Principle Stash

Where Does Software Engineering Go From Here? (Part 1)

Mon, 07 Sep 2026 00:00:00 GMT

I’ve been sitting on this one long enough that a few of these predictions will read as obvious to anyone who lives on Twitter. I don’t think that’s because I’m slow, but because the half-life of a paradigm in this industry has collapsed from years to months to weeks. A take goes from contrarian to consensus to cliché quite literally before you’ve finished writing it down.

I’m putting this in writing anyway, mostly because of the massive gap between the ecosystem’s loudest factions: the “AI is slop” camp, still living in denial and typing at 60 wpm, and the “AGI soon” camp, rapidly tipping into genuine AI psychosis. As usual, both extremes are wrong, and the truth lies somewhere in the messy middle.

1. LLMs are not intelligent, and never will be

Let’s get this out of the way first. If you disagree with this premise, nothing that follows will land.

People define intelligence a dozen ways, but I always return to Naval’s definition: the only real test of intelligence is whether you get what you want out of life (while knowing what’s worth wanting is wisdom). The machine doesn’t want anything. There is no “life” for it to get anything out of, and a thing without desire cannot have intent. Intention is the root of automation and the exclusive domain of beings with a soul, the one thing automation cannot supply. Even when an LLM looks like it’s acting with intent, that intent is borrowed; it’s the echo of the spark you supplied the moment you typed the prompt.

Over a millennium ago, Ibn Sina made this observation. Fast forward to 2026, and Pope Leo XIV reached for the same distinction in Magnifica Humanitas, his encyclical on AI, noting that these machines “do not undergo experiences, do not possess a body, do not mature through relationships.” They don’t want. They don’t feel. You don’t empathize with your power drill; you don’t ask if it’s having a good day, or handle it more gently so its feelings don’t get hurt.

Yes, a computer does more arithmetic in a second than I’ll do in a lifetime. An LLM extracts, summarizes, and remixes the written output of our species better than any single one of us. It lives in the space between a search engine and a synthesizer, approximating the internet as it stood on some past Tuesday. Inside that distribution, it performs miracles. Outside the training data, past the edge of context, it devolves into confident nonsense. That’s not to say they’re not incredibly effective at probing the limits of our knowledge — in May 2026, an 80-year-old Erdős conjecture fell to a single prompt — but those wins belong to a particular class of problem: the answer is checkable, the search space is bounded, and being right is unambiguous. The world rarely hands you problems shaped like that.

The Unassailable Consensus

The same math that lets them do all this has a gravity that pulls relentlessly toward the median. To defend the “AI is just autocomplete” crowd, that is truly all this is: predicting the likeliest next token. And the likeliest answer is, by definition, the most typical one — the literal consensus of its training data. Then alignment training sands that down further, rewarding whatever a rater was most likely to nod at.

So treat it for what it is: an amplification tool for what already exists. Call it a “clanker” if that helps keep your distance; better an unfeeling machine catches a stray than a fellow human. But stop saying model X is smarter than model Y when you just mean it’s more effective. Attribute intelligence to a thing that has none, and the only thing you’ve shown is where it’s lacking.

2. English is a leaky abstraction; it will never fully replace code

Every rung we climbed up the abstraction ladder so far has been deterministic. Punch cards gave way to assembly, assembly to compiled languages like C, C to the languages you actually use, and at each step the higher layer compiled down to the lower one the same way every time. Feed a compiler the same input twice and you get the same machine code twice; obviously not the case with LLMs.

Joel Spolsky called this the law of leaky abstractions: the tax on every abstraction is that you still have to understand the layer below it for the day it inevitably leaks. You save time executing, not learning. Using English to drive software is that same bargain, cranked to the absolute maximum.

Ask your AI assistant to add real-time features to your app. How is it doing that? Is it polling, long-polling, SSE, or WebSockets? What’s the consistency model when two people edit at once? Last-write-wins, CRDTs, or OT? How is it handling reconnection, offline queueing, or presence? People write dissertations on just one of these topics. The model picks a lane and buries the rest, and you won’t find out which lane it chose until production breaks.

Say you catch the bug born from its wrong guess, and you write the assistant a prompt detailing exactly what you meant. If you do that well, the prompt gets long. It gets precise. It fills up with the constraints and edge cases you left implicit the first time around.

Eventually, it gets so long and precise that it stops being a prompt altogether. We already have a word for a document that perfectly pins down every architectural decision and edge case: it’s called code.

English as an Abstraction

3. Code is a liability; always has been, always will be

The very first thing that came to mind when chat interfaces started getting shoehorned into apps with perfectly good, decades-honed UX was: oh boy, a new interface to maintain. As if mobile, web, and desktop weren’t enough. A text box isn’t going to magically deprecate the GUI; it’s just another layer of the stack to test, design for, and inevitably patch.

Jaded? Maybe. But every line of code you ship is a building you’ve just committed to maintaining, perhaps indefinitely. Maintenance already eats 60 to 80 percent of a system’s total lifecycle cost, a metric that has held steady across decades of studies. The bulk of that cost isn’t even fixing bugs; it’s managing the sheer weight of what already works. Every feature is a new surface area that every future change must route around. If agents multiply the rate of construction, they multiply the upkeep right along with it. Generating code faster just means generating liability faster.

Side-project (AI Edition)

You could argue the cost of generating code is trending to zero, so volume no longer matters. But trending to zero isn’t zero. You are still paying the maintenance tax, whether it’s in LLM compute tokens or consultants you eventually have to hire to fix your slop. (If that’s you, shoot me a message, I know a guy.)

More importantly, typing the actual code was never where the time or money went; it occupied maybe ten to twenty percent of the job. The rest is figuring out what to build, safely wedging it into a legacy codebase, testing it, and keeping it running until the end of time. We haven’t automated software engineering; we’ve just sped up the typing.

The Paradoxes of Software Engineering

4. Context and tooling is what makes an agent useful, and control will be open source’s advantage

An agent’s usefulness scales with how much of your world it can safely see and touch. Mine became sharply more useful the moment I connected my calendar, email, projects, and apps, and better again when I let it onto my laptop to actually click around and do things.

You could argue that this data should never have been siloed in the first place. But those walls were the rational product of the previous software era, when companies bought the best tool for each job and every tool had an incentive to hoard its data like the business depended on it, because it usually did.

That incentive is now inverting. If a customer’s agent cannot reach the customer’s own data inside your platform, they will leave for one where it can. The wall that once kept them trapped is becoming the reason they churn.

So the frontier moves up a layer.

“Open” used to mean data sovereignty: prying your information out of someone else’s walled garden. Once access is table stakes, it comes to mean control: swapping memory stores (temporal knowledge graphs, LSM trees on object storage), swapping tools, or JIT compiling the harness itself, and making sure your Millennium Prize research doesn’t show up in someone else’s LLM.

But more is not always better. Shove too much information into an LLM’s context window and it struggles to separate signal from noise, much as a human would. It latches onto the stale note, the obsolete calendar invite, or the half-dead pattern buried somewhere in the repository.

The same tradeoff applies to tools. Give an agent tools that are too primitive and it has to reconstruct higher-level behavior itself, fumbling through ten brittle steps for something that should have been one reliable operation. Give it too many tools, or tools with sprawling and ambiguous interfaces, and it struggles to choose between them, passes the wrong arguments, or chains them together incorrectly.

The quality of the tools matters just as much as the intelligence of the model using them, and a mediocre model in a good harness can beat a frontier model dropped into a bad one. For example, a better search engine like Exa gives the agent cleaner raw material. A faster browser like Lightpanda, which throws out rendering entirely and does the same crawl in a fraction of the time Chrome needs, lets it explore more paths before the task times out. A narrow, well-designed API can outperform a general computer-use interface because it removes entire categories of possible mistakes.

Goldilocks Context

If this is starting to sound a lot like the same stuff software engineering has been doing for decades, that’s because it is. The durable advantage will belong to whoever gives you the most control over what the agent sees, what it can do, and how those capabilities are assembled. That is where open source has always been strongest.

5. Specialized models and harnesses will dominate, coordinated by a central orchestrator

For now it’s useful to talk about the model and the harness as separate things. The model supplies the raw capability; the harness decides what it sees, what it can touch, and how its work is checked. That split won’t last. Specialized models and the environments built around them will be trained, evaluated, and sold as one system, and asking where the model ends and the agent begins will be like asking where a database ends and its query planner starts. Technically answerable, but mostly irrelevant to the person using it.

These specialized systems will win for three reasons.

First, the money. Today’s token spend is subsidized by VCs and trillion-dollar balance sheets terrified of missing the next platform shift, and that era is closing fast. GitHub moved every Copilot plan to usage-based billing, heavy agentic users are reporting bills 10 to 50 times higher, Microsoft pulled its own engineers off Claude Code after per-seat costs hit $2,000 a month, and Uber burned its entire annual AI budget in four months.

The days of cheap Uber…I mean, cheap token spending are over.

Meanwhile, capable open-weight models keep arriving with absurd token efficiency, and using the largest model available for every task is like hiring a Nobel laureate to alphabetize your filing cabinet. Most work doesn’t need everything the model knows, just a narrow slice of it, delivered cheaply and reliably.

Effective Utilization of LLMs

Second, the math bends toward cheap. Approximating data you already have is exactly the kind of problem that gets relentlessly cheaper: distillation squeezes a big model’s behavior into a small one, and inference at a fixed level of quality keeps collapsing in cost. What doesn’t get cheaper is new knowledge. A human still has to discover it, prove it, and write it down before a training run can absorb it.

Third, and thorniest, proprietary data is the moat. The open web that trained this generation of models is filling with their own output, and training on that sludge gives you a photocopy of a photocopy, blurrier with every pass. The way out is data nobody else has: your codebase, your incident history, your customers’ actual support tickets, the fixes your best engineers make on instinct, the tribal knowledge nobody ever wrote down clearly enough to scrape. As the public corpus rots, the premium on a clean private well only rises.

Tomasz Tunguz compared where this leads to databases: one category that fragmented into relational, document, graph, time-series, and vector stores, each shaped around a workload. Models will fragment the same way.

Skill files are a crude preview. They package domain knowledge into something the model loads at runtime, which means paying to reread and reinterpret the same material on every task. A fine-tuned model is that skill file compiled into the weights; Callstack’s Apex, tuned for React Native, is an early example. The likely destination is a mix: stable knowledge compiled into the model, changing context left in the harness, and lightweight LoRA adapters in between, so a skill becomes a module you load, combine, or drop without retraining anything.

How Many Times Do I Have to Tell You?

You can already see early signs of this: Skill-to-LoRA replaces runtime skill text with skill-specific adapters, and Microsoft’s MoLoRA loads several at once and learns which to activate for each token. A Rails adapter. A Terraform adapter. A company adapter trained on your code, incidents, and conventions.

The harness is going the same way. Today it’s hand-built and fixed, a guess about the average task made before any task has arrived. JIT-Agent is the first serious attempt to kill that guess: a model trained to emit harnesses, composing memory, planning, and tool modules per task and repairing them when execution goes sideways. In a generated harness, DeepSeek-V4-Flash beats GPT-5.6 on DeepSearchQA by 9.1 points, GLM-5.2 gains up to 20.2, and the generated harnesses hold their own against Claude Code and OpenCode. The cheap model in the bespoke harness beat the expensive model in the generic one.

The famous counterargument is Rich Sutton’s “bitter lesson”: general methods that scale with compute have repeatedly beaten systems where humans tried to hard-code domain expertise. But the lesson assumes compute keeps getting cheaper, data stays abundant, and you have time for generality to win. Put a meter on every token and a deadline on every task, and the calculus flips. A specialized model isn’t a return to hand-written rules, either; it’s the same learning machinery aimed at a narrower distribution. The bitter lesson still governs the base layer. The products built on top of it are another matter.

Which finally brings us to the orchestrator.

The orchestrator doesn’t need to be the best model at anything. It needs to understand the task, break it apart, route each piece to the right specialist, and check that the pieces fit back together: architecture questions to a frontier model, implementation to a framework-specific one, security review to a third, routine repository chores to something small enough to run locally. That’s how the economics resolve: frontier capability only where ambiguity demands it.

The models already contain this architecture in miniature. A Mixture-of-Experts model doesn’t fire every parameter for every input; a router picks a few specialized experts per token, so the model carries enormous capacity without paying for all of it at once. The agent stack will do the same one level up, routing tasks between models, adapters, tools, and harnesses instead of tokens between experts. Whether the answer came from the weights, an adapter, a tool, or the context, the user will neither know nor care.

From the outside, it will simply feel like competence.

5b. Open source gets a new business model

This new paradigm also creates a potential way to monetize open source.

In January 2026, Tailwind laid off three of its four engineers while the framework was more popular than ever, powering a huge share of the sites produced by the vibe-coding boom.

Its business depended partly on documentation traffic. Developers searched for how to center a div, landed on the docs, and some fraction bought Tailwind UI or another paid product. AI assistants now answer the question directly, so the framework can remain widely used without sending attention or revenue back to the people who built it.

Open source has always struggled with this mismatch. A project can sit underneath half the internet while capturing almost none of the value it creates. AI makes the mismatch worse by removing the few remaining points where attention flowed back to maintainers.

Dependency: AI Edition

But the same shift that destroys the documentation funnel may create its replacement.

Depending on where the tooling lands, that replacement could be a model fine-tuned on the framework itself, a set of LoRA adapters, or a maintained skill and plugin package that drops the project’s conventions directly into whatever agent you use. The implementation may vary, but the business model is the same: the people who built the framework, and understand it better than any general model scraping its documentation, sell the canonical way to wield it.

Instead of selling access to an answer one consulting hour at a time, maintainers can encode their expertise and judgment into something an agent applies repeatedly. Keep the package current as the framework evolves, charge for access or updates, and that knowledge becomes a scalable product rather than a service constrained by the number of hours in a day.

You could even move advertising into the tooling itself. That may sound like blasphemy, but a single sponsored status line in your terminal is a small price to pay if it keeps the open source project underneath your entire stack alive. I’m exploring a version of this model now. If you maintain an open source project and the idea sounds interesting, reach out.

6. The role of a Sofware Engineer will keep changing, but demand will continue to outpace supply

Repeat after me: LLMs are not intelligent. They cannot set intentions, decide what is worth building, cook up truly novel ideas, or take accountability for their decisions. There will always be a need for human operators to supply direction.

And for those humans to participate meaningfully in the software development lifecycle, they still need to understand the software being built. You cannot specify what correct means, judge whether an abstraction fits, or recognize when an agent has confidently wandered into nonsense without understanding the system beneath it. The code may increasingly be written by machines, but responsibility for it cannot be.

It may not feel like demand is outpacing supply because the software job market is brutal right now. But we are still coming down from two overlapping distortions: Covid forcing years of digital transformation into a few quarters, and zero-interest-rate overhiring that treated engineers like a speculative asset. The correction was always going to be painful, with or without AI.

You could argue that company leaders are laying off engineers because they see where the market is headed. I personally don’t buy it. Record profits, repeated re-hiring after cuts, and a general lack of understanding of how software actually gets built, all suggest something simpler: AI became a convenient story for cost cutting that was already going to happen.

If you are skeptical, ask yourself a simpler question: do you, or any engineers you know, work less since you started using AI?

Probably not.

The classic explanation is Jevons paradox. When a resource becomes cheaper or more efficient to use, consumption often rises enough to offset the savings. More efficient engines did not reduce coal consumption. They made coal economical for more uses.

Software follows the same pattern. When building a feature becomes ten times cheaper, companies do not build the same number of features with one tenth of the staff. They attempt ten times as many features, automate work that was never worth automating before, maintain more internal tools, personalize more workflows, and expand the frontier of what counts as economically viable software.

The Back-Backlog

On a micro level, practices that once felt too expensive or tedious to apply consistently will make a comeback. Test-driven development, domain-driven design, formal specifications, property-based testing, exhaustive observability, threat modeling, and all the other acronyms we invented and then selectively ignored because deadlines were real.

In the past, a startup could get away with poor engineering practices because it wrote code slowly enough for humans to keep some rough picture of the system in their heads, so the mess accumulated gradually. When agents can generate a month of implementation in an afternoon, the same lack of discipline becomes catastrophic. You can create a codebase faster than anyone can understand it.

Best practices therefore move earlier in the development lifecycle. Tests are written before implementation not because someone finally won an argument about methodology, but because agents need a machine-readable definition of correctness, an eval if you will. Architecture has to be made explicit because the system can no longer rely on conventions being absorbed slowly through osmosis. Documentation, invariants, schemas, and constraints stop being chores performed after the work and become the inputs required to do the work safely.

We also lose most of our excuses for tolerating technical debt beyond the phase where we are still discovering the right abstraction. If a language is unsuitable for the workload, an agent can help migrate it. If a legacy library is holding the system back, an agent can replace it. If a critical path needs a more performant component, an agent can write and benchmark several versions before lunch.

That does not mean every migration becomes wise, as you still need to decide whether the benefit justifies the risk, define the migration path, preserve behavior, and know which ugly parts of the old system were accidental and which were load-bearing. But at least you can automate the boring parts.

This is also the engineer’s side of the bargain. Every hour not spent on grunt work and boilerplate is an hour spent understanding the technology one layer deeper, and that understanding is precisely what extends the agent: sharper specs, better tools, evals that mean something, more work that can be offloaded safely. Less grunt work buys more understanding, and more understanding buys more leverage. It is a virtuous cycle of learning, but only for those who spend the time they get back on learning rather than on more prompting.

Time Well Spent?

More broadly, the job moves up the stack from writing code to specifying systems. Engineers will define invariants, design verification, decide what correct means, and audit fleets of agents the way a line manager audits stations on a production floor. “Everyone is an architect now” gets thrown around a lot, for good reason, though the defining job of an architect was never drawing diagrams. It was managing risk and preserving optionality: making the few decisions that are expensive to reverse slowly and deliberately, and structuring everything else so it stays cheap to change your mind later. That skill becomes essential the moment agents can pour concrete at ten times the speed, because when implementation is nearly free, the cost of a system is how difficult it is to change.

This will require a psychological reorientation. Engineers have started describing themselves as the bottleneck, and if you accept that framing, the logical endpoint is getting rid of yourself.

The better frame is rate limiter, quality gate, and direction setter. A bottleneck is a flaw in the line. A rate limiter is part of its design. Engineers who see review as an obstacle start cutting corners. Reviews become shallow, agents merge more code than anyone understands, and soon nobody is watching the system evolve.

The scarce skill stops being syntax and becomes judgment, which mostly means taste: knowing when a feature is bloated, when an abstraction is premature, when a release should slow down, and when not to ship at all. Agents raise the stakes on both sides of that judgment. Every option you preserve gives the fleet more room to operate. Every irreversible decision made carelessly becomes a foundation a tireless machine can build a thousand things on before anyone notices.

7. Small companies will operate like Big Tech, and Big Tech like small companies

Trying to be Google when you are not Google has always been a good way to drown in processes you did not earn. Google could throw an org at a problem. You had three engineers and a Notion page.

Agents invert that equation.

A five-person company will be able to operate capabilities that once required entire departments: platform engineering, QA, security, data infrastructure, internal tooling, compliance, and customer support. Small teams will gain much of the operational surface area of large organizations without inheriting their headcount.

They will maintain more products, support larger customers, run more experiments, and build internal systems that would previously have been dismissed as too expensive for a company their size. Not because the work disappears, but because a small number of people can direct far more execution.

The reverse happens inside Big Tech.

Large companies were built around the assumption that execution was scarce enough to justify enormous coordination costs. Every new problem could support another team, and every new team brought managers, planning cycles, roadmaps, dependencies, status meetings, and approval chains.

Agents make execution cheap. Coordination remains stubbornly human.

A company with fifty teams does not automatically produce fifty times as much. It may produce fifty competing priorities, duplicated systems, and six months of cross-org alignment before anyone can change a button. The faster agents make each team, the more obvious the cost of coordinating all of them becomes.

The giants will discover that much of what looked like scale was coordination overhead made tolerable by enormous margins. As smaller competitors begin shipping with a fraction of the people, large companies will be forced to collapse management layers, combine functions, and give smaller groups ownership over much larger surfaces.

The basic organizational unit starts to resemble a startup: a few people with broad responsibility, supported by agents instead of a pyramid of narrowly scoped specialists.

This will also change what management means. A manager whose primary job was moving information between layers becomes difficult to justify when agents can summarize, plan, report, and coordinate routine work. The valuable manager will look more like the engineer described in the previous section: someone who sets direction, resolves ambiguity, allocates scarce attention, and takes responsibility for decisions that are expensive to reverse.

Small companies will gain the leverage of Big Tech. Big Tech will be forced to rediscover the speed and ownership of small companies.

The companies in the most danger are those caught in the middle: carrying the coordination costs of a large organization without the capital, distribution, or margins that once made those costs survivable.

8. The current AI leaders will be martyred in the process

Anthropic in particular has been on an absolute tear, assembling what may be the most star-studded roster in the history of tech and shipping at a pace that makes most companies look asleep.

But no matter how cracked the roster, mere men cannot conquer market forces.

This is not an argument that Anthropic is badly run. On the contrary, it may be the best-run frontier lab, which is why it makes the best example.

Anthropic was built to win the frontier-model era: train the most capable general model, own the infrastructure to serve it, charge by the token. In May 2026 it raised another $65 billion at a $965 billion valuation to keep doing exactly that. Every part of the machine assumes frontier intelligence stays scarce.

Everything above runs against that assumption: cheaper models, better open weights, knowledge compiled into adapters, orchestrators that reserve frontier capability for the few tasks that need it. The model becomes a component, and the value moves to the data, the workflow, and whatever decides which model to call. Customers don’t want to maximize their frontier-token bill. They want to minimize it without degrading the result.

One Last Task

Anthropic clearly understands this. Claude Code is a harness, and it may be the clearest evidence anywhere that the product is worth more than the model beneath it. So it’s moving up the stack: Claude Cowork, Claude Design, and vertical workflows like legal. OpenAI is making the same move with ChatGPT for Financial Services.

Neither is likely to win the war against the open model. A developer in Montreal making CAD 60,000 can take an open model at a hundredth of the cost, tune it around one narrow workflow, wrap it in tools built for that job, and match the general product. In a narrow enough domain, beat it. Anthropic attacks the same market with a team that costs millions a year, the overhead of a frontier lab, and infrastructure expensive enough to justify that lab’s existence.

The solo developer, SMB, and even Enterprise customer uses whichever model is cheapest and switches whenever the market moves. Anthropic can’t, because its products exist to protect the model business, and a product pivots more easily than a balance sheet. Cowork, Design, Legal, and whatever comes next don’t just need to be good businesses. They need to carry the frontier lab beneath them. The startup is free to optimize for the customer. Anthropic must also optimize for Anthropic.

Now multiply that by every vertical it enters, each defended by people who have spent their careers on one design tool or one legal workflow. Talent, distribution, and brand buy an excellent horizontal interface, but good packaging only makes people tolerate subpar software for so long.

History rhymes; Netscape taught the world to browse, then lost the market when Microsoft bundled Internet Explorer into Windows. AltaVista made the web searchable, then lost it to Google. Sun sold the servers that powered the dot-com boom, then ended up inside Oracle. None of them were staffed by idiots. They were optimized around the bottleneck of one era just as the value migrated to the next.

The model labs are building the same bridge: historic sums spent industrializing the production of intelligence at the very moment intelligence becomes abundant, modular, and interchangeable. Their success accelerates the commoditization that threatens them.

That said, Martyrdom doesn’t have to mean bankruptcy. Anthropic could stay enormous as infrastructure for someone else’s products. But relative to the value creation that’s going to happen, their slice of the pie will look increasingly small.

At least the early employees will be able to dump their bags on to the general public post-IPO before reality catches up, and with the money and talent coming out of the comapny’s carcass, the Anthropic Mafia is going to be unbelievable.

9. Power law amplification in the broader market

Want to see what democratized creation produces? Look at social media. Everyone has a camera and everyone posts. Why is everyone not an influencer?

The mistake in “everyone will be a software engineer” is assuming people want to be. This may come as news to some of you, but we’re very weird people. Like, really weird. The average person does not, believe it or not, enjoy spending a Sunday morning tinkering with a new JavaScript framework.

The same way most people do not want to do their taxes, most people do not enjoy the act of developing software. Sure, the initial rush of building that got most of us into software engineering is now more accessible than ever, but the difference between production software and a nifty project has never been larger, and most people would rather outsource it to someone who is good at it and not think about it at all.

App Store submissions jumped 84 percent in a quarter, enough to buckle Apple’s review pipeline, and yet attention and revenue remain concentrated among the few who treat the work as a craft. The tools spread to everyone. The rewards do not.

The explosion in quantity also makes room for an explosion in quality because the marginal cost of polish is collapsing. Case in point, this post has illustrations I would never have made by hand. The people who care about the work will not just outlast the flood. They will use the same tools to raise the ceiling.

That probably means an even wider divergence in pay. When one exceptional engineer can direct the output of what used to be a team, their economic value starts looking less like a normal salary band and more like professional sports. Big Tech already paid its best engineers multiples of everyone else. Agents give the best operators even more leverage.

The other place where the social media analogy holds is distribution. As building gets cheaper, getting anyone to care gets harder. Having an audience, a recognizable voice, and an active presence online becomes part of the product whether you like it or not.

Engagement Farming

Software is becoming abundant, while our attention is scarcer than ever, and it’s practically an all-out brawl over it.

10. We’re still very early

As an LLM would say: this isn’t a cop-out, this is an acknowledgment.

I’m betting both time and money on these predictions, but the truth is that we really don’t know what the true second, third, and fourth order effects of releasing these amplification machines unto the world are going to be, nor what breakthroughs in their core tech await us, nor what applications that previously seemed implausible have suddenly become possible.

The software world is unique in the sense that it’s already 100% digital. But even then, LLMs have cranked up exactly one station, code generation, while every adjacent station still runs at the old speed. In legal, Legora and Harvey have made research and drafting dramatically faster, while discovery, negotiation, filing, and the court’s own calendar move at precisely the speed they did a decade ago. Speeding up one station on a slow line doesn’t give you a faster line, it gives you a pileup, and I suspect the coming decades will be about finding where those pileups accrue and equipping workers with the skills to run the stations that are still stubbornly human.

There’s also still a massive skill issue as we learn to harmonize, as AI usage reports read like early computing reports: lots of perceived productivity, not much real productivity. Experienced open source developers using AI tools turn out to be 19% slower while believing they’re 20% faster. We’ve run this experiment before. Businesses spent the 70s and 80s pouring money into computers while the gains stayed invisible in the productivity statistics. They only showed up in the late 90s, once firms restructured work around the machines instead of bolting them onto old processes. The companies that did the restructuring saw gains several times larger than the ones that just bought the technology.

But what’s clear today are the limits of our current processes, since we’ve made writing code fast, and left everything around it slow.

Which is where the next part picks up: what a tool built for that problem — rather than for faster typing — actually has to look like.

On the Separation of Money and State

Mon, 15 Jul 2024 00:00:00 GMT

“It is well enough that people of the nation do not understand our banking and monetary system, for if they did, I believe there would be a revolution before tomorrow morning.”

The above quote is often attributed to Henry Ford, though there is no concrete evidence that he actually said it. The sentiment nonetheless holds true; the inner workings of our financial system are so opaque and convoluted that even the most financially literate among us struggle to understand it, and those that do are often too busy profiting from it to speak out against it.

Deep down, you already felt that something was off.

How is it that despite decades of concentrating in classrooms, crazy hours climbing the corporate ladder, continuous side hustles, careful investment, cutting costs and clipping coupons, you’re somehow still teetering on the brink of financial collapse, just one bad day away from bankruptcy?

Yet in stark contrast, the world’s richest continue to grow their wealth exponentially, and governments can somehow always come up with the budget needed to pursue their latest moral failing. Is it crazy to think that there’s some connection there? That there’s something wrong with the system?

I’m here to tell you that you’re not crazy, and that the system is indeed fundamentally flawed.

Take the story of imperial Japan in 1942, when the Japanese government issued and forced the use of “invasion money” in its colonies, both as a tool of unification and wealth extraction. Backed by nothing but their colonizer’s questionable guarantee, these “Mickey Mouse Dollars” - as they became known at the time - were introduced at an arbitrary valuation relative to the colonized nation’s original money, then continuously devalued to fund the Japanese conquest. To make matters worse, every time Japan lost a battle, their value would plummet, until they finally became worthless.

Piles of Japanese-issued Pesos prior to being destroyed, Manila, c 1945, Pfc Glenn Eve.

This is more or less how our current money (aka fiat money) functions today.

If you’ve spent any time around software, you already know this pattern by another name.

Cory Doctorow coined “enshittification” to describe the way platforms die. A service launches and is genuinely good to its users, because it needs them. Once enough of them are locked in and leaving would cost too much, it starts treating them worse in order to court the businesses that actually pay the bills. Once those businesses are locked in too, it claws the value back for itself. Then the whole thing collapses under the weight of what it has extracted. Search results become ads. Feeds stop showing you the things you subscribed to. The app that used to work now wants a subscription to do what it did for free last year.

The part people tend to miss is that none of it works without the lock-in. A platform can only degrade as far as your ability to leave allows. Enshittification isn’t a story about greed - greed is a constant - it’s a story about what happens to greed once exit becomes impossible.

Our money is beyond enshittified.

Paper notes started out genuinely better than the gold they represented; you could settle a transaction across an ocean in the time it took a telegram to arrive, and nobody had to ferry bullion through pirate waters to do it. That was a real improvement, and people adopted it for real reasons. Then came the lock-in: legal tender laws that compel you to accept the currency, capital controls that stop you from moving out of it, and in the United States an executive order that made owning the alternative a crime. Only once the exits were sealed did the extraction begin, and it hasn’t stopped since.

What follows is that arc, at the scale of a civilization and across a century. The good years are the classical gold standard. The lock-in is 1914, then 1933, then 1971. The extraction from users is inflation, which quietly moves the value of your work to whoever happens to stand nearest the printer. The extraction from business customers is the IMF, which does to entire countries what your bank does to you, with better branding. And the terminal stage is the 44 hyperinflations that have happened since the peg broke, each one a currency that finally ran out of people to extract from.

The difference is that you can delete an app. You cannot delete the dollar. Every platform that has ever enshittified was, in the end, opt-in, which is why the pattern is a nuisance in software and a catastrophe in money.

To understand how we got here, it’s necessary to understand the history and evolution of money, and answer the deceivingly simple question of “what is money?”

💰 A Condensed History of Money

Money is a medium of exchange that serves as a unit of account and a store of value. This is perhaps the most widely accepted and succinct definition of what money is.

But it took ages for humans to arrive at this definition, and it all began with the humble mental ledger. Abel did a favor for Cain. Cain made a mental note to pay him back later with something equivalent (though probably not how he would have liked). This meant that these favors were fungible; interchangeable and equal in value to other units of the same denomination.

As more people began to roam the Earth, it became difficult to track who owes what, and so a physical record of these exchanges emerged out of necessity. The exact mechanism did not matter; maybe it was lines drawn in the sand, maybe it was flowers, maybe it was seashells, as long as it was salable; easily tradable and accepted/recognized by most people as having value.

As time marched on, the need for these records to withstand our forgetfulness and the elements was tested, and only the most durable survived.

As humanities’ geographic footprint expanded, we began to travel ever increasing distances, so mediums of exchange that were portable and divisible were largely preferred.

Inevitably, for almost every medium of exchange that emerged, someone realized they could forge, or extract, a large quantity of it relatively quickly. This in turn leads to a rapid devaluation of the medium and loss of its status as a medium of exchange, so scarcity could not be ignored.

Over time, different kinds of money with different characteristics that were in demand at that moment came and went, but only the ones scoring the highest on these 6 attributes survived, until we eventually landed on gold as humanities’ money of choice.

Gold continued to hold its dominance for millenia, and while its unwieldiness for day-to-day use meant that other forms of money pegged to its value remained in circulation, such as silver, it wasn’t until the emergence of telecommunications in the 19th century that things would really start to change.

A new, previously hidden attribute of money suddenly came in to the equation: speed of settlement. This was the first time in human history where information could travel an order of magnitude faster than the human carrying that information could; thus it became a pressing problem not only to perform transactions quickly over long distances, but to have them settle as well. This, along with the perils of ferrying tons of gold bullions across tumultuous waters, and the rise of nationalism, lead to the introduction of the paper, and then digital, bills we use today.

These bills were essentially “IOUs” issued by local bank-like entities (ie. goldsmiths), regional networks (ie. Hawalas), and eventually indirectly via government controlled central banks. These bills could be redeemed for a fixed amount of physical gold at any time, hence the British Pound Sterling’s old moniker of “as good as gold”.

A £1 English banknote issued in 1814 by the now defunct Gloucester Old Bank.

This is perhaps one of the only, if not the only, time in history where a “softer” form of money won out against a “harder” form in a free market, which shows just how important speed of settlement is. And despite our regression to an overall inferior medium of exchange to satisfy our need for speed, this system worked well for a while.

Until it didn’t.

👑 The Loss of The Gold Standard

The era of The Classical Gold Standard from 1870-1914 was overwhelmingly a time of economic prosperity and growth, characterized by monetary stability, low inflation, economic integration within and across borders, and rapid technological progress.

The start of World War I would put an end to this, when the British and their allies became the first to de-peg their currency from gold in order to fund their wartime efforts.

Other countries took notice of this, and realized that it would be impossible to continue waging war, much less have a shot at victory, without doing the same and bolstering their military spending.

Info

Decades later through analysis of the central bank’s ledgers, it came to light that the UK government’s initial attempt to raise capital via traditional wartime bonds fell pitifully short, only managing to collect £236.5 million of the £350 million required from the private sector. In a coordinated cover up, the government would abandon its principles of transparency and adherence to free markets, discretely inject the remaining funds using the central bank’s reserves, and follow it up with a media blitz claiming the wartime bonds were a huge oversubscribed success.

Since time immemorial, mainstream media has been a mouthpiece for the establishment.

In a private memo to the Secretary of The Treasury at the time, this blatant embezzlement was praised as “masterful manipulation” by none other than the deviant whose ill-informed economic theories would be propped up by governments and central bankers around the world for the coming century, John Maynard Keynes.

You cannot make this stuff up.

The departure from the gold standard was meant to be a temporary wartime measure, but as the saying goes, there’s nothing more permanent than a temporary government solution. To give credit where credit is due, the next century would see several countries re-peg and de-peg their currencies to gold, but each attempt was met with more resistance and hardship than the last. It was particularly unpalatable when governments passed laws that made it illegal for their citizens to own gold, after which they would impose hefty fines and forcibly confiscate it, in an attempt to shore up central bank reserves and enforce more control over transactions. But there was a clear lack of incentive from central planners to make unpopular decisions or take away power from themselves - there’s no putting the toothpaste back in the tube.

Executive Order 6102, signed by President Franklin D. Roosevelt on April 5, 1933, forbidding the hoarding of gold coin, gold bullion, and gold certificates within the continental United States. Remember kids, it’s not theft if the government does it.

Fast-forward to around the end of World War II, when the US was at the height of its power and influence. In an effort to stabilize the ravished post-war global economy, a new framework was created with the stated goal of “promoting economic cooperation and stability”. The Bretton Woods system would introduce a layer of abstraction over gold and instead have the world’s currencies pegged to the USD, itself convertible to gold held in Fort Knox at a fixed rate of $35 per ounce, while governments would retain the right to adjust their pegs under specific conditions.

No nation or individual (that we know of) has ever even attempted to break into Fort Knox, since no one in their right mind would think of challenging US military might directly on their home turf. In fact it’s so impregnatable, that not even auditors have been able to get inside to complete a full audit since 1953. But don’t worry, the US government assures us it’s all there.

Predictably, putting the reigns of the world’s money in the hands of a few of individuals did not end well, culminating in 1971 when the US refused to allow other nations to withdraw their deposits and ended the USD’s convertibility to gold, driving the final nail in The Gold Standard and ushering in the current era of wonky economics, hyper financialization, and fractured fiat money backed by nothing.

By ending paper money’s redeemability to gold, the state now had the ability to infinitely “print money” to fund itself, and by extension infinitely inflate away citizens’ savings, with nothing more than a keystroke to edit a record in a database. This has created an inherently fragile financial system prone to dramatic boom-and-bust cycles, in what is essentially a return to a more controlled, better regulated version of Japanese Invasion Money. Except now, the state isn’t waging a financial war on other nations, but its own citizens and their descendents.

The Evolution of Money

There isn’t necessarily malicious intent and secret collusion by the powers that be (though historical and contemporary records show an abundance of that), but the allure of alleviating problems in the short term via increased money printing is irresistible for any governing body. As an indirect result of this, we’ve seen 44 of the 60 recorded instances of hyperinflation occur since that fateful day in 1971 (keeping in mind that this number does not include the many instances of very high or borderline hyperinflation, such as what recently happened in Türkiye).

Despite this, many mistakenly believe the USD, EUR, and other “hard” currencies are a safe store of value. While there is some truth to the USD in particular being the best form of fiat due to its status as the world’s reserve currency, it only seems like a decent store of value when compared to other fiat abominations. If you zoom out, the USD has been inflating away for decades, devaluing by more than 95% against gold in the last 100 years.

Some examples of how currencies have fared against the USD over time. From the top right going clockwise: hyperinflation, persistent inflation, the rare few that have managed to “hold their value”, and the peg. It’s worth noting that the Swiss Franc is unique in that it was the last to de-peg from the gold standard in 2000. Source: Xe

The loss in value of the USD against gold and loss of purchasing power over time mirror each other. Meanwhile, you’re able to buy much more goods with gold today than you could 100 years ago. It seems that markets intrinsically understand the value of gold, even if the average person doesn’t. Sources: Macrotrends and Visual Capitalist.

Though there are massive advantages that come with being the world’s reserve currency, chief among them the ability to sanction and devalue not just your own citizens’ savings but the savings of entire nations if push comes to shove, it’s not without downsides. The US needs to maintain a trade deficit to keep the world awash in dollars, which means it needs to import more goods and services than it exports. This has led to the hollowing out of the US manufacturing sector where the real productivity lies and the rise of the financial sector, the latter being in many ways a drain on the real economy.

The opposite can happen too, where a government can artificially weaken their currency to keep exports competitive, as China is currently doing with the Yuan. This means that Chinese citizens are not able to enjoy the full fruits of their labor, as the world becomes increasingly reliant on their cheap exports, though this strategy has proven quite successful at the national level.

🏦 An Aside on Fractional Reserve Banking

Almost as early as people began depositing their gold for safekeeping and for the convenience of paper notes, depositories realized that not everyone will withdraw all their money at once. This meant that it was possible to lend out a fraction of these deposits to other borrowers or businesses, and make some extra income on the side.

Depositories that engaged in this risky behavior managed to extract more profits and expand their businesses more quickly, while simultaneously offering their customers lower fees and more competitive financial products, with the caveat that they expose themselves and their customers to market forces, and most importantly the possibility of a bank-run should their customers doubt their ability to keep their promise and attempt to withdraw all their deposits at once, in which case the last to pull out their deposits will be left holding the bag.

Nowadays fractional reserve banking is how nearly all banks operate, most holding less than 10% of their reserves at any point in time (even zero reserves as of recently, which sounds insane until you remember that fiat money was never backed by anything to begin with).

By mandating a threshold of reserves, governments can control how much a bank can lend. When a bank makes a loan that is only partially backed by a real asset, it is temporarily introducing new money into circulation, putting the system in a leveraged state. It generates revenue from various fees when issuing this loan as well as interest arbitrage, i.e. by giving you a higher interest rate than the one it received. This creates an incentive for the bank to give as many loans as possible, but is balanced by the risk of the borrower defaulting, and in the past, the limited supply of assets backing these loans.

How fractional reserve banking increases the money supply. In this example, what started off as a $100 deposit has become $171 in loans, $27.1 in reserves, and $12.2 in eventual interest payments and fees.

The risks of fractional reserve banking tend to manifest acutely during times of economic turmoil. For example, in the US, the “Free Banking” era from 1837-1863 saw half of all banks collapse with a measly average lifespan of 5 years, which prompted the creation of the Federal Reserve in 1913 to act as a central bank and lender of last resort.

But even that didn’t prevent the Great Depression from 1929-1931, where over 9000 (no that’s not a typo) banks failed. This prompted the creation of the Federal Deposit Insurance Corporation (FDIC) in 1933 to insure deposits up to a certain amount, which has since been increased to $250,000 per depositor per bank.

While it’s true that banks today are better regulated than they’ve ever been and failures are far less frequent, the industry has become extremely consolidated so they tend to be larger and more painful than in the past; case in point, the 2008-2012 Global Financial Crisis saw 465 banks fail, and last year (2023) saw the collapse of 6 US banks, with two of them being the 2nd and 3rd largest bank failures in the country’s history. As long as a fractional reserve system is in place, the risk of failure always looms.

Moreover, even if the depositors were eventually made whole thanks to FDIC insurance, the FDIC holds around a cent and a quarter for every insured dollar - a fund of $137 billion against some $10.7 trillion of insured deposits, and that bailout money has to come from somewhere. In fact, I’d take this a step further and assert that fractional reserve banking has never truly worked, and only became somewhat workable due to the existence of a lender of last resort that can create money out of thin air (and inflate away peoples’ savings ad infinitum).

Warning

State economists will try to convince you that bank failures are a good thing, as they allow the market to “cleanse” itself of bad actors and inefficient businesses. But this is a false equivalence: banks are not like other businesses, they are the foundation of an economy, and their failure can have catastrophic consequences for the average person and every business that relies on them, efficient or otherwise - look no further than the existential threat startups recently faced with the collapse of Silicon Valley Bank.

Paradoxically, these same economists will also argue that banks are “too big to fail” and that they need to be bailed out at all costs. Perhaps rather than trying to justify the unjustifiable and introducing layers of beauraucracy and government institutions designed to polish a turd, we should be looking for a system that doesn’t constantly fail and doesn’t require bailouts in the first place.

There are other less visible but no less insidious side effects to this practice: for starters, depositors have little to no say on how their deposits get invested, leading to a situation like the one we find ourselves in today where every major US Bank has a stake in or has provided financial support to weapons manufacturers supporting genocide.

As stated earlier, the ability to create unbacked money creates a leveraged financial system. The result of this is that economic output and real value are divorced from valuation. Roughly speaking, if you know how long something takes, and how much effort it takes, you should be able to price it. But by pumping money into the system and continuously devaluing your time and effort, prices get distorted and this equation breaks down. Just like you can’t do science without standard units of measurement, you can’t properly measure economic output without a stable unit of account.

Are companies really worth “more”, or is the value of the dollar just worth less? Guess attending a Jerome Powell watch party during work hours was the productive thing to do after all. Source: Bloomberg, Apollo Chief Economist (BBG ticker: .GLMOSUPP G Index, SPX index)

What’s more, the interest that banks attach to these loans leads to a concentration of wealth and exasperates societal inequality. It saddles individuals with debt they end up having to pay many times over, significantly impacts their quality of life, and imposes what is essentially debt slavery in all but name. Today we see this manifest in everything from student loans to mortgages to credit cards, where individuals and businesses alike are heavily incentivized to borrow and maintain debt on their balance sheet, since the value of a dollar today is worth more than the value of a dollar tomorrow, which in turn leads to more money in circulation and higher prices, which in turn leads to more borrowing, which leads to higher prices, which leads to more borrowing, and so on and so forth. In such a system, your wealth is primarily determined by your proximity to the debt creation mechanism, formally known as the Cantillon Effect.

The Borrowing Cycle

The Cantillon Effect in action: the top 0.1% of the US population has seen the dollar amount of their wealth multiply by 10 times since 1980, while the bottom 50% have stagnated. Source: Board of Governors of the Federal Reserve System (US)

Perhaps unsurprisingly, these same dynamics created by fractional reserve banking at an individual and business level show up at the level of nation states.

🇺🇳 The Rise of Financial Neocolonialism

Picking up where we left off: despite the failure of the Bretton Woods system, the International Monetary Fund (IMF) and World Bank that were created to oversee it and provide loans to countries facing difficulties continue to exist today, albeit in a much more sinister form.

If the last section was the extraction of users, this one is the extraction of everybody else. A platform that has taken what it can from the people using it turns next on the businesses that came for those people; a monetary system that has taken what it can from its own citizens turns next on everyone who has to hold its currency in order to trade.

Originally they would provide financial assistance for post-war reconstruction and development processes. However, as the decades went by, the IMF and World Bank would become the primary lenders to developing countries, and the conditions they imposed on these loans would become increasingly predatory.

With the dispassion of a private equity firm that is focused on nothing but increasing revenue, they would impose austerity measures alongside these loans that would in aggregate lead to mass poverty and unemployment, included but not limited to:

Requiring countries to open up their markets to foreign investment
Privatizing state-owned enterprises
Cutting government spending on public services like healthcare, education, and social welfare programs
Shifting agricultural production away from food crops for domestic use and towards cash crops for export
Limiting or restricting the export of certain natural resources or manufactured goods that the country could otherwise use domestically
Mandating that a certain percentage of key export commodities must be sold on the global market

All while applying heavy pressure on leaders to debase their currencies and yet accepting payments in gold.

Only under a fiat standard can the world’s food basket be reduced to importing basic necessities. Source: howMuch

This creates a vicious cycle where indebted countries need to take out more loans to pay off previous ones, saddling them with an even greater burden and even harsher “structural adjustment programs”, causing them to remain in perpetual debt to the IMF and World Bank, and by extension the countries that control them, ultimately turning them into vassal states.

Albert Einstein once said “Compound interest is the 8th wonder of the world. He who understands it, earns it; he who doesn’t, pays it.”, and boy are these countries paying it. What’s more, the IMF imposes “surcharges” which add an extra 2 or 3 percentage points to the most indebted borrowers’ loans. Sources: Visual Captialist and CEPR

To make matters worse, very few of these loans are put towards their intended use. The vast majority are needed just to pay off the interest on previous loans, while the rest are used to line the pockets of corrupt politicians and their cronies, as they live in extravagance with palaces, yachts, and supercar fleets, all while increasing the state budget for policing their own citizens and clamping down on any dissent.

Further adding insult to injury, not only are these misused loans taken out without the consent of the people, but they are also taken out in their name, and they and their descendants are the ones who end up bearing the brunt of the consequences with none of the upsides.

US President Donald Trump’s favorite dictator Abdel Fattah el-Sisi’s new palace and prison. Somehow the government is always able to find the funds for these types of projects, yet Egypt remains in perpetual poverty and political repression, despite possessing immense resources and human capital, and being one of the largest recipients of IMF loans and US military aid. All this before the latest round of loans and currency devaluation too.

Even in democratic countries where the elected leaders are aware of the dangers of relying on IMF loans, most if not all still end up taking them, and they can hardly be blamed. The need to alleviate immediate suffering and a political environment that favors short termism to boost re-election chances means they probably couldn’t get away with anything else.

Beyond the geopolitical motives, the only reason the IMF and World Bank are able to recklessly give out loans that have questionable benefits at best, disastrous consequences at worse, and no hope of being paid back, is simply because they’re incentivized to do so. Just like your local bank, they are able to conjure up infinite dollars at close to or zero interest, then lend them out at much higher rates, an incentive further strengthened by their ability to levy structural adjustments to force the wants of global superpowers over the needs of the indebted, resulting in an amped up massive scale Cantillon Effect.

Info

Ironically, the IMF published a working paper in 2012 advocating for a return to full reserve banking, which just goes to show that individuals within these institutions are aware of the problems they’re causing, but are powerless to do anything about it.

Unfortunately, I don’t see a top-down mandated return to full reserve banking happening anytime soon, not only because it would mean the IMF, World Bank, and every state actor who benefits from the current system would have to give up their power, but because unwinding the leveraged system would require a bailout >184% of GDP in the US alone (and this was back in 2012, it’s likely an order of magnitude higher now).

Even a bottom-up approach is seen as a threat, which is the exact reasoning the US Fed recently gave when they denied the application of a new bank that wanted to operate on a full reserve basis, fearing that if they gained traction, it would be “too disruptive to the financial system”.

Thankfully, there has been progress on other types of licenses that allow full reserve banking (with some limitations on the kinds of products you can offer), and a growing number of institutions being denied full-banking status are appealing the Fed’s unconstitutional decisions, and I’m hopeful that they’ll be able to operate soon.

🌐 Towards a New Global Ledger

It’s easy to look at the state of our financial system and wonder how we managed to screw it up so badly, and blame our predecessors for being simpletons with zero foresight. But it’s the height of hubris to think that things would be any different had you or I been in charge. The unwieldiness of using physical gold as the monetary standard was a major challenge which fiat provided a much needed solution for, and should we replay the events of the last century even while knowing what we do now, it’s unlikely that we’d end up with a materially different result.

Every actor in the system is simply playing the best hand they’ve been dealt; the global superpowers are trying to preserve or expand their hegemony, the banks give out loans because their mandate is to increase revenue, and people/countries take those loans to satisfy their immediate needs and wants. That’s not to say some entities aren’t more blameworthy than others, just that there isn’t necessarily a single entity responsible for all this, but an all corrupting system with fiat at its core.

Rather than look to the past, it’s more prudent to look to the future and ask ourselves, if we were to create a new global financial standard, what would it look like? To answer that, we need to start with a simpler question: what currency or commodity would be at its core?

For starters, it would need to score high on the characteristics of money mentioned earlier, so it would need to be:

Scarce, with a high stock-to-flow ratio or even a fixed supply
Indestructible
Able to instantly settle over long distances
Divisible into any number of units
Widely recognized as having value

And since we’re letting our imagination run wild, why don’t we also make it so that it’s:

Permissionless - no one can stop you from using it or taking custody of it if you’re the rightful owner
Censorship Resistant - no one party can arbitrarily modify it
Strikes a balance between optional privacy (for transactions) and transparency (for auditability)

Alas, synthesizing such a material in the physical world is simply beyond our current scientific means.

But in the digital world, it’s a different story.

In 2008, on the heels of yet another global financial crisis, an anonymous programmer under the moniker of Satoshi Nakomoto would bring us to the closest we’ve ever been to this ideal form of money and create his magnum opus, Bitcoin. Though not quite there yet, it nonetheless possesses all the monetary properties of gold and then some, but none of the physical ones.

Originally called “Electronic Cash System”, it was far from the first attempt at implementing a form of electronic money (in fact writings on decentralized money go as far back as 1976, and the idea of a new form of money suitable for the information age goes even further back to 1845), but it was the first to solve the double spending problem without the need for a trusted third party, and in the coming 16 years would gain market acceptance as the money movement protocol of the internet age, proving itself to be the most secure, decentralized, and censorship resistant form of money to date.

Money Tier List

I’ll delve into the exact mechanism of how it works in later blog posts (if you can’t wait, watch this) in addition to a deep dive into the technical details and current challenges/shortcomings, but for now the best way to understand it is to compare it to another type of money used in the past on the island of Yap in the federated states of Micronesia: Rai Stones.

The Rai were massive circular limestones that were challenging to move around. Instead, the stones would remain stationary in one location and the islanders would orally keep track of who owned which stone. When a transaction was made, usually for a large expense such as a funeral or a wedding, the communal ledger would get updated. Such a simple system was surprisingly robust, as the effort and danger involved in extracting and transporting stones limited the amount of new supply that could be introduced, and unless more than half the tribe conspired against you, no one could steal or claim possession of your Rai. It wasn’t until the introduction of modern sailing ships that the scarcity of the Rai would end and the system would show its first signs of collapse, followed by German colonization that took control of the stone ledger and forced the use of fiat money.

Despite losing its monetary value, the Rai is still used in ceremonial exchanges to this day.

Though a bit of an oversimplification, Bitcoin is just a modern, digitized, and decentralized version of The Rai with a known total supply (21 million, though the amount is arbitrary) and a set, front-loaded “mining” schedule (every 210,000 blocks, i.e. ~4 years, the supply of new Bitcoin halves, with ~19.8 million already in circulation). It’s a single ledger with a single source of truth, distributed in such a way that it can be verified by anyone with an internet connection at any point, and is near impossible to be tampered with. Instead of a community of islanders who ensure the integrity of the ledger, you have an army of computer programs (known as Nodes) validating transactions, and instead of attributing ownership to an individual it is attributed to a wallet address.

Warning

It’s important to emphasize that Bitcoin is not the same as “crypto”, let alone “blockchain” (which is just a data structure), something media outlets tend to frequently botch. It’s better to think of Bitcoin as a commodity backed by the energy it takes to secure the ledger, rather than the various cryptocurrencies - most of which are not even currencies - that have been the centerpiece of many an internet scam. Even if you ignore the obvious scams and celebirty shitcoins, many of these projects are just misguided attempts at decentralizing something that doesn’t need to or even benefit from being decentralized.

That’s not to say these so-called Web3 projects don’t have their use cases, specifically those that aim to provide a utility. For example:

Democratizing access to censorship resistant compute, storage, and data feeds, ie. BitTorrent, Rndr, and Chainlink
Tokenization of real world assets, which allows for permissionless exchange of value and instant settlements, ie. USDC or PAXG
Creating and propelling the adoption of standards in industries that benefit from network effects, ie. making virtual assets interoperable across different games

But being a sound money, much less a store of value, is not and will likely never be their use case.

What makes Bitcoin particularly unique is not just the underlying technology. The sequence of events since its inception, leading up to today, are outright impossible to replicate. Don’t believe me? Then go ahead and clone the Bitcoin codebase, change the name, launch your own version, and see how far you get.

Despite an endless torrent of replicas and competitors, even if some are technically superior, none have been able to dethrone Bitcoin, neither in terms of market cap nor usage, and unless an extremely powerful entity backs them, it’s nearly guaranteed none will. It’s not just the first mover advantage, but the network effect and community that has been built up over the years, the battle-tested trust that has been established, and the careful balance of incentives for ecosystem participants, all which have made it so that there is a single recognized canonical implementation of Bitcoin.

Tip

As alluded to earlier, if you think of Bitcoin as a money movement protocol, it becomes easy to recognize its value and why it’s not going away any time soon; the most technically advanced protocol does not always win. One of the most notorious examples of this is the SMTP protocol responsible for sending and receiving emails; it’s horribly inefficient, security was literally non-existent, and yet it has become a cornerstone of the internet, and it’s unlikely to be replaced anytime soon despite there being far superior alternatives. Instead it’s being used alongside other protocols that achieve a similar purpose (ie. RCS/MMS/SMS for messaging), and there are infrequent updates and improvements to the standard, plus new layers that are built on top of it to improve its functionality and security. Tens if not hundreds of billions of dollars worth of businesses and infrastructure exist just to make it workable.

The Crypto Midwit

While far from perfect, because of all these properties, Bitcoin has the potential to be the foundation of a new and sound global financial system, and that potential has been reflected in its market price. Even if it doesn’t end up living up to its promise, we now have the blueprint for what sound money in the digital age would look like, and the most suitable candidate to be the base layer of a new global financial system will be one that resembles Bitcoin, if not Bitcoin itself.

Not pictured: the ~$144 trillion US treasuries market. In the last 20 years, Bitcoin has gone from being worth less than a penny to being worth more than the GDP of most countries, briefly surpassing silver’s global market cap, and with a growth rate only matched by GPU maker Nvidia. Source.

But how would an economy not based on fiat or fractional reserve banking even function? How would people and businesses engage in economic dealings?

What if I told you that you don’t need to imagine such an economy, because the blueprint for it already exists?

🕌 An Abridged Primer on Islamic Finance

Many non-muslims are surprised to learn that Islam has a lot to say about money and finance. It’s not just about praying 5 times a day or avoiding pork and alcohol; it’s a comprehensive religion with guidance on every aspect of life, including how you earn, spend, and save money.

The topic of morality of money is a vast and complex one, and I’m not qualified to delve into it here. However, I can provide a brief overview of the most relevant principles to us today.

Islamic Finance, at its core, is about promoting real economic activity, discouraging speculative behavior, upholding financial stability, fostering community development, encouraging low time preference thinking, and ultimately aligning economic activities with moral values.

To ensure this, it lays out a set of rules, which if followed, inevitably lead to a sound financial system. The most relevant to us today are:

An extremely strict prohibition on engaging in usurious transactions (e.g. interest bearing loans), whether issuing or receiving them.
Price fixing is prohibited, the value of goods should be determined by supply and demand.
Contracts and financial transactions need to be built on the basis of equity and risk sharing. This renders explicit that there is no such thing as a “risk-free” business or investment.
Financial transactions should be backed by a known amount of tangible assets. Selling that which you do not own, or trading in derivatives, options, futures, and other highly speculative contracts is (in most cases) prohibited.
Investments in businesses deemed unethical or harmful (e.g. alcohol, gambling) is prohibited.

Even if you’re not a Muslim or don’t ascribe to Islam’s teachings, taking these rules in isolation, it’s impossible to refute their wisdom. Had they been applied correctly, we would still be on something akin to a gold standard today, there would be no neverending wars and financial crises, businesses would be allowed to go bankrupt to usher in value creation for more than just the boomers, and companies like Kalshi and Polymarket wouldn’t be wrecking havoc on the young and vulnerable.

Unfortunately, the current system is anything but sound, and Islamic banks are no exception. In fact, the overwhelming majority are just conventional banks with a few tweaks to make them “shariah compliant” while ignoring the tainted foundation they’re built upon, and the majority of Islamic scholars have begrudgingly given their blessing to fiat money, fractional reserve banking, and other practices that are clearly at odds with Islamic principles. Even if you don’t participate in the economy in any way except for using a checking account, it doesn’t matter because fiat money has interest baked in at the protocol level; it’s nearly impossible to escape the dust of usury.

Again, it would be hubris to assume we would’ve been able to create a better system or to dismiss the efforts of our predecessors, or to claim that the fatwas (formal rulings) on the permissibility of fiat were issued by scholars that lacked deep economic knowledge or were in the pockets of the state. The much more likely explanation is that our predecessors simply lacked alternatives, and the current system was the best compromise they could come up with.

Nonetheless, with the emergence of a technology that resolves the core issues of fiat, I believe there’s no longer a good excuse for continuing to use the broken system we were all born into, at least beyond what is strictly necessary, and that there’s an opportunity to re-work the financial system from the ground up in a way that is beneficial to all of humanity, not just Muslims.

To borrow a metaphor: if you found yourself stuck on an island where there’s nothing but pigs, then it’s permissible for you to consume them to avoid starvation. If you one day discover the existence of rams, the pork is once again forbidden and you’re obligated to consume the rams instead.

We’ve found the rams 🐏.

🛡️ Common Objections to Bitcoin, Answered

Though I could write entire articles about it, it would be remiss of me not to address the objections that come up as soon as people start taking Bitcoin seriously.

You can already guess what my opinion on the matter is, but keep in mind that I’m extremely biased. I’ll simply share a summary of the common objections and my commentary. You need to do your own research and decide for yourself; don’t trust, verify.

As with any novel technology, especially one that doesn’t cleanly fit into our existing frameworks, skepticism is the default reaction, and most of it doesn’t hold up to scrutiny. I’ll go a step further: not only does Bitcoin clear the bar, I’d argue it’s the best, most technologically advanced form of money humanity has produced.

Nonetheless, let’s go over some of the common arguments against Bitcoin, and why I believe they are flawed:

It is unclear if Bitcoin is a commodity or a currency; this drives a lot of confusion, as the two are treated differently under most legal and financial frameworks. To this I say, it is a commodity that can be used as a currency, and the fact that it can be used as a currency is what gives it value. This is not a new concept, as gold has been used as a currency for millenia, and is still used as a currency today in some parts of the world. Though one could argue that gold had a use case before it was used as a currency and still does, it’s not these use cases that give it such a high market value, but its monetary properties.
Bitcoin is not backed by anything; not only is the premise of this argument false as Bitcoin is backed by the energy it takes to secure the ledger, and the energy cost of securing the ledger is increasing over time, it ignores the fact that fiat currency is also unbacked, and neither was gold or any other commodity that acted as the lynchpin of a financial system.
You cannot physically hold Bitcoin; this is true, but it’s also true of fiat money, as most money in the world today is digital, existing only as numbers in a database. Moreover, we already recognize that digital assets, such as domain names, logos, and patents have value, and that they can be bought, sold, and traded. Bitcoin is no different.
Bitcoin is used for illicit activities; again, this is true of every other asset, including fiat money. In fact, humanity’s worst crimes are funded by fiat printing, and the US dollar is stained with blood. If anything, the transparency afforded by the blockchain and the ability to view the holdings of every wallet address makes Bitcoin an inferior option for illicit activities.
There is no central authority controlling Bitcoin; this is a feature, not a bug. The lack of a central authority means that no one can manipulate the supply of Bitcoin, and that the rules of the system are enforced by the network itself, rather than a handful of individuals. This makes Bitcoin secure and censorship resistant, enforcing a sort of forced cooperation and preventing one country or entity from unilaterally devaluing it. This means that just like gold, the value of Bitcoin is determined by the market. What’s more, the decentralization mechanism is crucial to maintaining the integrity and immutability of the blockchain. What can be improved, then, are the laws and regulations around Bitcoin, which are still in their infancy, and the lack of a clear legal framework in many jurisdictions can be a cause for concern, but this clearly isn’t an issue with Bitcoin itself but with our antiquated bureaucracies.
The energy consumed in Bitcoin mining is wasteful; once again, a false premise, as the energy consumed in Bitcoin mining has the critical function of securing the ledger - think of the massive amounts of processing power guarding the network as the equivalent of the tanks guarding Fort Knox. They make it so that anyone attempting a brute force attack would first have to spend billions of dollars to even stand a chance at succeeding, not that there would be any incentive in doing so, since a security failure is sure to negatively impact the value of Bitcoin. What’s more, Bitcoin mining actually has some benefits, as it stabilizes the grid, makes use of energy offshoots, encourages the development of clean energy sources, and allows for the near-perfect transfer of the value of energy. Even if you don’t agree on the benefits of Bitcoin’s energy consumption model, the energy used in mining gold or securing the banking system is an order of magnitude greater than that used in Bitcoin mining, and the environmental impact of gold mining is far worse.
Bitcoin cannot scale and will never be used in day-to-day transactions, thus it is not a serious currency; this is a valid concern, as Bitcoin’s current transaction throughput is limited to 7 transactions per second, and the fees can be high when the network is congested. However, the Bitcoin network is constantly being upgraded, and there are several solutions in the works that may allow it to scale to meet the demands of a global currency, the most promising being the Lightning Network, a second layer solution that allows for instant and near-free transactions, which is already live and being used by millions of people (though it too has its own set of shortcomings). This also doesn’t take away from the scenario that Bitcoin could still be used as a reserve currency rather than a day-to-day currency, just as silver was used alongside gold.
Bitcoin’s volatility makes it unsuitable for day-to-day transactions, and instead makes it a speculative asset; there’s no doubt that storing your earnings in Bitcoin could expose you to significant losses, though you’d have be a particularly unlucky individual to somehow still lose money on an asset that appreciated from $0.000764 to tens of thousands of dollars. Nonetheless, it remains the case today that many use Bitcoin speculatively, though this is also true of every other asset. Even fiat has day-to-day volatility that we are not aware of, and depending on where you’re from the currency you use is likely more speculative and prone to collapse than Bitcoin, bending to the whims of a few (likely corrupt) individuals. Meanwhile the price of Bitcoin is driven by a combination of global speculation and fundamentals, with the latter becoming more dominant over time. It has already overcome some serious challenges that could have shut it down, yet instead they’ve only served to strengthen the case for its use as a global reserve currency. To date, it has survived being outlawed in several countries (most prominently China, since it conflicts with their new global digital ledger project), numerous high profile hacks, the collapse of several multi-billion dollar exchanges (most notably FTX), several attempts to make undesirable changes to it’s source code, and still its percentage in cash holdings globally continues to rise. That’s not to say that something in the future won’t happen that seriously degrades the integrity of the network, but as more people use Bitcoin, the price becomes more stable, and the speculative nature decreases. If you’re a risk averse individual, this caution probably aligns with your thinking, though I’d encourage you to recognize the risks associated with holding any other asset.

Thus I posit that upon closer inspection, most arguments against Bitcoin tend to be based on a misunderstanding of how it works, or a lack of understanding of the current financial system, and have likely been misrepresented by those consulted on the matter. The arguments for Bitcoin tend to be based on a deep understanding of both, and the realization that most arguments against apply even more strongly to fiat. Regardless of your opinion, one cannot deny that it’s simply the best alternative we have to the current system, despite not being perfect.

🔑 The Exit

Fiat’s record isn’t mixed, it’s perfect - in the way a disease with no survivors is perfect. Every fiat currency ever issued has lost purchasing power. Not most of them, and not only the ones run by obviously corrupt regimes: all of them, across every form of government, every century, and every level of national wealth. The dollar has kept roughly three of the hundred cents it was worth in 1913, and the dollar is the success story - the reserve currency, the one every other currency is measured against. What looks like a survivor is simply a currency that hasn’t finished failing yet. There’s no counterexample to reach for, no country that found the discipline, no decade where the temptation was declined - and at some point a 100% failure rate across two centuries and every nation that ever tried stops being a run of bad luck and starts being a property of the design. A government that can print, prints. Debasement is the only tax that never has to be voted on, and nobody who has ever held that lever has left it alone.

For the first time in a century, money has an exit that doesn’t require moving gold across an ocean or asking your government’s permission. You still can’t delete the dollar. But you no longer have to hold it, and once you understand what holding it actually costs you, you’ll be racing towards the exit.

📜 Acknowledgements

Many of the insights in this post were only made clear through the scholarship and thought leadership of individuals who chose to make their work accessible to all, most notably Dr. Saifedean Ammous, Lyn Alden, Harris Irfan, Ray Dalio, and Alex Gladstein.

The Bitcoin community by and large, especially Bitcoin Design.

And of course, those friends that wouldn’t stop talking about Bitcoin.

Is Your Idea Worth Pursuing?

Sun, 07 Jan 2018 00:00:00 GMT

Photo by Rohan Makhecha on Unsplash.

There are no lack of great ideas floating around.

In some sense, it’s our responsibility to transform these gifts from the muse into a reality. But when you’re always trying to solve problems or improve on what’s out there, you’ll likely be struck with an epiphany of THE NEXT BIG THING™ at least once a week. You’ll get excited, do the research, formulate a business plan, pitch to your friends, get some people on board, and start building an MVP (and not necessarily in that order).

You’ll then be struck by an equally powerful, opposing epiphany: your idea actually sucks. And when two opposing forces act on the same point, they cancel each other out, taking you right back to where you started. The muse may be generous with her gifts, but there’s no way to really know the contents until you look inside. You can, however, make a damn good guess.

So how does one weed out the seemingly great from the truly great ideas? How do you ensure they then actually go somewhere?

When you choose to pursue an idea, you’re investing in it. But rather than investing money, you invest something far more valuable: time. That’s why the same principles used in monetary investing can just as easily be applied.

The following is the framework I developed through much research and experience to decide if an idea should be pursued. Most seemingly great ideas will pass at least half these criteria. Drop those. Don’t underestimate the amount of time and dedication it takes for your idea to materialize; this will be your life for the foreseeable future. Only when an idea passes all the criteria should you seriously consider pursuing it.

1. Would I use it? Am I addressing my own pain points?

Whether in the context of a business or commercial product, if it’s not something that you envision yourself using, don’t bother. But what if you’re not the target market? All the more reason not to pursue it. You’ll never be able to completely understand the customers’ way of thinking.

2. Do I have an unfair advantage?

Make that advantages. The more the better. You want to be driving a Tesla while everyone else is riding a bike with a flat tire. Not only will your journey be far more comfortable, you’ll achieve the same results at a fraction of the effort. Whether it’s creating a blue ocean of uncontested market space, access to exclusive resources, or an extremely deep understanding of the domain, so long as it’s something that the competition can’t hope to emulate.

3. Where do I see this going a few months from now? A few years?

Timing is arguably the most important factor for the success of any idea. It doesn’t matter how genius it is if the market isn’t, or won’t be, ready for it.

Unfortunately, humans are horrible forecasters. The longer we project into the future, the more offbeat our predictions become, which is compounded by the fact that we view our own ideas through a rose-colored lens. That’s why one should make their predictions on a spectrum from best-case to worst-case, and assume the worst. If it’s an all-or-nothing bet, it’s probably nothing.

4. What am I really trying to achieve? What is the underlying vision? Is it something that provides value to others?

Ever wonder why Uber for laundry isn’t a thing yet? One possible reason is that their vision statement would be something like “Enabling the upper-middle class to embrace their laziness”. Your idea doesn’t have to be grandiose, but your vision does.

Embedding your idea with an authentic sense of purpose beyond “make a cash cow” is how you’ll survive the inevitable setbacks and periods of doubt. Moreover, it’s what will inspire the truly committed to join you. Lastly, a strong vision provides the foundation for your strategy and affords you more opportunities to pivot, diversifying the ways your idea can evolve down the line.

5. Do I understand the underpinnings? Am I out of my league?

The most difficult criteria to fail, because failing here requires you to accept that, at your current level, you lack the skills necessary to steer your idea from abstract thought into reality.

For some reason the conventional wisdom nowadays is to “figure it out as you go”, when it should really be “figure out parts of it as you go”. You wouldn’t toss yourself into the middle of an ocean if you’ve never been a pool; but if you have, then you’re more likely to successfully navigate the current and swim back to safety.

Remember, it’s not about what you know, it’s about what you don’t know.

6. Trust your gut.

Human beings are irrational, and any good decision making framework takes that into account; you mustn’t discount the significant emotional aspects of decision making.

Very, very, very rarely will an idea that hasn’t passed all the above criteria continue to nag you. Despite everything indicating otherwise, there’s something in your gut insisting this will work. If that happens, then there’s just no helping it; you have to give it an honest try. You’ll regret it otherwise, and peace of mind is well worth any investment. Who knows, it could just be that one in a million crazy idea that actually works.

On the other hand, sometimes an idea will pass all the criteria, but your gut will insist this won’t work. It could just be that you’re overly cautious, lack confidence in your abilities, or are afraid to take the next step. Or it could be that your conscious thinking has yet to catch up with what your subconscious already knows. Use your veto power sparingly.

At the end of the day, most successfully realized ideas are black swan events. No two will develop under the same circumstances and the outcome is unpredictable by nature. As such, no framework can encompass the complexity of the real world. The point of these criteria is not to give you analysis paralysis or provide absolute laws, but to get you to think deeply about and be mindful of yourself and your surroundings. If you find you’re saying ‘no’ to a lot of great ideas, then it’s working.

This article was created to make my life easier and provide a concrete framework for evaluating ideas. If you put it to practice and found it useful in evaluating your own ideas, be sure to share it with others who might benefit!