Retrospective on a dying technology cycle, part 3: Venture Capital and an economy on life support

20 Mar 2023

After leaving my job to build apps in 2009, life took on some hard math. I only had so much money saved up. While I was making around $1,000 monthly from the App Store, the gap between my income and expenses was leaving me in the red.

So my maximum budget for a meal was $3. If I could get under that number, I tried.

This meant clever use of bulk buying at Costco and lots of meal prep. Going out for meals was an occasional indulgence, but the three dollar figure remained a constraint. Spending more than that meant having to bring home enough leftovers to break even.

I’d figured I could make some extra money working part time while I got the details of my app business figured out. To this point in my life, part-time work had been easy to come by. I took it for granted, like water from the tap. It would be there if I needed it.

But the financial crisis meant stiff competition for any work. I’d moved to Bend, Oregon for its low cost of living and beautiful scenery. Unfortunately, no one was hiring, and as the new kid in town, I had no local network to lean on. When a Kohl’s opened up, it made the evening news: newly laid-off job applicants wrapped around the block hoping for a chance at even 15 hours of work per week.

I wasn’t just competing with part-time job seekers. I was competing with people who needed any work they could get to keep a roof over their families’ heads.

It was a scary, precarious time. Unemployment peaked around 10% nationwide, even greater than in the dot-com bust, with millions of workers unable to find jobs. I came within weeks of losing my housing.

This desperate climate catalyzed the fuel for the last technology cycle. Central bankers around the world slashed interest rates nearly to zero. Without traditional means generating returns on their money, pensions, institutional investors and the wealthy sought new vehicles for growth.

The Venture Capitalist had just what they needed. On behalf of these limited partners, a VC would secure a packet of investments in technology firms, one or more of which might just burst into a multi-billion dollar valuation.

They called these “unicorns.”

Cheap scale

Software solves a problem through automation.

When you build a spreadsheet to solve your reporting needs, you’ve just made yourself some software. It’s got decent leverage: instead of manually filling out a chart, some formulas summarize your data and nicely format it, saving you a few hours a week.

You make a trade for this leverage. Making the report manually maybe costs you an hour a week. Building automation into the spreadsheet might take you much longer: a couple of afternoons. But within a few months, the leverage has paid for itself in time savings. Within a year, you’re ahead.

This simple example captures the economic value of software, and its potential scales well beyond an individual’s problem at the office. Whatever costs you sink in creating software, you get many multiples back in whatever work it accomplishes.

You can make office productivity tasks more efficient, you can build social interaction that promotes a particular style of engagement, you can deliver entertainment through games or other media, and you can even make cars show up in particular places where people want them.

Any task that software can do once, it can again as many more times as you want. On paper, this happens at near-zero marginal cost.

In practice, what this means is complicated. As Twitter’s Fail Whale demonstrated, just because scale is cheap doesn’t make it free. Architecting and re-architecting software systems to be resilient and responsive requires specialist knowledge and significant investment as your customer base transitions from one order of magnitude to the next.

Copying code costs almost nothing. Running code on the web, and keeping it running, is more expensive. It requires infrastructure to execute the code, and it requires human labor to troubleshoot and maintain it in flight.

The bargain of Venture Capital

This is where the Venture Capitalist comes in.

Writing code is expensive: it needs a full-time team of specialist engineers who know their way around the required technologies. The engineers also need to know what to build, so other specialists, like designers and strategists, are needed to document and refine the goals and end-product.

The code continues to be expensive once it’s written. Again, information is physical: it lives somewhere. The data that software handles has a physical location in a database and has to be moved to and from the user, who may be quite far away from the server they’re interacting with.

Thus, more specialist labor: engineers who can design technical infrastructure that gracefully avoids traffic jams as a web-based service becomes popular. Of course, there are more specialists after that. People to tell the story of the software, people to sell it, people to interact with the customers and users who are struggling to use it.

Code needs humans. It can’t be born without them, it needs them to grow, and it can’t stay running forever in their absence. Automation is powerful, but it’s never perfect, and it exists in a human context.

Humans are expensive. We need shelter, food, rest, recreation, partnership, indulgences, adventure… The list of human needs is vast, and like it or not, we meet those needs with time and money.

So the Venture Capitalist buys a chunk of a growing software firm, injecting it with money in trade. The money can be used more or less freely, but much of it goes into paying for humans.

Despite plunging tens or hundreds of millions of dollars into a firm, if the VC does their job right, they come out ahead. Software that can solve problems effectively, that captures a customer base that will pay for those problems to be solved, can produce billions of dollars in value. So long as the lifetime value of a customer is meaningfully greater than what it costs to earn them, you've got a powerful business.

So spending a few million to support an infant firm can turn into orders of magnitude more money down the road.

Not every startup achieves this success. These are outlier results. The majority will fail to achieve the scale and value needed for a win, in terms of venture expectations. But when the outliers are that lucrative, the overall model can work.

So long as you pick an outlier.

Going whackadoodle

GitHub was unique. Founded in 2008, GitHub was profitable almost immediately, without needing any venture funding. They are perhaps the ultimate modern success story in software tools:

GitHub simultaneously defined and seized the central turf of developer collaboration.

Their strategy was devastatingly effective. Git was an emerging technology for version control, which every software project large and small needs. Version control maintains a history of project progress—who changed what, on which date, and why—while allowing developers travel back in time to a previous state in the project.

Historically, developers needed an internet connection to interact with the history of a project, so they could connect with a server that stored all of the project data. Git’s great innovation was to decentralize this process, letting developers work in isolation and and integrate their changes with a server at their leisure.

Originally developed to serve the needs of the Linux kernel project, one of the largest collaborations in open source history, git had both enormous power and daunting complexity. Still, git would be the future not just of open source, but of software development altogether.

Seeing this, GitHub’s founders built a service that provided:

A central server to host git-based projects, called repositories
Convenient user interfaces to browse, change and duplicate repository content
Collaboration tools to discuss, improve and ship code with both teammates and strangers

On the strength of this developer experience, GitHub fast became the most popular way to build software, no matter the size of your team or your goals.

Its power and value earned it leverage. Four years into its life, it raised money on wildly favorable terms. It was a safe bet: enormous traction, plenty of users, massive goodwill, and a whole enterprise market waiting to be conquered.

Then its cultural debts caught up with it.

There’s a situation here

GitHub built a replica of the Oval Office. They also built a replica of the White House Situation Room. Then there was the open bar, offering self-serve alcohol at any hour of the day. The rest of the space was of the typically fashionable vibes you’d expect of a startup, but these indulgences caused onlookers to raise their eyebrows.

Many clowned GitHub for the White House stuff—myself included—but here is a place where I will now defend their intent. America is lots of things, and primarily an empire. The symbols of an imperialist state have a lot of complicated (that is, bloody) history and it’s a peculiar sort of hubris to adopt them whole-cloth.

But America, in its most aspirational of modes, sees itself as a fabric for creating prosperity and shared growth. Having spent a couple years there, I really think that’s where GitHub was coming from in its office cosplay decisions.

They wanted to be a fertile landscape for shared prosperity, and picked up the most garish possible symbols of that aspiration. It’s not how I would have come at it, but I get it how they landed there. It’s impossible to imagine the last chapter of software without GitHub as a productive substrate for all the necessary collaboration.

This is a tidy microcosm for the cycle. Incredible ambition and optimism, tempered by social challenges few founding teams were equipped to address. An industry whose reach often exceeded its grasp.

GitHub’s blindspots, coupled with its incredible early power, would conspire to make it both a distressed asset and a notorious place to work, claiming one of its founders in the process.

The mess in its workplace culture made it challenging for the company to recruit both the experienced leadership and specialist labor needed to get it to the next phase of its evolution.

GitHub did the work to patch itself up, recruiting the team it needed to get back to shipping and evolve into a mature company. With the bleeding stopped and stability restored, Microsoft purchased the company for—you guessed it—billions of dollars.

This stuff was not unique to GitHub

Zenefits had people fucking in stairwells, Uber spent $25 million on a Vegas bacchanal, and WeWork… they did a whole a docudrama about WeWork.

With nowhere else for money to go, successful founders had a lot of leverage to spend as they pleased. VCs needed to compete for the best deals—not everything would be a unicorn, after all—and so the “founder-friendliness” of a firm was a central consideration of investor reputation. No founder wanted the money guys fiddling around and vetoing their vision and VCs feared losing out on the best deals.

So investors kept a light touch on the day-to-day management of firms, so long as they were pursuing a strategy that kept hitting their growth metrics.

Meanwhile, you’ll notice I keep emphasizing specialist labor. Some problems in technology are particularly challenging, with a limited number of practitioners able to make meaningful traction on them. Thus, startups had to work hard to recruit their workforce. Founders would argue here, with a certain degree of legitimacy, that differentiating their experience of work through generous perks was an essential lever in winning the competition for limited talent.

Thus was the frenzy for hyper-growth that continued for more than a decade. Money pumped into the system, searching for its unicorns. Founders with more technical acumen than management experience tried to recruit the teams they needed. Investors kept the show running, hoping to grow money for their limited partners.

But this was mere prelude to the wackiest money of all.

Crypto craze

The way venture historically worked:

Invest in a young company
Support it through relationships and more cash, if its progress was promising
Wait around for quite awhile for it to reach maturity
Get paid upon a “liquidity event”—when the company goes public (lots of work) or is acquired (a bit less work)

Cryptocurrency, non-fungible tokens, and other so-called web3 projects were an irresistible optimization on the normal cycle. Going public required stability, clean accounting, a proven record of growth, and a favorable long-term outlook for a company’s overall niche. It required stringent regulatory compliance, as a company transformed from a privately-held asset into a tradable component of the public markets.

Keeping a financial system healthy takes work, after all, and bodies like the SEC exist to ensure obviously-toxic actors don’t shit in the pool.

Less arduous was an acquisition. Sometimes these were small, merely breaking even on an investor’s cash. Still, multi-billion dollar acquisitions like Figma and GitHub do happen. To integrate into a public company like Microsoft or Adobe, a startup needs to get its house in order lots of ways. Accounting, information security, sustainable growth—it takes time to get a company into a level of fitness that earns the best price.

Cryptocurrency investment promised to short-circuit all of this.

Investors could buy “tokens” as a vehicle for investing in a company. Later, when those tokens became openly tradable—through a mostly unregulated, highly-speculative market—investors could cash in when they decided the time was right.

Rather than waiting seven or ten years for a company to reach maturity, investors in crypto firms could see liquidity within half that time.

With cloud services full of incumbents, while mobile matured, it was hoped that cryptocurrencies and web3 would represent an all new paradigm, starting a new cycle of investment and innovation.

More than a decade since the 2008 crisis, central bankers had to pare interest rates back down to zero once again, this time in response to the Covid-19 pandemic. This created a fresh rush of money once again in search of growth, and cryptocurrency-adjacent companies were happy to lap it up.

What comes next

Of course, we know how this ends. Covid-era inflation soared, and central bankers cranked rates back up to heights not seen since before 2008.

Crypto came crashing back to earth, NFTs became a tax loss harvesting strategy, and now we muse about what else was a “zero interest rate phenomenon.” I’d be shocked if crypto was entirely dead, but for the moment, its speculative rush has been arrested.

In the next and final installment, I want to take stock of what we can learn from all this, and what comes next.

Retrospective on a dying technology cycle, part 2: Open Source and the Cloud

17 Mar 2023

[Previously: Mobile]

Though it rescued me from insolvency, I grew to hate my job at Aurora Feint. Two months into the gig, Apple Sherlocked us, sending the company into a series of endless pivots that lasted four to six weeks each.

The engineers were crunching, working evenings and weekends, only to throw away their work and start again, seemingly based on how well one press release or another gained attention. Meanwhile, I was far from the action in an ill-defined product management role. While I knew the emerging mobile product space as well as anyone, I had limited vision into the various rituals of Silicon Valley.

What I really wanted was to build mobile software.

I interviewed for months, but with no pedigree and a strange path into technology through self-teaching, I didn’t get very far.

After a particularly dispiriting day at the office, I sniped the attention of Adam Goldstein and Steve Huffman, then founders of a YC-funded travel search startup called Hipmunk. I wrote a flattering blog post about their product, posted it to Hacker News, and within weeks had a new job: leading mobile product design and engineering for a seed-stage startup.

They never asked for a resume. They checked out my existing work on the App Store and made me an offer.

I was the third employee. The entire company of five could fit into a San Francisco apartment's living room. And did, regularly, in its first few months.

Despite the small crew and modest quarters, thousands of people were buying airfare through Hipmunk already. Its great innovation was visualizing your flight options on a colorful Gantt chart, ranked by “Agony”—factors like layovers and overall flight time.

Two things gave this team enough leverage to do it: open source software and cloud computing. A credit card was all they needed to provision server capacity. An internet connection gave them access to all the software they needed to make the servers productive, laying the foundation for building Hipmunk's unique approach to finding flights.

What is Open Source, anyway?

Open source has permanently changed the nature of human productivity.

Open source describes code you can use, according to some basic licensing terms, for free. Vast communities may contribute to open source projects, or they may be labors of love by one or a handful of coders working in their spare time.

In practice, open source code is encapsulated labor.

One of the fundamental economic levers of software in a globally-connected community is that it scales at zero marginal cost. Once the software exists, it’s nearly free duplicate it.

Thus, if you have software that solves a common problem—provides an operating system, handles web requests, manages data, runs code—you can generalize it, and then anyone can use it to solve that category of problem in their own project. This combination of open source projects, called a “stack,” is sometimes packaged together.

Such open source stacks made the last cycle possible. The Facebook empire could be founded in a dorm room because Zuckerberg had access to Linux (OS), Apache (web server), MySQL (database), and php (language/runtime) as a unified package that solved the problem of getting you productive on the web.

This represented tens of thousands of hours of labor that startups don’t have to recruit or pay for, or even wait for, dramatically accelerating their time to market.

The details of serving HTTP requests or handling database queries aren’t central to a startup’s business value—unless you're making a platform play where your performance on these categories of tasks can be sold back to other firms. But for most, there’s no advantage to reinventing those wheels, they just have to work well. Instead, borrowing known-good implementations can quickly provide the foundations needed to build differentiated experiences.

In Facebook’s case, this was a minimalist social platform, for Hipmunk this was visual travel search, while Uber dispatched cars. There was no innovation to be found merely existing on the web. Instead, to survive you had to be on the web while also providing unique, differentiated value.

Open source is an informal commons of these known-good implementations, regularly refined and improved. Any party finding a project lacking can lobby for refinement, including proposing direct changes to the code itself.

I would argue there is societal impact from this that rivals other epochal moments in information technology, like the advent of writing or the discovery of algorithms. Being able to store, replicate and share generalized technical labor is a permanent reorganization of how millions of people work. While capitalists were among the first to seize on its power, over decades and centuries it may permeate much more than business.

The power of open source was constantly animating the last cycle. Node.js, a server-side runtime for JavaScript first released in 2009, took one of the most well-known languages in the world and made it viable for building both the user interface components of a web application and its backend server as well. Thus open source could not just encapsulate labor, but amplify it in motion as well. Suddenly, many more people knew a language that could be used to build complex web services. As a result, Node became among the most popular strategies hobbyists and startups alike.

Meanwhile, it wasn’t enough for code to be easy to build on. The code actually has to live somewhere. While open source was a crucial engine of the last cycle, cloud computing running this open source code was just as essential.

It’s in the cloud

Information is physical.

It needs somewhere to live. It needs to be transformed, using electricity and silicon. It needs to be transported, using pulses of light on glass fibers.

Software requires infrastructure.

Historically, this had meant shocking amounts of capital expenditures—thousands of dollars for server hardware—along with ongoing operating expenditures, as you leased space in larger facilities with great connectivity to host these servers.

Scaling was painful. More capacity meant buying more hardware, and installing it, which took time. But the internet is fast, and demand can spike much more quickly than you could provision these things.

Amazon, craving massive scale, had to solve all these problems at unique cost. They needed enough slack in their systems to maintain performance at peak moments, like holiday sales, back-to-school, and other seasonal surges. With the size of Amazon's engineering division, meanwhile, this capacity needed to be factored into tidy, composable services with clear interfaces and rigorous documentation, allowing any team to easily integrate Amazon's computing fleet into their plans and projects.

Capital abhors an under-leveraged asset, so they decided to sell metered access to these resources. Though Amazon was the first to use this approach, today they compete with Microsoft, Google, IBM and others for corporate cloud budgets all over the world.

With access to “the cloud” just a credit card away, it’s easy for small teams to get started on a web-based project. While scaling is commercially much easier than ever—you just pay more to your cloud vendor, then get more capacity—it can still be a complex technical lift.

The arcane art of distributing computing tasks between individual machines, while still producing correct, coherent output across large numbers of users, requires specialist labor and insight. The cloud isn’t magic. Teams have to be thoughtful to make the most of its power, and without careful planning, costs can grow devastating.

Still, this power is significant. The cloud can itself be programmed, with code that makes scaling decisions on the fly. Code can decide to increase or trim the pool of computing capacity without human intervention, adjusting the resources available to create maximum customer impact and system resiliency. Netflix is a notable specimen on this point, as one of the earliest large-scale adopters of Amazon’s cloud services. They created an infrastructure fabric that is designed to tolerate failure and self-heal. They even built a technology called “Chaos Monkey” to continually damage these systems. This forces their engineers to build with resilience in mind, knowing that at any moment, some infrastructure dependency might go offline.

(Chaos Monkey is, naturally, available as open source.)

For companies in infancy, like Hipmunk, the cloud opened the path to a fast-iterating business. For companies entering maturity, like Netflix, the cloud provided a platform for stability, resilience, global presence, and scale.

The world had never seen anything like it. In fact, there was early skepticism about this approach, before it came to dominate.

Today, the cloud is a fact of life. Few would even bother starting a new project without it. What was once transformational is now priced into our expectations of reality.

Hipmunk would not become a galloping success. It sold to Concur in 2016, eventually scrapped for parts. Steve Huffman would return to lead Reddit, his first startup. Reddit sold to Condé Nast for peanuts, only to spin back out as an independent venture as, stubbornly, it just wouldn’t die.

Meanwhile, six months after my departure, the renamed OpenFeint was acquired for $100m by a Japanese mobile company in need of next-generation mobile technology. I didn't get a dime of the action—I left before my cliff kicked in, and I couldn't have afforded to exercise the options even if I they were open to me.

That’s the circle of life in venture capital. Not every bet is a winner, death is common, and sometimes acquisition is bitter.

But when the winners win, they can win big. After a couple of years, I left Hipmunk for a couple more gigs leading mobile teams. Under all the VC pressure and SF's punishing cost of living, I burned out hard.

Somehow, barely able to get out of bed most mornings, I landed on a unicorn. In the next piece, we’ll explore how venture capital was the fuel of the last cycle, creating growth in a zero interest rate global economy by harnessing startups that, impossibly, created billions of dollars in value.

Retrospective on a dying technology cycle, part 1: Mobile

16 Mar 2023

I missed the introduction of the original iPad.

Instead of watching Steve Jobs unveil his bulky slab of aluminum and glass, I was picking my way through the streets of downtown Burlingame. I’d spent what was left of my available credit card balance on a ticket to SFO.

I was due for my first tech interview.

Before he created the chat juggernaut Discord, Jason Citron founded a mobile game startup. After middling success selling games themselves, Aurora Feint had pivoted to middleware: client- and server-side social technology to serve the growing ranks of independent, mobile game developers.

A few blocks’ walk from the Caltrain station, I’d found my destination: a plain office building hosting a bank and an insurance broker on the lower floors. Hidden away upstairs, thirty employees of Aurora Feint were toiling away, trying to capture turf in the unspooling mobile revolution.

The interview was a grueling, all-day affair running 10:30 to 4, including one of those friendly lunches where the engineers try to figure out if you’re full of shit.

It was 2010, and the inertia of the financial crisis was still on me. Weeks away from homelessness, I somehow closed the day strong enough to earn my first startup job.

As a result, I got to watch the last technology cycle from its earliest phases.

It transformed my life. I wasn’t the only one.

Animating forces

The last cycle was activated by convergent forces:

The mobile revolution: An all new set of platforms for communication, play and productivity with no incumbents. A green field for development and empire-building
The rise of cloud computing: In the past, computing resources were expensive and provisioning was high-friction. By 2010, cloud computing offered low-risk, highly-scalable capacity for teams of any size.
The power of open source software: Open source transformed the economics of building a software business. A vast, maturing bazaar of software resources, free to use, gave founding teams a head-start on the most common tasks. This amplified their power and allowed small seed investments to produce outsized results.
An economy on life support: In the wake of the 2008 financial crisis, economic malaise dominated. Central bankers responded with historically low interest rates. Growing money needed a new strategy, and the growth potential of software empires was irresistible.

Each of these, alone, promised incredible power. In combination, they were responsible for a period of growing wealth and narrow prosperity without precedent. As the cycle draws to a close, we find each of these factors petering out, losing sway, or maturing into the mundane.

As a consequence, the cycle is losing steam.

This wasn’t a bubble—though it did contain a bubble, in the form of cryptomania. The last cycle was a period of dramatic, computing-driven evolution, whose winners captured value and power.

New platforms and a digital land rush

The 2007 introduction of the iPhone is a pivotal moment in the history of consumer computing. An Apple phone had been rumored for years, and on the heels of the popular iPod, it was an exciting prospect.

Mobile phones of the day were joyless, low-powered devices that frustrated users even for the most common tasks. While Palm, Blackberry and Danger were nibbling around the envelope of what was possible, none of them had command of the overall market or culture of cellphones.

So when Jobs, barely suppressing a shit-eating grin, walked through the demo of Apple’s new iPhone—easy conference calling, interactive maps, software keyboard, convenient texting—the audience gasped audibly, cheering with unrestrained delight. The user experience of the cell phone was changed forever.

To accomplish this transformation, Apple had pushed miniaturization as far as possible for the day. Using a high-efficiency processor built on the ARM architecture was a typical move for mobile computing—even Apple’s failed Newton from 15 years earlier used this approach. The revolution came from porting the foundations of Apple’s existing operating system, Mac OS X, into a form that could run within the narrow headroom of an ultra-efficient, battery-powered portable computer.

On this foundation, Apple shipped a cutting-edge UI framework that supported multitouch gestures, realtime compositing of UI elements, lush animation, and a level of instant responsiveness that had never existed in a handheld context.

While the iPhone wasn’t an overnight success—it would take years to capture its current marketshare—it was an instant sensation, eventually demolishing the fortunes of Palm, Blackberry, and Microsoft's mobile strategy.

And without Apple’s permission, people began writing code for it.

The promise of a vast green field

Hobbyist hackers broke the iPhone’s meager security within months of its launch, and went straight to work building unauthorized apps for it. Regardless of Apple’s original plans, the potential of the platform was unmistakable, and within a year of its 2007 introduction, both the iPhone OS SDK and App Store launched.

This was yet another sensation. Indie developers were crowing about shocking revenue, with the earliest apps taking home six figures in the first month. Apple had created a vast, fertile market with turn-key monetization for every developer.

The rush was on.

Google soon joined the fray with Android, offering their own SDK and marketplace.

So by 2010, mobile was a brand new platform strategy. It used new libraries, targeted new form factors, and had all new technical constraints.

There were no incumbents. Anyone could be a winner in mobile, so loads of people tried. Moreover, the use-cases mobile presented—always connected, with realtime geolocation and communication—were entirely novel.

The story of Uber is impossible without the substrate of mobile. March of 2009, 18 months into the launch of the iPhone OS SDK, Uber was hard at work building unprecedented, automated infrastructure for car dispatch. The iPhone 3G, the most recent model of the day, had integrated cell tower-assisted GPS, allowing any device to know its exact position in realtime, even in urban areas, and report that information back to a central server. Add this to an advanced mapping API included in every iPhone, and Uber had leverage to build a category of business that had never existed.

We know the rest of the story. Uber transformed urban transportation worldwide, adopting a take-no-prisoners approach and burning through vast sums of cash.

The gig economy and the automated supervisor

Of course, Uber wasn’t the only player leveraging this power to the hilt.

Lyft and other “rideshare” startups built their own spins on the model. But the power of automation delivered through pocket-sized, always-connected computers created a new category of labor and business, which the press dubbed “the gig economy.”

Using software automation, companies could animate the behavior of legions of dubiously-categorized independent contractors. Software told these workers what to do, where to go, and when they were needed. Software onboarded these workers, software paid them, and using ratings and algorithms, software could fire them.

In an unexpected but inevitable twist of Taylorism, modern automation replaced not the worker themselves but middle managers.

Gig work created new categories of service, and invaded existing ones as well. There were apps for delivery of all foods, apps for cleaning and home handiwork, even an app that handled packing and shipping of goods, sparing you a trip to the post office. Seeimingly every commercial relationship where Silicon Valley could interpose itself, through the lure of convenience, it tried.

Reporting not to other humans but faceless, algorithm-driven apps, gig workers were precarious. Their pay was volatile, their access to opportunity was opaque. The capricious whims of customers and their ratings—a central input to the middle-management algorithms—dictated their ongoing access to income. They had limited redress when things were bad for them because the apps that defined their daily work weren’t built to express curiosity about their experiences.

And they had no direct relationships to their customers.

Gig work companies responded to these complaints by touting the flexibility their platforms offered. The rigid scheduling of legacy labor was replaced by the freedom to work when you wanted to, even at a moment’s notice.

That was just one of many opportunities created by an always-online, always-with-you computer.

The attention economy

In 1990, the only way to reach someone waiting in line at a bank was to put a poster on the wall. Most banks weren’t selling wall space, so this was their exclusive domain to sell financial products like small business loans, mortgages and savings accounts.

20 years later, a customer waiting in line at a bank had a wealth of smartphone apps competing for their attention.

Social media platforms like Facebook and Twitter were in their mobile infancy, but by 2010 anyone with a smartphone could tune out their immediate surroundings and connect with their friends. It wasn’t just social platforms, either. Mobile games like Words With Friends, Angry Birds, and Candycrush competed vigorously to hold the attention of their enthralled users.

Mobile computing created an entirely new opportunity for connecting to humans during the in-between times. Historically, digital businesses could only access humans when they were at their desks in their homes and offices. With mobile, a 24/7 surface area for commerce, advertising, culture and community arrived overnight.

Mobile ossification

So we see that the new paradigm of mobile computing opened the door to previously impossible feats. It was a space where the primary obstacles were time and funding to create something people wanted.

Today, mobile is full of incumbents. Instead of a green field, it’s a developed ecosystem full of powerful players worth trillions in total. It is a paradigm that has reached maturity.

More than a decade after smartphones became a mass-market phenomenon, they’ve grown commonplace. Yearly iterations add marginal improvements to processing speed and incremental gains on camera systems.

When Facebook craves the metaverse, it’s because their totalizing strategy for growth—two billion users to date—is approaching a ceiling. They badly want a new frontier to grow in again, and they’d love to control the hardware as much as the software.

But VR doesn’t solve everyday problems the way smartphones did. No one craves a bulky headset attached to their face, blocking reality, with battery life measured at a couple hours. Augmented reality might have greater impact, but miniaturizing the technology needed to provide comfortable overlays with the same constant presence as the smartphone in your pocket is still years away. Could it be less than five? Maybe, but I wouldn’t bet on it. More than ten? Unlikely.

That range doesn’t present immediate opportunities for new growth on the scale they need. Indeed, Facebook is retreating from its most ambitious positions on the metaverse, as its strategy committed billions to the effort for limited return.

There will be new paradigms for mobile computing. But we’re not there yet. Even so, Apple—16 years gone from their revolutionary iPhone announcement—is poised to release their own take on VR/AR. And like Facebook, they're stuck with technological reality as it is. Perhaps their software advantage will give them the cultural traction Facebook couldn't find.

In my next post, I’ll get into the twin engines of cloud computing and open source, which were crucial foundations to the mobile revolution.

Native code wasn’t enough. True power was orchestration of opportunity between devices, which needed server-side technology. Through the power of open source code running in the cloud, any team could put together a complete solution that had never been seen before.

Developer success is about paying the rent

18 Feb 2023

I look at developer experience as a nested set of microeconomic problems. A well-optimized DX makes the most of the individual and team labor needed to:

Learn developer tools, from APIs to design patterns to integration strategies
Apply these tools to the developer’s specific domain and goals
Debug outcomes to arrive at a reasonably robust, predictable, correct implementation
Successfully ship artifacts on whatever terms the developer cares about (creative, commercial, collaborative)
Iterate on these artifacts to refine their effectiveness and power
Collaborate across all these facets, either as a distinct team, or with fellow developers on the internet working on distinct projects

In short, DX is about creating the conditions for developers to be successful.

There are many ways, meanwhile, to define developer success. But for the specific context of a developer tools business, I would argue the definition is simple: developer success is a rent check that clears.

While times are turbulent in the technology industry, the fact remains that software skills are a high-leverage form of labor. When you create software, you’re creating impact that can scale at near-zero marginal cost, serving dozens of customers up to billions. This has power and the need for it exists across every industry sector.

Every business needs software, and most businesses will use it in some form to deliver value to their customers. To accomplish this, they hire the best developers their labor budgets, networks and cultures can attract.

Thus, if you’re making devtools for money, you have to understand how your tools are competing against alternatives that make developers more effective at creating results that pay their rent and advance their careers.

Onboarding is an essential lever in this equation. If the activation energy needed to become productive in your tool is too great, it’s going to lose against alternatives that provide more immediate results. You want the time between a developer reading your marketing page, and that same developer making their first API call, successful build, or other hello-world-like activity, to be as short as possible.

A few minutes is ideal.

A few seconds is even better.

But even if you optimize your basic mechanics to make that moment of traction easier, there’s more to do. Your project needs to demonstrate its design patterns in a way that lets developers import the mental model of your expectations for a successful integration.

Mere docs of the existing functionality are not enough. Tutorials are better. But best is a collection of reference implementations—Glitch.com calls these “starters”—that let developers begin tinkering with a system that’s already complete and working, even if its scope is limited or simplistic.

Tinkering your way into a proof-of-concept is fairly cheap from there, and when the artifacts are good, pretty fun as well. This is how you can convince a developer that your solution has the tangible qualities they need to invest the considerable labor of learning and integrating your stuff into both their project and their longer term professional destiny.

Remember, too, that your competition isn’t just other developer tools. It’s developers avoiding your tool by solving the problem the hard way, because short-term, at least they are feeling traction on their progress.

So as we build developer tools, we have to think about:

Who will be made more successful when this ships?
How will we make the case to them they’ll win with us?
How will we demonstrate the superiority of our approach?
How can we limit the costs of initial exploration and proof of concept?
How can we maximize the impact of the investments developers make to integrate us into their workflows, products and future outcomes?

Hey, I never promised it would be easy. But it is worth it. There’s nothing more powerful than a legion of grateful developers whose rent checks clear because once upon a time, you made them successful.

Grasping the true scale of inequality

11 Nov 2022

There's a problem with understanding inequality at the modern scale.

Our minds struggle to make sense of a "billion" things. Much less tens or hundreds of billions of things. Our minds further struggle to compare how a billion of this might compare to a million of that.

As a result, the everyday person has no idea just how much more money the wealthy have:

The average American believes that the richest fifth own 59% of the wealth and that the bottom 40% own 9%. The reality is strikingly different. The top 20% of US households own more than 84% of the wealth, and the bottom 40% combine for a paltry 0.3%. The Walton family, for example, has more wealth than 42% of American families combined.

This viral video tries to visualize the drama:

There have been more attempts to make the differences tangible. For example, you've probably seen the viral TikTokker who quantified extreme wealth with grains of rice. For example, if a single grain is worth $100,000, Jeff Bezos has 58 pounds of the stuff.

I'd like to offer an alternative to these laudable approaches to solve this difficult problem.

Inequality is foremost a matter of time

Work is a trade: time and energy for some amount of money. We give up irreplaceable time in our lives to pursue the goals of someone else. In exchange, we get enough money to, we hope, pay for our basic necessities: food, shelter, clothing, medical care. If we're very lucky, we make more than we need, and can use the rest for comforts and saving.

In the United States, one year of work yields $70,784 in the median case.

Aside: median vs average

As a refresher, the median value in a set of numbers describes where the middle is. In other words, there are as many values that come before as after. Medians can be helpful in statistics around inequality because they prevent extreme values at either edge from disorting the picture.

Median annual income as the unit of time-for-work

So at the middle of the pack, $71k isn't quite prosperous, but it is enough to rent a one bedroom apartment in every US state.

Like grains of rice, we can use this number to slice up inequality into numerical scales we can actually understand. So for every $71k you have stored up, that's a year of self-determination or leisure time available to you. A year buffering you from poverty and desperation.

Time leverage by annual compensation

To start, let's look at annual compensation as it yields a unit of median US income. In other words, how much does a year of work buy you in terms of the power not to work if you don't want to?

Elon Musk is doing pretty well: in just a year he was paid 142 millennia of median income. In other words, Elon made enough leisure money for 23x the duration of all human civilization.

Musk is an outlier, certainly, but you can find plenty of other dramatic examples.

Tim Cook made enough in a year for 12 millennia of leisure, and Sundar Pichai got enough for 40. Dave Clark has 800 years of leisure time at his disposal, while Satya Nadella gets 700.

Parasitizing the American healthcare system isn't a bad gig, either. CVS CEO Karen Lynch has almost three centuries of leisure coming her way, as does UnitedHealth CEO Andrew Witty. This makes sense: people will do anything to keep themselves and their loved ones alive. It's a profitable protection racket.

Everyday workers, meanwhile, earn less. Walmart, Amazon and McDonald's workers' median incomes are less than half of the national median, while Apple's is 80%.

As Starbucks workers fight, often successfully, to build a union, it's worth noting that the CEO there gets almost three centuries of leisure, while the typical barista scrapes by with only half the national median income.

Dave Clark gets eight centuries of leisure in a year while half his workers don't even get a single year. In fact, it's worse than that, because working in an Amazon warehouse can cost future earnings, due to injuries and fatalities.

I'm comfortable with the argument that being an executive of a public company takes certain specialized skills not everyone has, and therefore is due certain additional rewards. But centuries of leisure potential every year? While the typical worker doesn't even hit the median income, much less stack up any extra? That's taking so much and leaving so little.

Time leverage by wealth

But annual compensation inequality is nothing compared to wealth inequality.

During the 2016 campaign, Trump argued that he received a "small loan" of $1 million from his father, and that he'd worked hard for his wealth. Let's take the claim at face value, ignoring all his other generational advantages. That's 65 years of income, or more than an entire lifetime. Imagine what you could build with an entire lifetime of income loaned to you at the beginning of your career.

Meanwhile, back to Musk. He has 2.7 million years of buffer time stored up. That's more time than has passed since the first humans evolved. Bezos, Buffet and Gates each have around 1.5 million years.

If it sometimes seems as though Nancy Pelosi—disdainful as she is of progressive policy goals—is out of touch with the typical American, it's worth noting she has as much as 15 centuries of cash. That's enough money to last from the fall of Rome until today. McConnell isn't doing too badly, either, with 500 years stashed away. Joe Biden could retire, meanwhile, for over a century on his current haul.

AOC, it should be noted, may still be in the red thanks to student loans. She's much closer, therefore, to the typical American, which has just two years of cash buffer. It's harder for those who didn't attend high school: they have almost no buffer, at median net worth of $20k. Having a college degree, meanwhile, brings the median to four years of buffer.

This doesn't even touch the mechanics of financialization, like stock buybacks. Apple has transferred 132,000 lifetimes of wealth in this form, or 5.9 million years of the median US income.

Inhuman leverage

So we have some people, in American society, with almost no buffer at all.

Meanwhile, some among us have so much excess power in the form of time that they'll die centuries before they could come anywhere close to using all of it. This is disorting our world in dramatic ways, between the chaos of the Twitter acquisition to the fallout of Citizens United, to the ongoing consolidation of essential services.

As a whole, we're able to produce so much wealth. Does it really make sense for the most powerful among us to be so gluttinous with the rewards? Does Elon need 61,000 lifetimes of wealth? Do Bezos, Buffet and Gates need 30,000 lifetimes of wealth?

What would happen if they shared the pie a little more? How would the resulting tax revenues improve our communities through infrastructure and education spending? How would everyday lives improve with less stress, more leisure time, more time with our loved ones?

It's helpful for the wealthy that the numbers are so incomprehensibly big it takes a whole spreadsheet just to begin the conversation.

When your salary requires you not understand the labor movement

18 Oct 2022

I’ve been reading Daring Fireball for something like 18 years now. I appreciate John Gruber’s insights on Apple, and find him more right than not in analyzing their products, strategy and motivations. Hell, I survived a layoff in 2020 by buying an ad on his site.

But I’ve been scratching my head at this recent remark about union drives at Apple’s retail operation:

This public enthusiasm for labor unions is manifesting in high-profile unionization drives at big companies like Starbucks, Amazon, and now Apple.

This is a strange logical construction to me, but it mirrors a larger challenge I find among pundits in understanding the current moment and movement in labor.

In one of my favorite quotes of all time, noted 20th century troublemaker Upton Sinclair wrote “It is difficult to get a man to understand something, when his salary depends upon his not understanding it!”

The most insightful people in the game are struggling to make sense of a resurgent labor movement. But it’s not that hard to follow—if your incentives aren’t too bound up in the interests of the people who already have a lot of money.

Trouble is, that’s a hard line to walk while getting paid to write. I'm sympathetic—and unaffected. Maybe I can help.

Unions aren’t forming because they’re popular; they’re popular because they’ve become urgently needed and they’re forming for the same reason

In most people’s interactions with a workplace, the company takes too much and gives too little. The only recourse for labor is to form structures of counter-power to try and balance the equation.

You can stop reading there. All I’m going to do next is prove the point several ways, but if you came here to understand why unions are both forming and popular, you’re good to go.

CEOs, as agents of Wall Street and other financial interests, are paid hundreds of times what their workers make every year. In Apple’s case, Tim Cook took home $100m in 2021 alone. The typical Apple Store employee, making $22 an hour, would need to work 2,367 years to match Tim’s compensation.

This isn’t unusual to Apple, though. CEO pay is at an all-time high, but that’s not even the worst part. When workers create profits for corporations, what doesn’t go to the CEO is too often sucked up by shareholders in the form of stock buybacks.

Supporters of the status quo will argue that guys like Tim Cook create outsized value for companies, and deserve outsized compensation as a result. I can accept that Cook is a uniquely talented person with unique insights. Gil Amelio, Michael Spindler and John Sculley are proof enough that not everyone is suited to run Apple.

Nevertheless, I struggle with the idea that Cook deserves that much more of the pie than the people who make it possible for him to move the vast quantities of hardware and services that allow Apple to post its billions in quarterly profits.

This isn’t an argument in the abstract, either. It’s becoming harder and harder to afford the basics of life—housing, food, transportation, childcare—in the United States, precisely because of this inequality. For example:

Rents have increased an unprecedented 10%
Consumer price index on food has risen 11.2%
Childcare costs have surged 41%

The people with money are living the high life while wage workers are struggling to get by. But this is about more than money. Employees of large corporations are separated from decision makers by enormous gulfs of reporting structure and policy, with limited say in their day-to-day work.

Apple’s workers don’t just want more money, they want things like better scheduling and career advancement. The timing of when you work is everything: it impacts your ability to rest, to be with friends and loved ones, to meet educational goals, and otherwise determine the course of your life.

Scheduling in a recurring theme in many recent retail labor disputes, as in the case of Starbucks.

Amazon presents perhaps the most extreme example of how precarious today’s workers are. Six warehouse workers died when a tornado struck a distribution center in Illinois last year. Desperate drivers with no slack in their schedules have to piss in a bottle to meet their delivery quotas, as the company admitted to lawmakers. The company’s idea of worker well being is, in a bit that would go too far even for Severance, a phone booth-sized cubicle where workers can watch mindfulness propaganda.

Self-determination is an issue for wage earners across many sectors. The US sits on a knife’s edge as rail workers—over-scheduled and fighting for the basic right to do things like visit the doctor once in awhile—contemplate a nationwide strike that would grind logistics infrastructure to a halt. Those guys, at least, have a union.

To recap, workers are struggling with:

The basics of reliable scheduling and paid time off
Soaring costs of the essentials
Their ability to advance their careers
All the surplus value they create going to CEOs and Wall Street

In an economy that has produced enormous gains over the last decade, all of the fruits are going to the richest people in the system. After a global pandemic, in which frontline workers kept entire global economic order afloat, the rich are richer than ever, while workers are scrambling to pay the bills.

That’s why unions are popular. That’s why unions are happening.

There’s just no other recourse for such a wide-ranging, unfair, structurally entrenched bargain.

labor