Vertical integration versus modularity

James Allworth places some of the blame for Boeing's 787 Dreamliner problems on Boeing's decision to modularize its production and design so early on.

In the creation of any truly new product or product category, it is almost invariably a big advantage to start out as integrated as possible. Why? Well, put simply, the more elements of the design that are under your control, the more effectively you're able to radically change the design of a product — you give your engineers more degrees of freedom. Similarly, being integrated means you don't have to understand what all the interdependencies are going to be between the components in a product that you haven't created yet (which, obviously, is pretty hard to do). And, as a result of that, you don't need to ask suppliers to contract over interconnects that haven't been created yet, either. Instead, you can put employees together of different disciplines and tell them to solve the problems together. Many of the problems they will encounter would not have been possible to anticipate; but that's ok, because they're not under contract to build a component — they've been employed to solve a problem. Their primary focus is on what the optimal solution is, and if that means changing multiple elements of the design, then they're not fighting a whole set of organizational incentives that discourage them from doing it.

Conversely, if you're trying to modularize something — particularly if you're trying to do it across organizational boundaries — you want to be absolutely sure that you know how all the pieces optimally work together, so everyone can just focus on their piece of the puzzle. If you've done it too soon and tried to modularize parts of an unsolved puzzle across suppliers, then each time one of those unanticipated problems or interdependencies arises, you have to cross corporate boundaries to make the necessary changes — changes which could dramatically impact the P&L of a supplier. Lawyers will probably need to get involved. So too might the other suppliers, who could quite possibly be required to change the design of their component, also (chances are, you've already contracted with them, too). The whole thing snowballs.

I just returned from Asymconf, and one of the things Horace Dediu likes to say is that vertical integration is the optimal strategy up until a product is good enough. Once it's good enough, then it's more ideal to modularize, or to start outsourcing more pieces of production.

Apple would seem to be taking the reverse approach with production of the iPhone. Earlier models were assembled from parts from a variety of suppliers, but for competitive reasons, especially vis a vis Samsung, Apple has started to bring more of the parts production in-house. Allworth believes the difference is that Apple "has mastered the art of managing design as an integrated process, while still utilizing outsourcing." It's also a strategy deeply ingrained in Apple's DNA.

One other company comes to mind, one that also takes a consistent vertical integration approach to their production in an industry in which the dominant model is more one of breaking production up. It is related to Apple and also located in the Bay Area. Any guesses?

Yep: Pixar.

You are not alone

I dig this live leaderboard of popularity of articles across all of Gawker's properties (the Big Board). It updates in close to real-time with a simple reader count. Clicking on any article brings up more information on engagement with that article.

Back when I was at Amazon.com, we had tinkered with the idea of making shopping on Amazon seem less lonely and more social, like shopping in the real world at a mall. That was back in an age before social networking services like Facebook and Twitter whose live feeds give one the sense of other people online alongside you as you browse. The Gawker implementation is another way for sites that aren't inherently social to achieve the same feeling.

Nowadays, with most of the studios licensing their movies to a whole slew of channels across the cable dial, movies seem to loop endlessly. Something's always playing on some channel. However, when I was in elementary school, back in the age when we only really had the three major networks on TV, back before DVRs and VCRs, ABC would occasionally air a James Bond movie, and it felt more momentous because I felt the presence of all the other viewers in the country watching that movie at the same time. No time-shifting, no DVD version to pop in at a later date. If you wanted to watch the movie, you had to sit there with everyone else in the country, and we were all watching the same moment in that movie. The attentional synchronicity felt magical, even if I'd never want to go back to that age of strictly scheduled entertainment.

At Hulu, when we were tinkering with design concepts for a next generation programming guide for television, we had played around with offering a channel/programming sort based on the volume of viewers at that very moment, with extra weight given to those programs being watched by your friends (according to your social network of choice). The Gawker Big Board reminded me of some of those concepts.

The best at what they do

The thought experiment is to compare players across sports. I.e., are basketball players better at basketball than, say, snooker players are at playing snooker?

Unless you count being tall as one of the things NBA basketball players “do” I would say on the contrary that NBA basketball players must be among the worst at what they do in all of professional sports. The reason is simple: because height is so important in basketball, the NBA is drawing the top talent among a highly selected sub-population: those that are exceptionally tall. The skill distribution of the overall population, focusing on those skills that make a great basketball player like coordination, quickness, agility, accuracy; certainly dominate the distribution of the subpopulation from which the NBA draws its players.

...

When you look at a competition where one of the inputs of the production function is an exogenously distributed characteristic, players with a high endowment on that dimension have a head start. This has two effects on the distribution of the (partially) acquired characteristics that enter the production function. First, there is the pure statistical effect I alluded to above. If success requires some minimum height then the pool of competitors excludes a large component of the population.

There is a second effect on endogenous acquisition of skills. Competition is less intense and they have less incentive to acquire skills in order to be competitive. So even current NBA players are less talented than they would be if competition was less exclusive.

That's Jeff Ely on whether basketball players are better at basketball than other athletes are at their sports (h/t Marginal Revolution). Click through to see which sports he considers to have the best at what they do.

It's commonly said that one reason the U.S. can't field a world class soccer team is that our best athletes go into football and basketball instead because they are more glamorous sports in the U.S. That might serve as an additional complicating factor.

The equivalent in the tech world seems to be that design talent disproportionately flocks towards consumer rather than enterprise apps and services, or so it's often said. Based on my experience using enterprise apps as compared to consumer apps, it certainly feels like the crucible of competition has run hotter on the consumer side, resulting in superior user experience.

It's good to be (one of several) kings

Yesterday I got a text message from AT&T today saying I'd just gone over $20 worth of text messages this month and that it would've been better for me to have paid $20 for one of their monthly text message subscriptions.

Most months, paying a la carte for text messages has worked out for me, but not this month. iMessages hasn't worked as often as it has in the past so some of my text messages to and from Apple iMessages users have ended up going the conventional route, and I was at Sundance where I received a ton of group texts from some Android users in my house.

Whenever I express my disgust at paying AT&T for text messages, people bring up the ridiculous mismatch between the cost of transmitting that infinitesimal volume of data over the telecom network and the actual cost they charge consumers. It's a mistake to think of text messages as goods that are priced on a cost-plus basis.

Telecom network operators invest a lot to build out their network. Once they have the infrastructure in place, and especially if they are part of an oligopoly, which many telecom operators were for quite some time, they can really charge whatever the market will bear for their goods. They're not looking to take the cost of a text message and mark it up by some percentage. They're looking to make as much as possible off of each of their subscribers across voice, data, text messages, and all the other services they offer.

Consumers routinely make the mistake of thinking the cost of their data, voice, and messaging plans are tied much more closely to the cost of delivering those services than they actually are. The pricing plans themselves contribute to that confusion because the more messages or data you want, the more you pay.

AT&T and Verizon no longer offer unlimited data plans anywhere in their cellular packages, and they'll blame people overloading their networks for doing so, but the truth is they are just pricing things to extract more revenue from their customer base.

This is the same economics mistake Brian Stelter of the NYTimes makes in his article Rising TV Fees Mean All Viewers Pay to Keep Sports Fans Happy. On the surface, it seems like he's right. Since ESPN is the most expensive channel for cable operators to carry (at last check when I was at Hulu, the SNL Kagan rate for ESPN was something like $5 per subscriber per month, by far the priciest of all cable channels), it seems like people who don't watch ESPN are paying to subsidize those of us who do watch ESPN.

However, almost everyone I know only watches a small fraction of the hundreds of channels they have to pay for in the cable bundles available to them. You could make arguments that ESPN viewers are subsidizing people who watch Lifetime and the History Channel.

Cable and satellite companies, like telecom network operators, have built out this expensive network infrastructure, and they are also fortunate to be part of a legal oligopoly. They work hand in hand with content owners to push the prices of your cable subscription up every month, and until now, the cord-cutting has been minimal. As your monthly cable package prices goes up, the cable/satellite guys make more money, and they hand more of it over to the content owners. Why wouldn't they both continue to squeeze more money from consumers as long as possible?

The same applies for your broadband internet access. The cost of your monthly internet access isn't based on the cost of delivering that bandwidth to you. The margins on broadband internet have been estimated to be upwards of 90% to 100%. Most of that infrastructure was paid off years ago from cable customers. Carrying internet data over the same cables and being able to charge for that service is just gravy. They can charge those extortionate rates because most consumers have very few broadband alternatives, if any, where they live.

Trying to lower the amount you pay for monthly text messaging, cellphone voice and data, cable/satellite subscription, and your internet broadband is almost impossible for consumers. If enough consumers started to cancel their text message plans because of iMessages or WhatsApp or something else, AT&T would just gouge us for revenue another way to make it up, maybe by raising the price of data or voice plans. 

The only reliable way to drive down price of these services is to offer real competition to these legal oligopolies. The FCC really failed U.S. consumers in allowing them to end up beholden to these communication oligopolies (enterprising reporters out there, this is a subject worth a long article). To build out the types of infrastructure to compete with these companies is priced beyond the hope of all but a handful of companies.

Disruption has not come as quickly as we all would love. We all thought we'd be watching content for much less on our Google or Apple TVs, or that we'd be making voice calls over some ubiquitous wi-fi network blanketing the country, and it would all cost much less than we'd paid for the equivalent services before.

Failure just forces disruptors to be more creative, though, and businesses with outrageous margins like these oligopolies have tend to fall prey to the barbarians at the gate in the totality of time. I've recently seen some companies starting to take new attack vectors on the incumbents, and I suspect the challengers will be more successful this time around. That's a subject for another day, another post.

UPDATE: Timely, but perhaps some proof that competition is the only way to get these oligopolists to lower their prices: a Time Warner broadband customer living near the Google Fiber installation in Kansas received an unsolicited speed bump and price drop.

Now that the United States has the world’s highest reported rate of incarceration, many criminologists are contemplating another strategy. What if America reverted to the penal policies of the 1980s? What if the prison population shrank drastically? What if money now spent guarding cellblocks was instead used for policing the streets?

In short, what would happen if the rest of the country followed New York City’s example?

As the American prison population has doubled in the past two decades, the city has been a remarkable exception to the trend: the number of its residents in prison has shrunk. Its incarceration rate, once high by national standards, has plunged well below the United States average and has hit another new low, as Mayor Michael R. Bloomberg announced recently. And crime in the city has fallen by more than 75 percent, almost twice as much as in the rest of the country.

Whatever has made New York the safest big city in America, that feat has certainly not been accomplished by locking up more criminals.

From a NYTimes article studying New York City's success in both shrinking its prison population and lowering crime. As criminologist Lawrence Sherman notes in the article, the U.S. is the only country he knows of that spends more on incarceration than police. America now has a fifth of the world's prisoners.

One specific policing strategy that has yielded fruits is hot-spot policing, or focusing less on individual criminals but instead on areas where they tend to work.

“Crime doesn’t move as easily we thought it did,” Mr. Gajewski said. “If I’m a robber, I want to be in a familiar, easily accessible place with certain characteristics. I need targets to rob, but I don’t want people in the neighborhood watching me or challenging me. Maybe I work near a bus stop where there are vacant buildings or empty lots. If the police start focusing there, I can’t just move to the next block and find the same conditions.”

I wish San Francisco would spend more on police. For a few months, I worked at an office in SOMA that required me to walk through some of its sketchier blocks, and even as a man I didn't feel comfortable walking home late at night. Even during the daytime I'd sometimes encounter some crazy, scary characters, from an old bearded man who would curse me out for no reason to a gaunt, pale woman who often smoked a crack pipe and who could've passed for an extra on American Horror Story: Asylum.

Just in the past year, five of my friends have had their car windows smashed in and items stolen from their car while parked on the streets of San Francisco. Small sample size, sure, but I felt safer in Manhattan than I do in San Francisco. It honestly feels like San Francisco has just decided to let parts of the Tenderloin be our Hamsterdam.

We need to create more jobs anyhow, I'd be willing to take lower incarceration to free funds to hire more police.