you do spend on Facebook will be more valuable. And if we do the right thing, I believe that will be good for our community and our business over the long term too.

Forgive the longer-than-usual excerpt, but there is a lot here. Zuckerberg:

  • Implicitly admits that time spent on Facebook may not be “well-spent”, and cites research suggesting that many common activities on Facebook may not be good for you
  • Introduces the change as a shift in goals from delivering relevant content (a “perfect personalized newspaper”, as Zuckerberg called it in 2014)
  • Suggests that the time spent on Facebook may decrease due to these changes (sending Facebook’s stock down)

In an interview for the Daily Update, Vice-President of News Feed Adam Mosseri argued that this would benefit Facebook in the long run:

This change is primarily focused on doing right by our community, because we actually believe that by doing right by the community in the long run will be good for the business and so we just try to take a long term approach to any question like this.

I absolutely believe the last part of that quote: Facebook is taking a long-term view, and it would only make this change were it right for the business. I’m just not entirely convinced that Zuckerberg and Mosseri are telling us the entire story.

Facebook’s Believability

Start with Zuckerberg’s claim that this change will reduce “the time people spend on Facebook and some measures of engagement.” Mosseri said that would be mostly due to less time spent watching video, given that video content would likely be hurt by this algorithmic change.

That in and of itself is certainly interesting; Zuckerberg has been pushing the importance of video on earnings calls for some time now, and no wonder: TV advertising money remains the proverbial gold-at-the-end-of-the-rainbow for all advertising-based tech companies. Is Facebook giving up on its leprechaun dreams?

I don’t think so, and not just because forgoing all of that potential revenue would be quite unbelievable. Instead, I think the answer was laid out by Zuckerberg during Facebook’s Q1 2017 earnings call while answering a question about Facebook’s new video tab:

For the video tab, the goal that we have for the product experience is to make it so that when people want to watch videos or they want to keep up to date on what’s going on with their favorite show or what’s going on with the public figure that they want to follow, that they can come to Facebook and go to a place knowing that that’s going to show them all the content that they’re interested in.

So that’s a pretty different intent than how people come to Facebook today. Today, for the most part, people pull Facebook out when they have a few minutes, when they want to catch up and see what’s going on in the world with their friends and in the news and everything that’s going on. That’s very different from saying, hey, I want to watch video content now. And that’s what I think we’re going to unlock with this tab.

My takeaway at the time was that Facebook was effectively building two video products: one for content people wanted to watch (the video tab), and the other for content people watched because it was stuck in front of them (News Feed video).

I think that was right, but it also follows that the former would be easier to monetize: after all, people are more likely to put up with an advertisement for a video they want to watch, as opposed to one they are watching because it happened to be presented to them. Indeed, the latter could be actively harmful, reminding people to simply close the app. To that end, reducing the time users spend watching videos that Facebook would never monetize effectively doesn’t seem like a particularly large loss.

That’s not the only reason why it is hard to take Facebook seriously when it comes to proclamations of doom-and-gloom. Back in 2016, on the 3Q 2016 earnings call, Facebook CFO Dave Wehner said that Facebook would soon stop growing the ad load on News Feed and that advertising growth would “come down meaningfully.”

I wondered at the time if this meant Facebook’s ads were less differentiated — and thus had less pricing power in the face of increasing scarcity — than I expected. In fact, my initial analysis was spot-on: as I have been documenting in the Daily Update over the last year, Facebook’s price-per-ad has been increasing as ad impression growth has declined over the last year, strongly suggesting that Facebook has pricing power:

The growth rates of Facebook's ad metrics over time

So excuse me if I take Facebook’s pronunciations about the harm its business will soon befall with a rather large grain of salt. The company has already demonstrated it has pricing power such that its advertising revenue can continue to grow strongly even as the number of ads-per-user plateaus; moreover, that power further complicates any attempt to understand Facebook’s motivation.

Facebook’s Imperviousness

The key thing to remember about Facebook — and Google’s — dominance in digital ads is that their advantages are multi-faceted. First and foremost are the attractiveness of their products to users; that attractiveness is rooted not only in technology but also in both data and people-based network effects. Second is the depth of information both companies have on their users, allowing advertisers to spend more efficiently on their platforms — particularly on mobile — than elsewhere. The third advantage, though, is perhaps the least appreciated: buying ads on Google and Facebook is just so much easier. They are one-stop shops for reaching anyone, which means competitors need to not have similar targeting capabilities and user engagement, but in fact need to be significantly better to justify the effort.

These structural advantages lend credibility to Facebook’s contention that it is making these changes with its users’ best interests in mind. After all, it ultimately won’t matter to the bottom line. Indeed, note that Zuckerberg made no mention of these changes impacting revenue, as he surely should have were this change to have a negative impact; in contrast, Zuckerberg warned that hiring new content moderators would impact profitability on the last earnings call.

Of course one hesitates to give Facebook too much credit if this were the case: it would be a clear example of a Strategy Credit, where doing the right thing is easy because it doesn’t actually hurt the underlying business. That, though, may be reassuring in the short term, but it points to still more possible Facebook motivations.

Facebook’s Threats

For about as long as Facebook has been a going concern, the conventional wisdom about their downfall has remained largely the same: some other social network is going to come along, probably amongst young people, and take all of the attention away from Facebook. In fact, as I argued last year in Facebook, Phones, and Phonebooks, the social sphere has room for many players — including networks that garner huge amounts of attention — but that Facebook’s position was secure.

It is increasingly clear that there are two types of social apps: one is the phone book, and one is the phone. The phone book is incredibly valuable: it connects you to anyone, whether they be a personal friend, an acquaintance, or a business. The social phone book, though, goes much further: it allows the creation of ad hoc groups for an event or network, it is continually updated with the status of anyone you may know or wish to know, and it even provides an unlimited supply of entertaining professionally produced content whenever you feel the slightest bit bored.

The phone, on the other hand, is personal: it is about communication between you and someone you purposely reach out to. True, telemarketing calls can happen, but they are annoying and often dismissed. The phone is simply about the conversation that is happening right now, one that will be gone the moment you hang up.

In the U.S. the phone book is Facebook and the phone is Snapchat; in Taiwan, where I live, the phone book is Facebook and the phone is LINE. Japan and Thailand are the same, with a dash of Twitter in the former. In China WeChat handles it all, while Kakao is the phone in South Korea. For much of the rest of the world the phone is WhatsApp, but for everywhere but China the phone book is Facebook.

This isn’t a bad thing; indeed, it is an incredibly valuable thing: Facebook’s status as a utility is exactly what makes the company so valuable. It has the data to target advertising and the feed in which to place it, and it is difficult to imagine any of the phone companies overtaking it in value.

Make no mistake, in this analogy the phone book is where the money is at: Snapchat and Twitter are all struggling to monetize in large part because phones simply aren’t conducive to advertising.1 That, though, makes Facebook’s new focus even more interesting: if advertising struggles to find a place when users are more actively engaged (versus passively consuming content), why is Facebook seemingly going in the opposite direction?

One possible answer is that conventional wisdom is right: Facebook may still have a hold on identity, but the amount of time users — particularly the most valuable users — are spending on the network is steadily decreasing.2 That may not be a problem for the business today, but it certainly could be in the long run.

Another possible answer is that Facebook fears regulation, and by demonstrating the ability to self-correct and focus on what makes Facebook unique the company can avoid regulatory issues completely. The question, though, is how exactly would Facebook be regulated? There certainly is no crime in providing a free service that lets people connect with those they know. I suggested last year that perhaps Facebook’s monopoly power could be seen in its seeming inability to help publishers monetize or especially in digital ads, but those cases are far more theoretical (or in the case of publishers, fantastical) for now.

Perhaps there is a third motivation though: call it “enlightened self-interest.” Keep in mind from whence Facebook’s power flows: controlling demand. Facebook is a super-aggregator, which means it leverages its direct relationship with users, zero marginal costs to serve those users, and network effects, to steadily decrease acquisition costs and scale infinitely in a virtuous cycle that gives the company power over both supply (publishers) and advertisers.

It follows that Facebook’s ultimate threat can never come from publishers or advertisers, but rather demand — that is, users. The real danger, though, is not from users also using competing social networks (although Facebook has always been paranoid about exactly that); that is not enough to break the virtuous cycle. Rather, the only thing that could undo Facebook’s power is users actively rejecting the app. And, I suspect, the only way users would do that en masse would be if it became accepted fact that Facebook is actively bad for you — the online equivalent of smoking.

This is why I find Facebook’s focus on what is good for users to be so fascinating. On one level, maybe the company is, as they can afford to be, simply altruistic. On another, perhaps they are diverting attention from problematic trends in user engagement. Or perhaps they are seeking to neutralize their biggest threat by addressing it head-on.

I don’t know which of these motivations are correct — probably there is truth in all of them — which is precisely why I find this announcement so fascinating. This change could have been made and justified without even broaching the idea that Facebook might be bad for you; why did Facebook rest everything on that reasoning?

It certainly is hard to escape the election of President Trump. I have argued regularly that I don’t believe that fake news was a causal factor in Trump’s election, and I think that Facebook has been a convenient scapegoat for many.

On the other hand, I made the case back in the primaries that Facebook’s decimation of the media led to a correlated decimation of the parties’ ability to control the presidential candidate selection process, creating the conditions for a candidate like Trump to arise. In other words, I do blame Facebook for Trump, but for structural reasons, not causal ones. And even then, Facebook is a stand-in for the Internet’s effect broadly: were it not Facebook ruining media’s business model, it would have been some other company.

Zuckerberg, though, has always seemed to tilt towards the more utopian side of the spectrum when it comes to the Silicon Valley cliche of “changing the world.” The ardent belief that sharing and connecting will fix everything has been a fixture in Zuckerberg’s public comments ever since he emerged into the public sphere, and the CEO effectively declared at the 2016 F8 conference that Trump was in opposition to that.

In that light dismissing Facebook’s change as a mere strategy credit is perhaps to give short shrift to Zuckerberg’s genuine desire to leverage Facebook’s power to make the world a better place. Zuckerberg argued in his 2017 manifesto Building Global Community:

Progress now requires humanity coming together not just as cities or nations, but also as a global community. This is especially important right now. Facebook stands for bringing us closer together and building a global community. When we began, this idea was not controversial. Every year, the world got more connected and this was seen as a positive trend. Yet now, across the world there are people left behind by globalization, and movements for withdrawing from global connection. There are questions about whether we can make a global community that works for everyone, and whether the path ahead is to connect more or reverse course.

Our job at Facebook is to help people make the greatest positive impact while mitigating areas where technology and social media can contribute to divisiveness and isolation. Facebook is a work in progress, and we are dedicated to learning and improving. We take our responsibility seriously.

That, though, leaves the question I raised in response to that manifesto:

Even if Zuckerberg is right, is there anyone who believes that a private company run by an unaccountable all-powerful person that tracks your every move for the purpose of selling advertising is the best possible form said global governance should take?

My deep-rooted suspicion of Zuckerberg’s manifesto has nothing to do with Facebook or Zuckerberg; I suspect that we agree on more political goals than not. Rather, my discomfort arises from my strong belief that centralized power is both inefficient and dangerous: no one person, or company, can figure out optimal solutions for everyone on their own, and history is riddled with examples of central planners ostensibly acting with the best of intentions — at least in their own minds — resulting in the most horrific of consequences; those consequences sometimes take the form of overt costs, both economic and humanitarian, and sometimes those costs are foregone opportunities and innovations. Usually it’s both.

Facebook’s stated reasoning for this change only heightens these contradictions: if indeed Facebook as-is harms some users, fixing that is a good thing. And yet the same criticism becomes even more urgent: should the personal welfare of 2 billion people be Mark Zuckerberg’s personal responsibility?

  1. LINE actually monetizes decently well on a per-user basis, but has struggled to grow outside of its core markets ↩

  2. This has been rumored for a long time, but it is very difficult to pick out facts from motivated guessing when it comes to Facebook scuttlebutt ↩

Meltdown, Spectre, and the State of Technology

Posted on

You’ve heard the adage “It’s all 1s and 0s”, but that’s not a figure of speech: the transistor, the fundamental building block of computers, is simply a switch that is either on (“1”) or off (“0”). It turns out, though, as Chris Dixon chronicled in a wonderful essay entitled How Aristotle Created the Computer, that 1s and 0s, through the combination of mathematical logic and transistors, are all you need:

The history of computers is often told as a history of objects, from the abacus to the Babbage engine up through the code-breaking machines of World War II. In fact, it is better understood as a history of ideas, mainly ideas that emerged from mathematical logic, an obscure and cult-like discipline that first developed in the 19th century.

Dixon’s essay — which I’ve linked to previously — is well worth a read, but the relevant point for this article is perhaps a surprising one: computers are really stupid; what makes them useful is that they are stupid really quickly.

The Problem with Processor Vulnerabilities

Last week the technology world was shaken by the disclosure of two vulnerabilities in modern processors: Meltdown and Spectre. The announcement was a bit haphazard, thanks to the fact that the disclosure date was moved up by a week due to widespread speculation about the nature of the vulnerability (probably driven by updates to the Linux kernel), but also because Meltdown and Spectre are similar in some respects, but different in others.

Start with the similarities: the outcome for both vulnerabilities is the same — a non-privileged user can access information on the computer they should not be able to, like secret keys or passwords or any other type of data owned by other users. This is a particularly big problem for cloud services like AWS, where multiple “tenants” use the same physical hardware:

This multi-tenant architecture is achieved through the use of virtual machines: there is specialized software that runs on a single physical computer that allows each individual user to operate as if they have their own computer, when in fact they are sharing. This is a win-win: single-user computers sit idle the vast majority of the time (they are stupid really quickly), but if multiple users can use one computer then the hardware can be utilized far more efficiently. And, in the case of cloud services, that same concept can be scaled up to millions of physical computers sharing even more fundamental infrastructure like cooling, networking, administration, etc.

The entire edifice, though, is predicated on a fundamental assumption: that users in one virtual machine cannot access data from another. That assumption, by extension, relies on trust in the integrity of the virtual machine software, which relies on trust in the integrity of the underlying operating system, which ultimately relies on trust in the processor at the heart of a server. From the Meltdown white paper (emphasis mine):

To load data from the main memory into a register, the data in the main memory is referenced using a virtual address. In parallel to translating a virtual address into a physical address, the CPU also checks the permission bits of the virtual address, i.e., whether this virtual address is user accessible or only accessible by the kernel. As already discussed in Section 2.2, this hardware-based isolation through a permission bit is considered secure and recommended by the hardware vendors. Hence, modern operating systems always map the entire kernel into the virtual address space of every user process. As a consequence, all kernel addresses lead to a valid physical address when translating them, and the CPU can access the content of such addresses. The only difference to accessing a user space address is that the CPU raises an exception as the current permission level does not allow to access such an address. Hence, the user space cannot simply read the contents of such an address.

The kernel is the core part of the operating system that should be inaccessible by normal users; it has its own memory to store not only core system data but also data from all of the users (for example, when it has to be written to or read from permanent storage). Even here, though, the system relies on virtualization — that memory is the same physical memory users utilize for their applications. It is up to the CPU to keep track of what parts of memory belong to whom, and this is where the vulnerabilities come in.

Speculative Execution

I just referenced three critical parts of a computer: the processor, memory, and permanent storage. In fact, the architecture for storing data is even more complex than that:

  • Registers are the fastest form of memory, accessible every single clock cycle (that is, a 2.0 GHz processor can access registers two billion times a second). They are also the smallest, usually only containing the inputs and outputs for the current calculation.
  • There are then various levels of cache (L1, L2, etc.) that are increasingly slower and, on the flipside, increasingly larger and less expensive. This cache is located in a hierarchy: data that is needed immediately will be moved from the registers to L1 cache, for example; slightly less necessary data will be in L2, then L3, etc.
  • The next major part of the memory hierarchy is main memory, that is system RAM. While the amount of cache is dependent on the processor model, the amount of memory is up to the overall system builder. This memory is massively slower than cache, but it is also massively larger and far less expensive.
  • The last part of the memory hierarchy, at least on a single computer, is permanent storage — the hard drive. Solid-state drives (SSDs) have made a huge difference in speed here, but even then permanent memory is massively slower than main memory, with the same tradeoffs: you can have a lot more of it at a much lower price.
  • While not part of the traditional memory hierarchy, cloud applications often have permanent memory on a separate physical server on the same network; the usual tradeoffs apply — very slow access in exchange for other benefits, in this case keeping data separate from its application.

To be sure, “very slow” is all relative — we are talking about nanoseconds here. This post by Jeff Atwood puts it in human terms:

That infinite space “between” what we humans feel as time is where computers spend all their time. It’s an entirely different timescale. The book Systems Performance: Enterprise and the Cloud has a great table that illustrates just how enormous these time differentials are. Just translate computer time into arbitrary seconds:

1 CPU cycle 0.3 ns 1 s
Level 1 cache access 0.9 ns 3 s
Level 2 cache access 2.8 ns 9 s
Level 3 cache access 12.9 ns 43 s
Main memory access 120 ns 6 min
Solid-state disk I/O 50-150 μs 2-6 days
Rotational disk I/O 1-10 ms 1-12 months
Internet: SF to NYC 40 ms 4 years
Internet: SF to UK 81 ms 8 years
Internet: SF to Australia 183 ms 19 years
OS virtualization reboot 4 s 423 years
SCSI command time-out 30 s 3000 years
Hardware virtualization reboot 40 s 4000 years
Physical system reboot 5 m 32 millenia

[…]

The late, great Jim Gray…also had an interesting way of explaining this. If the CPU registers are how long it takes you to fetch data from your brain, then going to disk is the equivalent of fetching data from Pluto.

Gray presented this slide while at Microsoft, to give context to that that “Olympia, Washington” reference. Let me extend his analogy:

Suppose you were a college student interning for the summer at Microsoft in Redmond, and you were packing clothes at home in Olympia. Now Seattle summers can be quite finicky — it could be blustery and rainy, or hot and sunny. It’s often hard to know what the weather will be like until the morning of. To that end, the prudent course of action would not be to pack only one set of clothes, but rather to pack clothes for either possibility. After all, it is far faster to change clothes from a suitcase than it is to drive home to Olympia every time the weather changes.

This is where the analogy starts to fall apart: what modern processors do to alleviate the time it takes to fetch data is not only fetch more data than they might need, but actually do calculations on that data ahead-of-time. This is known as speculative execution, and it is the heart of these vulnerabilities. To put this analogy in algorithmic form:

  • Check the weather (execute multiple sub-processes that trigger sensors, relay data, etc.)
    • If the weather is sunny, wear shorts-and-t-shirt
    • Else wear jeans-and-sweatshirt

Remember, computers are stupid, but they are stupid fast: executing “wear shorts-and-t-shirt” or “wear jeans-and-sweatshirt” takes nanoseconds — what takes time is waiting for the weather observation. So to save time the processor will get you dressed before it knows the weather, usually based on history — what was the weather the last several days? That means you can decide on footwear, accessories, etc., all while waiting for the weather observation. That’s the other thing about processors: they can do a lot of things at the same time. To that end the fastest possible way to get something done is to guess what the final outcome will be and backtrack if necessary.

Meltdown

Now, imagine the algorithm was changed to the following:

  • Check your manager’s calendar to see if they will be in the office
    • If they will be in the office, wear slacks and collared-shirt
    • If they will not be in the office, wear shorts-and-t-shirt

There’s just one problem: you’re not supposed to have access to your manager’s calendar. Keep in mind that computers are stupid: the processor doesn’t know this implicitly, it has to actually check if you have access. So in practice this algorithm is more like this:

  • Check your manager’s calendar to see if they will be in the office
    • Check if this intern has access to their manager’s calendar
      • If the intern has access, access the calendar
        • If they will be in the office, wear slacks and collared-shirt
        • If they will not be in the office, wear shorts-and-t-shirt
      • If the intern does not have access, stop getting dressed

Remember, though, computers are very good at doing lots of things at once, and not very good at looking up data; in this case the processor will, under certain conditions, look at the manager’s calendar and decide what to wear before it knows whether or not it should look at the calendar. If it later realizes it shouldn’t have access to the calendar it will undo everything, but the clothes might end up slightly disheveled, which means you might be able to back out the answer you weren’t supposed to know.

I already said that the analogy was falling apart; it is now in complete tatters but this, in broad-strokes, is Meltdown: the processor will speculatively fetch and execute privileged data before it knows if it should or not; that process, though, leaves traces in cache, and those traces can be captured by a non-privileged user.

Explaining Spectre

Spectre is even more devious, but harder to pull off: remember, multiple users are using the same processor — roommates, if you will. Suppose I pack my suitcase the same as you, and then I “train” the processor to always expect sunny days (perhaps I run a simulation program and make every day sunny). The processor will start choosing shorts-and-t-shirt ahead of time. Then, when you wake up, the processor will have already chosen shorts-and-t-shirt; if it is actually rainy, it will put the shorts-and-t-shirt back, but ever-so-slightly disheveled.

This analogy has gone from tatters to total disintegration — it really doesn’t work here. Your data isn’t simply retrieved from main memory speculatively, it is momentarily parked in cache while the processor follows the wrong branch; it is quickly removed once the processor fixes it error, but I can still figure out what data was there — which means I’ve now stolen your data.

Meltdown is easier to explain because — Intel’s protestation to the contrary (Meltdown also affects Apple’s processors) — it is due to a design flaw. The processor is responsible for checking if data can be accessed, and to check too slowly, such that the data can be stolen, is a bug. That is also why Meltdown can be worked around in software (basically, there will be an extra step checking permissions before using the data, which is why the patch causes a performance hit).

Spectre is something else entirely: this is the processor acting as designed. Computers do basic calculations unfathomably quickly, but take forever to get the data to make those calculations: therefore doing calculations without waiting for bottlenecks, based on best guesses, is the best possible way to leverage this fundamental imbalance. Most of the time you will get results far more quickly, and if you guess wrong you are no slower than you would have been had you done everything in order.

This, too, is why Spectre affects all processors: the speed gains from leveraging modern processors’ parallelism and execution speed are so massive that speculative execution is an obvious choice; that the branch predictor might be trained by another user such that cache changes could be tracked simply didn’t occur to anyone until the last year (that we know of).

And, by extension, Spectre can’t be fixed by software: specific implementations can be blocked, but the vulnerability is built-in. New processors will need to be designed, but the billions of processors in use aren’t going anywhere. We’re going to have to muddle through.

Spectre and the State of Technology

I ended 2017 without my customary “State of Technology” post, and just as well: Spectre is a far better representation than anything I might have written. Faced with a fundamental imbalance (data fetch slowness versus execution speed), processor engineers devised an ingenious system optimized for performance, but having failed to envision the possibility of bad actors abusing the system, everyone was left vulnerable.

The analogy is obvious: faced with a fundamental imbalance (the difficulty of gaining and retaining users versus the ease of rapid iteration and optimization), Internet companies devised ingenious systems optimized for engagement, but having failed to envision the possibility of bad actors abusing the system, everyone was left vulnerable.

Spectre, though, helps illustrate why these issues are so vexing:

  • I don’t believe anyone intended to create this vulnerability
  • The vulnerability might be worth it — the gains from faster processors have been absolutely massive!
  • Regardless, decisions made in the past are in the past: the best we can do is muddle through

So it is with the effects of Facebook, Google/YouTube, etc., and the Internet broadly. Power comes from giving people what they want — hardly a bad motivation! — and the benefits still may — probably? — outweigh the downsides. Regardless, our only choice is to move forward.

I wrote a follow-up to this article in this Daily Update.

The 2017 Stratechery Year in Review

Posted on

The slogan for Stratechery’s sister podcast, Exponent, is “Tech and Society”; never has that felt more appropriate than 2017. This year I wrote 136 Daily Updates (including tomorrow) and 46 Weekly Articles, and, as per tradition, today I summarize the most popular and most important posts of the year: tech and society figure prominently.

You can find previous years here: 2016 | 2015 | 2014 | 2013

Here is the 2017 list.

The Five Most-Viewed Articles

Stratechery not only had record traffic in 2017, but year-over-year growth was also the highest ever; unsurprisingly, the first three articles were in Stratechery’s all-time top five in terms of traffic (Amazon’s New Customer was number one by a long-shot), and the other two in the top twelve.

  1. Amazon’s New Customer — The key to understanding Amazon’s purchase of Whole Foods is to understand that Amazon didn’t buy a retailer: the company bought a customer.
  2. Microsoft’s Monopoly Hangover — There are striking similarities between Microsoft today and IBM in the Lou Gerstner era, but today’s IBM should be a warning to Redmond.
  3. Alexa: Amazon’s Operating System — Money is made at chokepoints, and the most valuable chokepoints are operating systems; Amazon has built exactly that with Alexa.
  4. Facebook and the Cost of Monopoly — Facebook gave one of the worst keynotes in a long time: there was no vision, just the adoption of Snap’s. It’s the inevitable outcome of a monopoly.
  5. The Uber Dilemma — Benchmark’s lawsuit against Uber is extraordinary; that is because Uber, despite everything, remains an extraordinary company. Game theory explains the implications.

Five Big Ideas

These five posts very much capture the biggest themes of 2017: the power of aggregators, one of the most important acquisitions in media history, Bitcoin, sexual harassment, and Apple and China.

  • Defining Aggregators — Building on Aggregation Theory, this provides a precise definition of the characteristics of aggregators, and a classification system based on suppliers. Plus, how to think about aggregator regulation.
  • Disney and Fox — Disney’s rumored acquisition of 21st Century Fox is all about competing with Netflix; whether or not that is a good thing depends on your frame of reference. (See also, Disney’s Choice)
  • Tulips, Myths, and Cryptocurrencies — Did you hear the one about the tulip bubble? It’s almost certainly a myth. It is myths, though, that explain why cryptocurrencies are here to stay.
  • Goodbye Gatekeepers — Harvey Weinstein was a gate-keeper — a position that existed in multiple industries, including the media. That entire structure, though, is untenable on the Internet, and that’s a good thing.
  • Apple’s China Problem — Apple had mixed earnings: most of the world was great, but China was bad again. The reason is that in China WeChat matters more than iOS.

Five Posts About Technology and Society

The biggest theme of all, though, at least in technology, was the impact technology is having on society broadly, and how society might respond. At the center of everything was Facebook.

  • Manifestos and Monopolies — Facebook has long had too much power, but Mark Zuckerberg’s expressed willingness to use said power for political ends means it’s time to consider countermeasures (See also, the afore-linked Facebook and the Cost of Monopoly)
  • Tech Goes to Washington — Facebook, Google, and Twitter testified before a Senate committee: it provided evidence of how tech prefers power over decentralization, even if it means regulation.
  • Why Facebook Shouldn’t Be Allowed to Buy tbh — Facebook is acquiring tbh, another burgeoning social network; regulators erred in allowing the Instagram and WhatsApp acquisitions, but there is no better place to start enforcing the law than now.
  • Inspired Media — The key for products and politicians used to be maximizing paid and earned media. What matters on the Internet, though, is inspired media.
  • The Local News Business Model — Subscriptions are the future of local news: the key, though, is getting rid of newspapers (See also, Faceless Publishers)

See also: The Great Unbundling and the talk I gave at Recode Media:

The Year in Daily Updates

I am incredibly proud of the Daily Update this year: in my very biased opinion, in 2017 the depth and importance of Daily Updates easily matched Weekly Articles. Increasingly, Daily Updates were organized around a single company or topic, and often previewed themes that were later expounded on in Weekly Articles. Here are some of my favorites:

I also conducted four interviews for The Daily Update:

Finally, while I’m not sure if this will become an annual thing or not, for the first time I christened Tech’s Person of the Year: Susan Fowler. In addition, I have now made this Daily Update free to read — Fowler’s impact was extraordinary.

I can’t say it enough: I am so grateful to Stratechery’s readers and especially subscribers for making all of these posts possible. I wish all of you a Merry Christmas and Happy New Year, and I’m looking forward to a great 2018!