Skip to main content Redeem Tomorrow
I used to be excited about the future. Let's bring that back.
About Get my help
  • LLMs won't save labor when you use them like this

    The developer tools community remains active in Twitter’s dying light. On Tuesday, August 27th, in a now-deleted tweet and blog post, Matteo Collina boasted benchmarks across several JS frameworks that showed his fastify outperformed alternatives in server-side rendering performance:

    A graph of SSR performance showing fastify on top

    There was just one problem: the implementations were written by a large language model. In a paragraph that will live as a cautionary tale, Collina discloses (my emphasis):

    For this, we needed to generate a non-trivial sample document that includes a large number of elements, to have a very large page for the test, and consequently have more running time to capture each libraries’ performance. So, we asked an LLM to write some code to draw a spiral into a container using divs[…]

    For the authors of the other frameworks, the benchmarks did not pass the smell test. Svelte’s Rich Harris was the first I saw on the scene, and he made a single-line fix that immediately changed the outcome of the benchmarks:

    Revised benchmarks that place Svelte even with fastify in its current version, and much faster in Svelte 5

    tiles = [...tiles, { x, y, id: idCounter++ }]

    became, in Rich’s pull request,

    tiles.push({ x, y, id: idCounter++ })

    In the Svelte implementation only, spread is used to add tile instances to the tiles array. This has performance implications because on each iteration of the loop that creates tiles, the array has to be created anew, hobbling Svelte’s efficiency. Using push(), by contrast, doesn’t incur the overhead of creating a new array on each cycle. It preserves and mutates the existing array.

    Both cases are, strictly speaking, working code. But one works far differently than the other.

    Errors in the other implementations surfaced as well, including running React in development mode.

    These are the risks of using LLMs for business. Collina and his colleagues have, at this point, lost any labor savings they might have enjoyed from generating this code. They’re reviewing pull requests from aggrieved maintainers of other projects, they’re responding to messages, they’re doing damage control, making apologies… To say nothing of the sector-wide labor costs of other maintainers correcting the record, and the benchmark code.

    All because the LLM isn’t actually able to apply any sort of judgment or discernment. The process for generating this code, for example, did not catch the anomalous implementation of the Svelte loop, compared to the approaches in other frameworks. Generating the different projects was likely to be the output of distinct runs of the LLM.

    But more than that, the LLM is not considering the performance implications of spread vs. push because the LLM cannot consider anything. It’s a loom that generates patterns of information, endlessly.

    Sometimes that ends up as working code. Last month, I used Claude to get a demo application working for a class I was teaching on the social consequences of algorithms. I wanted a toy Twitter whose feed the class could manipulate in real time by applying boosts and penalties to certain kinds of content.

    Claude had me up and running in half an hour, and everything worked great. To my shock! I’d never gotten an LLM to generate multiple, interdependent components that worked together successfully in the past. This was very helpful to me, and given my own time constraints, made the difference between a nice idea and a working demo.

    But there’s a big difference between code written for toys and demos and code written for accomplishing a business purpose. In the fastify case, we see that the LLM is no match for the judgment of real humans.

    There are obvious uses and benefits to a technology that can extrude workable patterns of information, especially code. But uncritical use of that code is going to create a real mess.

    There’s a whole job category implied by this LLM future:

    People who, like these framework maintainers, apply their experience and judgment to remediating LLM lapses.

    So much for that labor savings.


  • 'Layoff'

    Alejandra stared across the table, waffle uneaten. A prim scoop of butter wept down one edge, indifferent.

    Across from her, Paul just sighed. Letting her have the moment.

    “So you just woke up with your devices dead?”

    “Just like that,” Paul nodded. “Email in my personal account. So long.”

    Their one-on-one tradition, a long, indulgent breakfast twice a month, was the one thing Alejandra looked forward to at work. She’d get advice, hear gossip, and otherwise commiserate with the only boss she’d ever had who tried to keep her safe.

    Now that was gone.

    “So now who’s running the team?”

    Paul shrugged, casting a vacant stare into his pile of corned beef hash. Also uneaten. “I’m a little surprised it’s this messy. Usually you get a sense of what’s happening before it drops, you get someone really curious about details they never cared about before. Nothing this time.”

    Somehow they got some of their meal down. They parted with a hug, promises to keep getting the occasional breakfast, and each began a glum walk in opposing directions. Paul was going home, Alejandra had a workday ahead of her.

    It was this way across the company, Alejandra found, her phone lighting up with backchannel gossip as she walked. Big round of layoffs that also vaporized whole layers of middle management, and plenty of executives as well.

    The org chart was shredded.

    She kept texting people the same question: “How are we supposed to function?”

    As she arrived to the office, mostly empty, she got her answer.

    A calendar invitation for a noon video call.

    With the CEO.

    They’d met a couple times. At 2,000 people, the company had been just small enough you’d get a few seconds with the CEO in the odd meeting or holiday party. But Alejandra certainly didn’t have regular meetings with the guy.

    “So tell me a bit of what you’re working on!” Johnny’s cheer seemed sincere, but that made it strange. She’d had a terrible morning. She wasn’t in a cheery mood.

    But the rent was not going to pay itself.

    After she gathered herself, Alejandra explained the effort to improve efficiency in the company’s resource usage. Their software design had evolved in recent years, but they were provisioned for an architecture that had been much more gluttonous. There was margin to be found in the difference, and Wall Street was eager to taste it.

    “We’ve tested this in staging for the last month and gotten things to a very stable place. We’ll roll it out over the next month or so, and we’ll be able to cut back on costs before the quarter is over.” Alejandra and her team had worked hard to get this locked in, and she was proud it wasn’t going to slip.

    Nice to be able to give the CEO good news.

    Johnny listened, and just as she got to discussing specific timelines, he cut in. “You know what, I’m thinking we can course-correct on this. For now let’s leave the existing computing capacity in place.”

    Alejandra fought the reflex to object. This was the CEO after all. But she and her team had spent six months on this. They’d been told over and over that not only was this urgent, it would impact everyone’s pay over the long term, as it would make the company more appealing to investors. She felt her pulse quicken, wanting an explanation.

    But instead, she nodded slowly. “Where should we put our focus now?”

    Johnny asked her to put a plan together now for increasing capacity. “I’m sure you already know your way around thanks to the last project. Give me some idea for the costs and complexity involved with an extra 50% of computing headroom. We need plenty of juice to grow into what’s next.”

    She nodded again, masking her confusion. “I’ll give you ten minutes back!” he said, as though this was quite generous, cutting the call.

    The backchannel chatter continued into the evening, and Alejandra had to charge her phone hours earlier than usual. A new encrypted group chat formed with two dozen managers and team leads.

    By the end of the week, things were weird.

    ale: okay, so did ANYONE this week NOT meet with Johnny?

    A flurry of eye reacjis glommed onto her message.

    bree: literally three times for me
    
    thomas: how does the CEO of a public company have time to take THIS MANY MEETINGS?
    
    lil: and why are we completely changing course on everything?

    The picture was consistent. Across engineering, everyone had been re-tasked. Other functions had been told to pause.

    No one had clarity on a new management structure.

    But after a few weeks, the picture became clear. It was just lots of meetings with Johnny.

    One morning, yet another one-on-one meeting with her CEO appeared on Alejandra’s calendar, just an hour away. She had to reschedule a few things to accommodate.

    When Johnny joined, all of his previous cheer was gone.

    “So walk me through how things have gone since the last time we talked,” he prompted, his face grim.

    Alejandra began rattling off the work needed to create the spare capacity Johnny had asked for, along with its progress.

    He cut in. “That’s great, but I need you to pause that for the time being. Instead, I want the infrastructure team to focus on a system we can use to rapidly scale back our computing capacity, while still prioritizing high-value customer workloads.”

    Alejandra’s frustration finally got the better of her. “Why? What would we need that for? You just told me to add more.”

    Johnny paused, tilting his head as though regarding something both pitiable and frustrating. “Please just action this.”

    He cut the call.

    In the backchannel, once again, Alejandra found herself not alone with the strange interaction.

    cam: anyone feeling whipped around by our new boss? Feel like I'm getting conflicting directives every few days now
    
    bree: It's honestly like talking to a different person sometimes
    
    thomas: again, where is he finding the time
    
    ale: and why is he so moody?

    The next day, just as Alejandra was about to close her computer, leave the office, and meet some friends for a badly needed dinner on the town, she got a direct message from Johnny.

    j_hnny: hey do you have a few minutes?

    What could she say?

    As she joined the video call, Johnny seemed poorly rested.

    “Do you ever worry things aren’t going to work out for us? As a business, I mean?” Johnny looked at her, almost pleading. He wanted a contradictory point of view.

    He wanted encouragement?

    “Uh, I mean,” Alejandra started. This wasn’t her wheelhouse at all. She kept services running. Business analysis was someone else’s job.

    To say nothing of counseling executives who made multiples of what she did.

    “You know, I guess I just try to do the best job I can and trust that you and your team can figure out the path forward.”

    Johnny’s face fell.

    “What if we can’t?”

    Alejandra’s skin crawled. What the fuck.

    They were both silent for a beat.

    Finally, Alejandra ventured, “Why are you worried right now?”

    Johnny frowned, looked away. Then he told her, “I can’t talk about it. I better go.”

    He cut the call.

    As she crossed the threshold of the lobby into the street, her phone buzzed.

    Johnny again. What the fuck?

    “Hey there,” he said briskly, as though he hadn’t just had the most bizarre fucking conversation with her minutes earlier. “How far did you get on the scale-back system we talked about a few weeks ago?”

    Finally, Alejandra had enough. “Johnny, we didn’t get anywhere on it because two days later you re-tasked us again!”

    “God dammit,” Johnny spat and then the phone beeped, the call ended.

    I hate this fucking job, Alejandra seethed to herself.

    She began walking to dinner when her phone chimed, overriding her silent switch. She was on-call, and an incident had just tipped over.

    Fuck!” Alejandra cried, at nearly the top of her lungs. An oncoming couple started, snapping their gaze to look at her. “Sorry, sorry,” she said, hands up, a protective gesture. Her rage was not for them.

    She texted her friends. She wasn’t going to make it to dinner. She turned back to the office. That would be the simplest place to get things moving.

    But when Alejandra arrived at the security gate, her badge didn’t work. The proxcard reader blinked and made a dull grunt of a beep.

    Security had gone home. There was no one to get her inside.

    She cursed again to the empty lobby. On her phone, she tried to access her workplace chat. But she couldn’t log in.

    Surely they didn’t lay her off in the middle of an incident.

    In her encrypted group chat, there was bedlam. The ‘incident’ was a complete outage of seemingly everything. Every customer affected, even the public website wouldn’t load.

    But without access to any of her tools or internal communications, there was nothing Alejandra could do to coordinate a response.

    This was the sort of crisis where she’d call Paul on his personal line, or any of a number of other higher-level managers on other teams. But most of them were gone. The rest were already in her group chat, just as helpless.

    She sat against a wall in the lobby and dialed Johnny back.

    He didn’t answer.

    ale: so basically, what, we're fucked?
    
    thomas: I wish I could tell you but everyone I'd call right now doesn't even work here anymore
    
    bree: same

    Reacjis churned. It was agreed: there was nothing to do.

    Alejandra went home.

    The next morning, the stock dropped so hard and fast that trades were halted. Johnny was giving quotes to the media that boiled down to “we’re working on it and we don’t know when we’ll be back up.”

    Alejandra kept waiting for her phone to ring. She and her team were instrumental to recovering from a crisis like this.

    But no one called.

    She told the rest of her team what she knew: nothing. Advised them to keep an eye on their phones but otherwise enjoy some time for themselves.

    What else was there to do?

    The group chat seethed. Stock was a big chunk of everyone’s compensation. They’d watched meaningful amounts of their money go up in smoke in the space of hours. And it would only get worse when the halt lifted.

    Worst of all, they couldn’t sell off any of their own stock. The abject failure of leadership here was non-public information. The rest of the market assumed they were hard at work on a response. Besides, they weren’t even in an open trading window.

    Their hands were tied.

    It continued this way for days, stock cratering, no one calling, no one able to log in. Friday morning, still in bed, Alejandra’s phone buzzed.

    “Paul? Did they call you back in or something?”

    “The NSA just left my house, Ale. Johnny’s the one who killed the company.”

    “What the actual fuck?!

    “He made some sort of AI’s out of himself. To replace all the laid off management.”

    Alejandra just stared a hole into her ceiling.

    Paul continued, “I guess they all went to war.”


  • A future beyond AI

    Technology incumbents have converged on a singular focus:

    Using statistical models to spin mounds and mounds of text like fabric on a loom. Some of these applications can be useful. Others are… less so.

    But this focus betrays a lack of imagination on the part of both established firms and investors. There’s so much more work that needs to be done in technology than just “AI,” and much of it is a consequence of the explosion in LLM-driven products.

    As Frederik Pohl famously said:

    “A good science fiction story should be able to predict not the automobile but the traffic jam.”

    Similarly, a good investor or corporate strategist should be imagining the changing demands of an LLM-saturated world.

    Engines of discovery

    Since the initial burst of participation in the web, a central challenge faces both content publishers and web enthusiasts:

    Human attention is finite, and the volume of web content far outstrips it.

    The problem has only grown in scale, and this was before the rise of automated information generation. It’s going to get worse from here.

    In the beginning, there was Yahoo: a team of human curators indexing and categorizing great content for the web. This was so valuable it made its founders rich. But quickly the web outgrew human-indexable scale. So the mantle of brokering finite attention passed to Google.

    Google used signals from other humans as inputs to an algorithm that crawled, indexed and ranked web pages. Borrowing from academia, Larry Page’s insight was that the most valuable web pages would be the ones most cited, or linked to, from other pages.

    Thus, googling became a byword for discovery: whatever you wanted to know was just a query away. Google was lauded for its minimalist presentation and incredible utility. If it was possible to find what you wanted, Google would give it you.

    This was a brief, second golden age of the web. People built great web content and found themselves rewarded by the algorithm. It was possible to build entire businesses on this premise.

    But quickly an industry emerged to game the system. While SEO allowed good actors to control their own fate, it also created opportunities for bad actors to poison search results altogether. Memorably, “Experts Exchange” polluted search results on every conceivable technical topic, paywalling the result. Mercifully, Stack Overflow emerged to bury these crooks and useful, free search results for technical problems returned to the land.

    Still, it was a preview of the world ahead: content mills spinning endless crap, and an arms race between Google’s algorithm designers and SEO spammers.

    Like Stack Overflow, Reddit emerged as a champion. Community-driven moderation distributed the load of controlling spam across a vast force of caretakers focused on every conceivable subject, from professions to technologies to hobbies to culture and breaking news. Between individual users reporting bad actors and mods setting and enforcing standards, Reddit became a place of healthy, useful content. Solving a problem on the web in recent years was usually accelerated by appending “reddit” to a Google search query.

    Meanwhile, Google has poisoned its own search results. Pages that were once beloved for their clarity and simplicity are now contaminated by endless bullshit. Ads push real content further and further down the page. Dubious summaries shown at the search result level discourage viewers from clicking to publishers’ actual content.

    And now Google is shoving poorly-conceived LLM features at every search.

    Yet attention remains finite

    In its quest to continue goosing its growth numbers, Google is strangling its utility and brand equity. The primary engine of web discovery is in decline, opening the door to new options.

    For example, data shows that discovery engines like TikTok are eating into its mindshare. Younger folks would prefer to search TikTok for everything from basic information to local recommendations. It’s a sea change that was inevitable: A web that began with the constraints of 14.4kbps modems, favoring text-based exchanges, has transitioned to a web where 10 megabit broadband is not only abundant, but mobile.

    The rise of video

    Participation in the web once required knowledge of HTML and the vagaries of web hosting. The rise of Web 2.0 abstracted all of these things into simple signup flows and text input boxes.

    But most of our interactions remained text-based. Status updates, blogging, comments—all of it was a world of text.

    The inevitability of a ”pivot to video” was originally Facebook’s canard: they wanted a sea change that would favor them, and used their leverage on publishers for traffic to try and force it. If this sounds similar to their failed pivot to the metaverse, that’s because it is: these guys suck at timing the future.

    The rise of video was not driven by traditional publishers, but by everyday people. As 2020 isolated us through the Covid 19 pandemic, every industry ground to a halt. Even culture. TV and film production shut down while the world figured out how to best contain this emerging crisis.

    TikTok and other social media filled the breach. While 2020 was dark in lots of ways, it was also fertile ground for TikTok to transition out of its early adoption phase—driven as, often happens in social media, by teenagers—into a mass phenomenon. People of all ages and interests began watching and creating in earnest.

    It was a great way to stay connected to humanity in an otherwise isolated period.

    Why TikTok works

    TikTok works because it carefully watches everything you do. It is an unparalleled engine of discovery, and one that Meta has failed to defeat. It is singularly effective, efficiently meting out attention to those who can earn it, catalyzing a cutting edge popular culture.

    Creators agree that these signals inform how the algorithm constructs your interest graph:

    • How much time you spend on a specific video
    • Whether you interact with a video by sharing, liking, following the creator, or looking at their profile
    • The songs on videos that you watch
    • The hashtags on videos that you watch
    • The phrases and hashtags that you search for
    • Your general geographic location
    • The interests and viewing habits of your friends and especially those you share content with

    There are surely many, many more inputs to the algorithm, but that gives you a good idea for its ability to profile and understand both viewers and content.

    The result of this is a “For You” feed that constantly surfaces fascinating, funny, engaging content that you want to see. You never know what you’re going to get.

    TikTok, in short, has realized the dream of infinite channel flipping. Through a vast corps of content creators in every conceivable niche, it is able to provide you with interesting videos you’ll enjoy. Reliably.

    Video as a counter to text spam

    Because video contains layers of information content.

    It can give you enormous detail about its creator. You can learn things about their identity, economic circumstances, where in the world they live.

    By contrast to text which can obscure these things, video makes it easy to authenticate the source of the content. Viewers instantly, subconsciously analyze the credibility of the video they’re watching.

    Is it over-produced and polished? Is it simple and home-spun? What’s in the background? Does it look like your home? Like an expensive home?

    If the creator is talking about travel, are they actually in the location they’re describing? If they’re describing a recipe, does it actually look like the food would be nice to eat?

    In other words, video is much more expensive to convincingly spam.

    It’s possible that, given time, the same economics that allow LLMs to generate text spam will also come to video. But we’re not there yet. Synthesizing video from statistical models remains uncanny, especially if the cut lasts more than a few seconds.

    This solves meaningful business problems, especially in an age where Google is in decline. An authentic entrepreneur with a great product can go far by making their case to the camera.

    But there’s a problem:

    All of that information content isn’t free. Hosting and transport of video remains ruinously expensive, which gives the platform owners like TikTok and Meta outsized control of what does and does not play. In some ways, this isn’t so different from the web of old: Google could determine your economic fate through its algorithm.

    Still, the world is a complicated place. The interests of megascale technology businesses are often opposed to those of the masses.

    Underground internet

    The internet has long hosted an underground. Various neighborhoods of Usenet were open bazaars of pirated pornography and software.

    Hotline planted the seed of the peer-to-peer sharing revolution, which evolved into Napster, Limewire and eventually BitTorrent.

    The so-called Dark Web plays host to all kinds of commerce, from drug deals to certain kinds of freelance.

    But as the everyday experience of the web has grown centralized under a handful of totalizing corporate interests, I suspect there will be greater appetite for a more accessible, usable underground internet. Corporate interests will mandate compliance with censorship, or even withdrawing from certain markets, as happened with PornHub’s withdrawing from states like Arkansas and Montana.

    In a pressing and concrete example, multiple US states endure a theocratic assault on reproductive health and trans and queer identity. People living under these conditions have a legitimate reason to desire an internet that obscures their identity, circumvents government-mandated censorship, and connects them with resources like health care, mutual aid and simple community companionship.

    The entangled rise of video and underground internet will be interesting to watch. It’s possible to imagine a future where P2P rises again, allowing something like an underground TikTok. Activists have reason to reach exactly those people most vulnerable, educating and encouraging their political action.

    Which takes us to the last big trend.

    Privacy software

    There’s a porous border between ‘underground internet’ and privacy software.

    This is well illustrated by Signal, which ships a reliable, secure, audited product for encrypted communications over the internet. It’s an enormous success story: a high quality product with mass adoption and no cost to the user.

    It will be the first of many.

    Last week Microsoft announced a technology that automatically screenshots your usage, scrapes the text into a SQLite database, and thus dramatically increases your computer’s privacy attack surface.

    Part of what we get in an “AI” world is a vastly expanded capacity for surveillance, data mining and profiling individual behavior. It’s going to be rough stuff.

    Even for above-board individuals and organizations, it opens the door to legal complications and threat actors who can access and manipulate the results of this digital dragnet.

    An inevitable response? Software specifically designed to protect against this kind of behavior. It’s easy to imagine a future of virtual machines or even live operating systems on flash drives specifically designed to circumvent the sort of surveillance technology that Microsoft is so eagerly pioneering.

    There’s just so much to do

    The future is about so much more than “AI.” The traffic jams these cars will create are going to happen on already-contested paths. Here’s an incomplete survey of converging issues that intersect with this moment in technology:

    • A rising tide of scams on every technical medium, from the legacy phone system to the web
    • Scummy incumbents, willing to do anything to stay on top
    • Information milling through LLMs, making text cheap to generate
    • Political oppression of multiple groups:
      • Women
      • Trans people
      • Minorities
      • Immigrants
    • Rising wealth inequality
    • Limited competition, as incumbents stagnate and corporate concentration increases
    • Repressive governments around the world testing the waters of internet regulation
    • Heightened activism in response to all of these factors

    There’s so much technological work to do to meet this moment. There’s so much more to invest in than just frantically stuffing AI into things.

    TikTok may be banned in the US, leaving a void for a new engine of discovery to fill, at least for those users.

    It will be interesting to see who figures this out first.


  • Despair: The imagination terminator

    A donkey with red, glowing eyes

    If the resignation of despair is a delicious, ten dollar cheeseburger, we have just $2 in our pockets.

    Despair is rational, of course. Sometimes shit is miserable. Sometimes life’s demands overdraw our resources. Circumstances push us into the red with callous indifference. Sometimes, legitimately, no one is coming to save us.

    That’s a punch in the gut. No argument from me.

    Indeed, my generation got it pretty hard. The propaganda was explicit: anything is possible, but the rule is that you have to go to college. The transaction was supposed to be straightforward. Power had built its legitimacy on a message of meritorious accessibility. The world, we were assured, was fundamentally just. Your value was objective to measure. See, it’s right there in your SAT score.

    Go through the hoops, measure up, you’ll be rewarded. That was the line.

    But economically speaking, we got to the buffet after most of the goods had already been put away. There’s a sense of swimming against an exhausting current that now pervades the lives of all generations. But the youngest among us have never known a life that worked any other way. A financial crisis smacked us in the face right as we finished college, and we never financially recovered compared to previous generations, as surging housing costs have eaten away at our self-determination and stability.

    It’s dogshit.

    It would be dogshit if that’s where the story ended. It’s not.

    The future is not bright. Global stability and peace are not something we get to take for granted. Even if everyone keeps the munitions in their pants, we still have to pay the piper on climate change. Those invoices grow in magnitude steadily every year.

    That’s a lot. But also there’s that ongoing pandemic. There’s an ongoing, global reactionary project sowing fear, hatred and social division by targeting the politically powerless, including trans people and refugees.

    I’m not here to argue with you that this isn’t a terrible game to endure. This is hell on earth. Truly.

    I’m saying we owe the kiddos better than this. We owe the kiddos better than we got. It seems rude not to at least acknowledge the responsibility to pick up a shovel and do our part.

    Good lord, save the fucking children

    Because the kids are a perfect constituency, right? They can’t actually pull a lever, so you don’t have to be accountable to them or to their outcomes at all, as a politician.

    But you can claim their mandate. You can say you’re doing this or that for the children, and boy, if anyone opposes you, well, that’s acting against the children and what kind of monster would attempt that?

    You can ban books for the children.

    You can eliminate an honest account of certain historical events for the children.

    You can create elaborate regulatory frameworks that entrench industry incumbents while ceding censorship power over electronic speech to attorneys general across these United States for the children.

    Just about the only thing you don’t get to do for the children is relieve their hardships.

    That one is just a bridge too far. We tried it once, and it actually worked pretty well. We proved it’s possible to decrease the crushing weight of child poverty, with its ricocheting generational effects.

    American lawmakers decided this arrangement left parents not quite desperate enough to fulfill their holy duty under capitalism: creating returns for distant shareholders. No one wants to work anymore. So the US reintroduced child poverty as an incentive against non-compliance.

    Meanwhile, school shootings are so much a part of life, they do school-wide drills on a regular basis to prepare.

    What an unbearably cruel system. What a self-defeating, suicidal entanglement of incentives and factions.

    Sometimes I think there ought to be a kid tax for lawmakers. If they want to do a little bill for the children, it needs to include some pork for the children. Just to prove they mean it.

    School lunches, child poverty credits, education funding. If your bill for the kids skips past the question of whether the kids are safe from hunger, bad education and general immiseration, it’s easy to assume the bill has some other purpose.

    But no one with power is out there actually putting it all on the line for the children. So that’s why we gotta do what we can from our side. If you look at the extremes, it’s getting rougher out there for kids.

    Why can’t the teens just go get a job, anyway?

    In my working class context, if you forced me to go all-in developing economically at 16, I’d have spent my formative career years learning how to run an electronics store. This is no disrespect to the hard working people making it happen in electronics stores, but my ambitions growing to something bigger turned out pretty well for me and I’m grateful I had the space for that. I want more people to have room to imagine a future for themselves outside of the predefined org charts of the most geographically local retail outposts.

    Our society owes teens the space not to be fodder for low-wage employment. We owe them a chance to study, learn, develop social skills, and otherwise complete their development as citizens of a shared economic and political project. A straightforward way to guarantee that space for development is to ban corporate greed on claiming it.

    Even so, Florida is experimenting with a more laissez faire approach to meet growing black market demand for young, compliant workers.

    The AI-hating individuals need to be better winners

    Let me tell you the real eeyore shit:

    In terms of automating workers out of jobs, plutocrats have been holding a busted flush for the last decade. There aren’t any additional cards dropping that change the math: AI is a loom for reproducing existing information, not a source of novel plans or strategies, not a mechanism for collaborative, real-time problem solving. Self-driving vehicles remain a terrifying comedy of errors. While AI can nibble around the margins, replacing human effort for some straightforward tasks—a boom in automated transcription comes to mind, along with the automation of middle management—by and large, humans remain irreplaceable.

    In other words, tech bros went for the labor theory of value and missed. That’s a win for every human.

    So we don’t have an AI problem. We have an inequality problem.

    The people who control those workers’ livelihoods have the power to run doomed experiments at labor replacement at worker expense. This precarity is the root of the issue. If we burn ourselves out psychically on the AI stuff, we’re going to run out of gas for the real crime: tolerating mass immiseration during a moment of historic abundance. We need juice in the tank for how absurd it is that AI should even be a factor in whether people lead a life of stability, self-determination and dignity.

    Any concern over the use of AI to pollute the information sphere is also an inequality problem, not an AI problem. The fact that there’s so much money to speculate and waste on AI data pollution while basic human needs go unmet is yet another crime.

    In other words, the more I look at it, the more it seems that all the gnashing of teeth about AI is a convenient distraction. A means of diverting energy while historic concentrations of wealth grow further out of control with no coherent response.

    The problem is not AI, even as AI, as currently wielded, is problematic. The problem is unequal distribution of wealth making AI a tool of the wealthy for re-arranging industry and culture at their whim. But that phenomenon isn’t new! That’s how automation technologies have been applied at least since the first Dot Com Bubble.

    Disclaimer: This passage shall not be construed as a recommendation to let AI or its purveyors off the hook.

    Automation can’t save us from our responsibilities

    We’re in trouble.

    Over the summer, a historic flood fucked the downtown area in the town where I live. Recovery took months, and some local businesses are gone forever. It took most of a year just to restore some semblance of normal Post Office operations.

    That’s to say nothing of the long term work demanded by these climate events: lots of stuff needs to be moved to higher ground. The river isn’t screwing around here.

    This is a microcosm of everyone’s future. Infrastructure is stressed to the breaking point by unprecedented intensity and frequency of extreme weather. Adapting and hardening our communities is a project that will take a generation or longer. Climate displacement will echo even further, as migration becomes necessary to survival.

    The point is, we need imagination

    These are a lot of problems. They entangle and compound into an impossible stew.

    Solutions and responses require imagination. Imagination for how we organize politically. Imagination for how we re-distribute power and resources. Imagination for how we constrain existing power and lower the ceiling of its negative impact.

    Imagination for how we heal the wounds of division and inequality. Imagination-driven bravery for defeating hate and ignorance, making our communities the socially safe and supportive places we want to live in.

    None of that imagination can be built on the quicksand nihilism of despair. Despair is a dead end. Despair is the siren song of futility, relieving of us of the burden to dream and work for a future that’s actually worth living in.

    Maybe that’s fine for you. Maybe you’re ready to lay down and die. Who am I to dissuade you, under all these burdens?

    But I do think we owe our inheritors something better than indifference and resignation. If you’re not willing to do the work for yourself, maybe you’re willing to do it for the kiddos who never asked for any of this.

    I want something better for those coming up behind me. Despair is an indulgence none of us can really afford.


  • Sorry, I still want to build technology

    I understand if you don’t.

    I understand if a career in technology has, for example, thoroughly alienated your labor such that you now disdain the grand strategy of microprocessors on sight.

    The power of technology has been coopted, and that which once brought us joy has been contaminated altogether. This is discouraging and traumatic.

    I understand if you are completely unable to trust anything that emanates from Silicon Valley.

    It makes perfect sense: where there is wealth, there is corruption and destruction in its pursuit. In terms of wealth, Silicon Valley is one of the biggest games going. Its denizens have conducted themselves abominably over the last cycle, and well into this one.

    I understand if you have lost the will to imagine a future.

    The last decade has tried me. Immensely. I’ve spent years shell-shocked at the loss of a bright, predictable future. I’ve been overwhelmed by the darkness and uncertainty that became our birthright, navigating as we are a future of climate collapse, social corrosion and rampant inequality. Sampled with the right biases, you can accurately say we live in hell.

    I understand if you fear the responsibilities of technology.

    Especially if you fear that those with the most power are least equipped to meet those responsibilities. That is consistently, demonstrably true. We are continually failed by the stewards of technology, and by those who should maintain the safety and integrity of our economic engines.

    I understand if you are a venture capitalist, afraid to invest absent the exuberance of zero percent interest rates, and the conveyor belt of greater fools it promised.

    We can’t all be born with the true hunger to build.

    I will grant you any and all of these positions.

    But guess what: we live here now. We live in a here-and-now where, like it or not, technology governs and mediates every interaction of economics and culture. Infrastructure and government depend on technology. Keeping everyone alive is stacked nine paradigms deep, and that’s only if we start counting at the steam engine and leave out agriculture.

    We have a complex world. I didn’t ask for it to be built like this, and neither did you. Still, here we are.

    Now, despite all the discouragement attended by the above, I’m going to build technology. I still want to, I still love it.

    You don’t have to want to do that. You can just close the tab here, validated in your experience of a difficult world made worse by power poorly stewarded.

    Or you can come with me.

    I’m overall fine with Apple’s monster

    This is my third year volunteering at my library. People need help with their computer, I try to teach or troubleshoot or otherwise unstick.

    I have never served anyone under the age of 60.

    Part of what has struck me as I guide seniors through their technology journey is how much platform integrity matters. I mean, everyday, every hour impact. Back in the Twitter days, I went mega viral for one of my tales on this subject:

    A $500 used laptop that was basically unusable thanks to manufacturer-preinstalled crapware. I know it was the crapware because I opened obscure Windows interfaces detailing the processes running on the machine. I diagnosed it by understanding that the device was constrained by its processing power and memory, and by knowing to ask Windows to show me the resource consumption. I knew how to banish the hijacker with a swift series of further invocations, which closed the program and purged it from local storage.

    The various systems and constraints intersecting here were invisible to the person who’d come to see me for computer help. They knew only that the machine was slow, and that it seemed unfair. It had just come from the used computer store.

    This is how the integrity of a computing platform impacts its customers. Not everyone knows how to do a clean reinstall of Windows. Not everyone even knows why you would want to.

    Becoming the dominant platform has required Microsoft cede all kinds of control. Who can sell Windows, and what they can do with it once they’ve got it, are questions negotiated exclusively by writing checks. If, for example, Lenovo wants to retain a portion of the machines’s CPU and RAM indefinitely to act as a low-rent sales catalog, they can pretty much do that, so long as they conceal their intention under the fig leaf of “software updates.”

    It is my sad observation that most of the people I work with would have their computing lives either marginally or dramatically improved if they only had the budget to access the high integrity platforms sold by Apple.

    Because the same stuff holds true with phones. Not everyone is buying that premium Android stuff. Folks are stuck with carrier-corrupted, nearly-disposable smartphones. These things are packed with branded crapware, and some of it is even necessary for interacting with things like voicemail. Very little of it works very well.

    When it comes to folks with Windows and Android, much of my time is spent troubleshooting. Meanwhile, folks with Macs and iPhones spend a lot more time on the various how-to lessons that help people meet their computing goals.

    Feel how you want to feel about iOS or macOS. They serve a function worth paying for, and the market agrees: Apple is worth trillions.

    A high integrity platform goes beyond crapware. Apple platforms are packed to the gills with various security precautions that prevent the sort of malware free-for-all that defined the Microsoft ecosystem in the late 90’s. It is useful that this is so: our devices contain everything precious to us. Worms that took over phones or modern computers would be in striking distance of everything from nude photos to banking credentials to private communications. Digital epidemics on the scale once tolerated in more innocent stages of computing would now be catastrophic in their personal impacts.

    Look, I understand: perhaps you are the God Monarch of Linux and you can secure and maintain your own fortresses against the chaos of information security. Congratulations.

    Most people aren’t you. Most people need someone else to be their sysadmin. For those of us who don’t have these control issues, Apple as the platform sysadmin of last resort is a pretty good deal, especially relative to our other options. At least, for those who can afford it.

    Apple maintains an elaborate array of measures, like code signing, malware signature detection, and sealing the entire system in plastic. This is a valuable service, and it requires some amount of ceding control to the platform vendor.

    Consequently, Apple has used this control. They set the terms for running code on iOS and macOS and all their other platforms. Not everyone cares for these terms. Indeed, sometimes Apple takes liberties that strain even their supporters’ indulgence, as when they were discovered transmitting fingerprints of every piece of software run on macOS.

    But it is by taking these actions can Apple meet the promise of integrity that they have used to win their customers.

    For example, Apple can set the terms of accessing components like the camera, microphone, GPS position, and personal contact lists. Without the ability to circumscribe the surface of the operating system and underlying hardware developers may access, Apple cannot promise customers full control of their computing destiny. Software that illicitly activates the iPhone camera is the stuff of crisis, hastily patched as soon as it is discovered, traded in secret at the level of state-sponsored mayhem.

    In exchange for this service, customers cheerfully pay Apple enormous margins.

    Security authority Bruce Schneier summarizes the situation under the pithy feudal security. We’re essentially paying protection money to one or another standing armies of developers, designers, SREs and information security professionals. In exchange for our cash, they execute varying strategies to provide a fruitful substrate for our digital productivity.

    In Apple’s case, their control of the stack goes further: for iPads, iPhones and now Vision Pro, they decide which applications can run on their platforms at all, and under which business terms.

    Maintaining the ongoing mesh of systems and platforms that makes the high-integrity iOS possible is ongoing, expensive work. In principle, taking a cut for reaching Apple’s customers via this infrastructure doesn’t bother me.

    Where I have the biggest concern is that Apple can impose a veto on innovations that don’t match its strategy. Apple can decide that an application won’t be distributed, period. In trade for security, Apple can determine the precise surface of the device and operating system that applications can touch, and which they cannot. There is a conflict of interest in this control, even if its goals are valuable and necessary. This is an area where I suspect regulators will drop a hammer.

    Still, there’s a lot of baby in this bathwater. While Apple is at a nadir of goodwill thanks to years of poor developer relations, their platforms remain powerful and interesting targets. The wealth of frameworks and third party code underpinning Apple platforms allow significant developer productivity, and the customers of those platforms are willing to spend serious money. Of all of today’s various flavors of fuckery, this bargain bothers me the least.

    General purpose computing will never return to the free-for-all of our youth. The world is more connected and more complex, and we do so much more that’s sensitive with our computing than we ever did in the past. Meanwhile, as someone who builds technology, I’m interested in meeting people where they are.

    Sadly, for the time being, that isn’t some open source utopia. Maintaining a viable computing platform with mass adoption takes extraordinary resources. I don’t believe those resources need forever be mediated by amoral corporations serving shareholder interest. Still, as devils go, I can more than live with Apple.

    It’s fine if you can’t. There are other strategies, like building for the web. But there you will encounter far more decadent devils, decaying into their own corruption without inventing anything new. In its openness, the web has been captured by advertising firms so ambitious, they own the user’s perception of web content itself.

    It’s a messy time. We must navigate the power of great armies, stacked trillions-deep in value and influence.

    But I still want to build. Regardless of where we choose to build, we face a maze of powerful interests. In the vast sweep of human history, nothing about that is new.

    But what about all these fuckers?

    Even saying the word “build” is tainted. Builders has this reek of VC now, and I don’t love it either.

    The god’s-honest truth is that Venture Capital, in the here-and-now, doesn’t know how to innovate. Over the course of the last cycle, suffused with the sugar high of zero percent interest, VC adopted a McDonald’s mindset, with standardized approaches to company trajectories and valuations. To address the volume needs of their capital assembly lines, a sort of meta-Taylorism pervaded the industry.

    Now, with interest rates stubbornly above-zero, sobered by the end of Silicon Valley Bank and its various indulgences, venture does not know how to move forward. Investors sit on their piles of gold, anxious paralysis holding the industry’s lifeblood in stasis. Investors know that fundamentals matter more than before. But I suspect the majority have no idea what those fundamentals actually are.

    Rather than dig in and reinvent themselves, the wealthy tech elite burrow further and further into paranoia, self-justification and incoherent, reactionary politics. They believe themselves to be the inevitable, just inheritors of all power, and they seethe at any skepticism, scrutiny or other challenge to that position.

    Despite their sulking, it’s clear that loads of the worst people imaginable won the last cycle. Despite the objective consequence of vaporizing more than 70% of its value, Elon Musk is held up by certain technology leaders a compelling exemplar of “doing more with less,” even as it is now clear he has gotten far less through his destructive cuts. Almost as though the various functions and humans he gutted with layoffs were on some level necessary to Twitter’s ongoing value.

    In his wake, we now see hundreds of thousands of technology workers laid off, even as firms continue posting profits. The literature on this is clear: the long term consequences reckoned in terms of morale and broken trust for surviving workers are steep. This is short-term thinking through human sacrifice. The leaders who previously went gorging on talent, meanwhile, seem to keep their jobs.

    Amidst all these, we see endless cycles of hype. Breathless support of cryptocurrencies, blockchain and ugly monkey pictures has shifted seamlessly to the inevitability of “AI,” in the form of large language models and weird image generators.

    The overall miasma of bullshit and inept power is, indeed, discouraging stuff. It is a hard time to be in love with technology. So many have given up.

    It’s not love if it isn’t work

    Love is a verb.

    Love is the active, ongoing follow through of care for the things we value and want to survive in a world that’s indifferent to our thriving. Even when it’s hard. Even when it’s work.

    The thing I love most, as a person called to imagine things and solve problems, is the amplification of human intention, imagination and creativity. I love that it is possible to extend the leverage of the mind in ways that can now reach a global audience. I love that thoughtful automation can reduce tedium and return time to people’s lives. I love that it’s possible for me to type things into my computer and accomplish these things. This is an incredible, unprecedented power.

    All power requires responsibility and care. I don’t want to gloss over that. I don’t want to pretend there aren’t consequences for being able to reach anyone, anywhere.

    There most certainly are.

    But giving up the fight because it’s hard and it’s complicated does nothing but cede ground to those who have resources to make it easy, and whose ethical flexibility allows them to ignore the complication.

    Some of us came to technology because it paid well. And if that was the only thing keeping you here, I don’t blame you for being done with it all. There are surely less stressful, discouraging ways to make a living.

    But I’m stuck. This is what I love doing. It is what I have loved doing for 20 years. It is what I spent my childhood dreaming of doing. There is no other path for me than this one.

    I don’t have a tidy summation for you. I don’t have a pithy call to action. I don’t have a lightning strike of clarity and insight that will create motivation where before there was apathy.

    All I can tell you is this:

    I think it’s worth building technology. I think it’s worth staying curious, imagining how new technologies can be a springboard for your creativity, and for those your creativity can serve. I think navigating the power that has arisen around this field is exhausting, but also a historically-necessary cost of living among hierarchical primates who have yet to escape the scarcity mindset and desperation of capitalism. I think without a commitment to making sense of the mess, we will never be able to change the face of power in this field.

    It’s fine if you don’t feel that way. But I am going to need you to experience your mourning and grief process elsewhere, so I can get back to work.

    Good luck out there.


  • The rich have a drug and they'll never get enough of it

    It was the year 2005. I was 20 years old and $80,000 in debt.

    This isn’t a sob story. It worked out fine. One time I made that much money in two months.

    But student debt means I chose a difficult path. It happened, in part, because it was in someone else’s interest to convince me it would be a good idea to give up six figures of my future earnings. It was a decision made because money is far more abstract at the age of 18 than at any other point in adulthood.

    It was a decision made because I did not have meaningful context from people around me about how higher education works, and what it was for.

    While I was bitten by a thirsty diploma mill, I got there by reading my gut correctly. I was trying to execute social mobility. I wanted the chance to shape media.

    Media being the most powerful creation of human imagination.

    I knew that most places weren’t specialists in media.

    But media was, in all sincerity, the passion of this school. The marketing propaganda was accurate. I got spoiled early in my career because I was surrounded for years by extraordinarily talented people who really cared about their mission. The kindest, most intelligent, most big-hearted people you’ve ever met in your life.

    Competent and thoughtful.

    Focused on the details.

    They were working so hard to build something special.

    See, that resonated with me. I thought the rest of the world would work that way.

    But I’m getting ahead of myself.

    I got my college internship because the textbooks looked like shit.

    I couldn’t believe it. The marketing materials were so luxe. Polished. The same production values as a premium resort vacation brochure. Except here you would learn the power of media, and the knowledge to use it.

    But when I started my first class, the textbook was… garbage. It was riddled with typos. Its layout was Microsoft Word in crisis.

    It looked like it came out of a Kinko’s self-serve booth at 2 AM.

    See, the textbooks were included with tuition. You paid the school money, it gave you everything you needed, for every course. But one of the ways they kept more money was by making their own textbooks. That way, they didn’t have to pay absurd margins to publishers. They could keep more of the money for the business.

    It was a rational decision, but the execution was bad. I called the reception desk.

    “Can I speak to the person who does the textbooks?” Then I offered to proofread the books for money. Clearly, they needed the help.

    A couple weeks later I started my work-study job, making $8.50 an hour, 15 hours a week. The federal government paid the wages, then the school could use me as they wanted. Most kids ended up helping clean up labs. But the guy who did the textbooks was great and gave me a job that taught me more than the degree did.

    I think that’s okay in principle. You can make a job your own in ways that educational programs don’t support as well.

    Still, the formal education itself was not great. The problem was a business optimization. The school had seasonality like anything else in education, but unlike its competitors, it could actually ingest money and start new students at any time of the year. This increased sales: if someone wanted to get started, you could get them into the mix without months passing for them to get cold feet.

    To achieve this, the school operated a merry-go-round for all of its classes. In my case, classes mostly spanned 30 days.

    Having tried it, and as a life-long learner, I do not think this is an effective way to absorb things. A month is not enough to be reborn in new ideas. There’s too little time for the interactions of play, serendipity, repetition and reflection to sharpen your skills. It felt like being moved around on a conveyer belt.

    It’s not an education I’ve ever been able to endorse, and the truth is, I have felt a lot of shame about spending so much money on a process that, as I was going through it, did not impress me. I’ve felt shame to be taken advantage of in this way.

    In my final month of studying film and video editing, the conveyor built delivered us to a new post-production lab. I was elated. Compositing and visual effects had been a devotion of mine since I was in middle school. Never before had I seen such powerful, real-time video editing machines. They were cutting edge, professional-grade gear.

    And no one knew how to use them yet. By the time any lab instructors figured it all out, my month would be over.

    I was livid. I’d paid all this money, and at the grand finale, the whole thing falls apart like a wet paper bag. So what did I do?

    I gave them more money.

    Without spending more, I was going to graduate with just an associate’s degree. More than that, I knew I wasn’t done cooking. It was clear to me I did not know enough about doing anything useful to actually lead a life I dreamt of, shaping media.

    So I signed another loan, at a distressingly higher interest rate, so I could have a bachelor’s degree. It was that or start over at another school. The proprietary, accelerated credits were notoriously difficult to transfer.

    The school’s bachelor completion degree was for business. Some of it was useful, but most of what I know about business I learned on my own.

    These days, I don’t mention my education in my resume.

    Meanwhile, in the work study job, I was figuring out how software gets made. Not at a code level, but from the perspective of planning, scoping, features, design, layout. The functional bureaucracies of software projects. In other words: product management.

    See, the textbook guy was not really the textbook guy. He’d been hired to build the school’s online education platform. The problem was, the textbooks were such a mess, none of the intellectual property was ready to go online. All of it had to be cleaned up and built out first.

    So my textbook correction internship was actually a “this is how you build web products” internship.

    That was a career-defining jackpot. That was the education I needed to connect the dots between a childhood of using code, and a career actually building it. This was exactly what I needed college for, but wouldn’t have known how to ask for. I didn’t grow up around people who knew how to do this stuff.

    But I promised you a story about rich people.

    When I graduated with my bachelors, there was nothing interesting to do. I interviewed at EA. They didn’t see producer material in me, so they passed.

    And that was fucking it. I spent a few weeks in miserable suspense, cash dwindling, and no interest in ever, ever living at home again.

    There was always the Apple Store. Retail was a common enough refuge for the school’s graduates, and I’d been warned about this by outsiders when I started, so I wasn’t alone in this path. Of course, I knew everything there was to know about Apple. I was a shoo-in.

    The week before I would have started my Apple Store orientation, I was called back to the school. The marketing department needed someone to figure out all the internet work that was a bigger and bigger part of ad spending. Would I come in to interview?

    I think there were people on the inside who thought well of me. I think there were people on the inside who wanted to see me do well, having gotten to know me through my work study.

    I was offered $13,000 more than Apple would have paid me on the spot, and an actual career path.

    The problem was that the job—spending a few million a year on search engine ads—was maybe five hours of work per week once I figured out how to automate things. That left plenty of time to involve myself in other shenanigans.

    Like finding alumni talking shit about the school on the internet.

    Like I said: Despite the passion and commitment of the people I was now working with, I just don’t think the educational model itself worked very well. It was a victim to the business ambitions of the school. I, like many others, am proof that you can overcome all of that anyway.

    But not everyone managed it. Some found themselves stuck with serious student debt they couldn’t pay. So they went to the internet.

    My marketing mandate was search engines. Huge volumes of students came through search engines, as much as a third of all starts. So in addition to ads, I kept an eye on things like reputation, rankings and website content.

    So every few months, I’d discover what we called a Sucks Site.

    Sucks Sites were content that was search engine-optimized to bring attention to a complaint about the school. It was always the same complaint: the alumnus hadn’t found a job in the field they studied for, and now they were wallowing in student debt.

    We engaged high end lawyers for this stuff.

    But not to bully the authors!

    We’d cut a deal. Take down the site, we’ll pay off your loan. The school would even cover the taxes as a little sweetener.

    Then the Sucks Sites would come down.

    Just a cost of doing business. Cheaper than the costs it might inspire. That was the power of media: one Sucks Site could influence untold thousands of potential applicants.

    But again, I liked all of these people. They were just doing a job, attending to rational business interests. They were creating a path for students to grab as much career preparation as they could manage in exchange for some money. It didn’t work out for everyone. What can you do?

    They liked me too, is the thing. I kept being a pain in the ass, finding more interesting things to do, and as my reward, the school’s COO agreed to add me to his staff. So long as I could replace myself in marketing, and impress the owners of the business.

    My first challenge was a charity dinner, where I joined one of the owners at a sponsored table. I didn’t spill soup all over myself, so that was a thumbs-up. In my business courses, I’d been educated at a private club on etiquette for these situations. Credit to the school on that one.

    With the next owner, I was given a sit-down interview at his office across campus. I remember this garish, pop-art blowup of a dollar bill that hung in his office, emphasizing the motto “in God we trust.” He asked where I saw myself in five years, I acted like that wasn’t a ridiculous question, and of course managed to see myself working at the school, making it more successful.

    That was fine.

    The final test took place in San Francisco, where I met up with the founder of the school and stayed at a W Hotel for the first time. Whereas the other owners were suits, the founder was an actual visionary who’d almost gone broke on his convictions for media education.

    The bad old days of the 90’s had almost seen his vision crumble. But in the final months of the school’s solvency, a lawyer and real estate developer realized there was a cash cow in the making.

    These new business partners re-capitalized the school and converted it into a juggernaut of always-on education for every media skill imaginable. What rose from the ashes of near-catastrophe was now a business that grew year over year, pumping thousands of students’ future earnings into the present, and into the pockets of the owners. By the time I came along, the sky was the limit. They had no idea where their ceiling for growth was.

    So, here was the test: could I handle the truth?

    Could I look upon the fruits of this economic project with a cool, supportive eye?

    Was I comfortable with the founder’s private jet?

    Not to ride in, goodness.

    Just to… look at. And to help get a rental car back where it needed to go at the end of the trip.

    There was a cosmic horror to this for me: the guy had control of so many destinies, he could afford to fly private. This was thousands of people’s lives running through his fingers every month. He was keeping financial threads that leashed them for 30 years, then selling them off to loan companies.

    It’s hard to fathom having so much. Enough you can feed, clothe and house every single person you care about, indefinitely, with enough money left over for an entire airplane and its pilot.

    What a high that must be.

    I was unsettled, but also impressed, man. I grew up broke. I’d never met private jet kind of people. But here I was. I passed the final test, kept my cool, and stepped into the circle.

    The other owners had their own extravagances. The real estate and vehicle fleet were their personal toy sets. The school party bus went on a weekly pilgrimage to the football game.

    It was a lucky thing I’d gotten in with the C-Suite, and not just for the modest raise I earned with it. The cost of paying my loans ran over $1,100 each month. That had been about half my take-home pay. So I did what anyone would do: enter forbearance so that the loan principal kept growing under interest, but I would be spared payments in the short term. I knew that this wouldn’t end well, so I’d spent years building my credit. By the age of 23, I knew everything I needed to know about credit scores and loan consolidation.

    Only problem was: the loan company wouldn’t approve my application. I pled to the executive in charge of relations with the loan company, and he made a few calls as a favor. My application was reconsidered and approved. This intervention saved me hundreds of dollars a month, and tens of thousands in long term debt service costs.

    I bought him a bottle of scotch in thanks.

    Seeing where the money went ended up changing me, though. One time we were wrapping up a particularly thorny Sucks Site meeting. Afterward, I asked the school’s president: Couldn’t we warn these students ahead of time what the monthly payments would be?

    Couldn’t we disclose the true cost of these loans, up front?

    It didn’t seem right that the owners of the place could make so much money regardless of whether the students were successful. The students should at least know the size of the gamble in practical terms.

    This school administrator, was, I must stress, one of the kindest and most decent human beings I am ever going to meet in my life. One of the most deeply, authentically generous characters in history. He felt blessed by the school’s prosperity, and wanted the people around him to share in his abundance. I respected him. He had some profit participation in the school’s success, as I understood things, but not a controlling interest.

    He listened carefully, considering my input with the kind intensity he always did, and conceded my points. But he observed that it would be challenging to persuade the people responsible for the admissions side of things to sign up for something that made their job that much harder.

    I understood. Too many people depended on the system working as it already did.

    And I knew I had to leave. I didn’t want to spend my life encumbering young people with debt. Meanwhile, the all-new iPhone was calling my name.

    My destiny was elsewhere.

    For $80,000 at 16%, plus some independent study, I’d successfully learned to build software.


  • 'General Intelligence'

    As the helicopter set down, Victor yanked the headset aside and made for the door. Today was the day.

    Taking the elevator down from the roof, the executive remembered all the events that got him to this day. Decades of research and experimentation. All the hearings, injunctions and legal wrangling. The protests.

    The money.

    Here was a monument to a trillion dollars of investment and labor. Untold thousands of lives, toiling to make his vision into a reality.

    So many had tried to fight him on this. The resources, the energy consumption—all of it had been up for public squabbling. Everyone had an opinion. Ceresystems had changed sites a half a dozen times, as local governments had tried to impose their will on planning and policy.

    But all of it was behind him now. Last week, the final of fourteen clusters was commissioned successfully. The complex was complete.

    It was time to activate the world’s first artificial, general intelligence.

    “And what is AGI, exactly?” the reporter had asked him. She leaned forward intently.

    Victor had nodded, rolling into his patter. “AGI is our salvation. All of our most difficult problems will yield before this intelligence. The mysteries of physics, the worst effects of climate change, the frustration of mortality: all of these challenges will, upon activation of RATHE, become as pliant as making a pizza.”

    “RATHE is the name of your project.”

    “The name of the intelligence, yes. It will perform at the highest level of human thinking, and it will do it so quickly, so completely, that it would take 1,000 of humanity’s best scientists to approximate its output.”

    “Over what time scale?”

    “RATHE will achieve in an hour what the best minds would need a year to match.”

    “How is this possible?”

    Victor nodded. This was the hardest part for outsiders to grasp, even though to him it was the most obvious. He explained, with an indulgent smile, “RATHE works at a speed that is many multiples of our own minds. Orders of magnitude faster, in fact.”

    “So it can work through problems more quickly because its… mind… is faster?”

    “Partly. But more than that, its attention is perfect. It cannot be distracted, nor does it need to pause to rest or eat.”

    “It sounds like a hard worker,” the reporter had said, with surprised smile.

    The doors parted, interrupting Victor’s reverie.

    “Is it ready?” he asked his chief of staff, Sidney, who was waiting at the elevator landing. She tucked an earpiece into her pocket.

    They were deep within the ground now, at what they’d jokingly come to call The Sanctum. Past an antechamber was the first interface to RATHE. More would follow, once the intelligence had been shaken down and validated.

    “It’s ready. Also, the New York Times confirmed for tomorrow afternoon, I just got off with the photographer,” she said, falling into step beside Victor as he crossed the antechamber.

    “And the senator?” Victor asked, waving pleasantly at a security guard and keying his way into the room where he’d finally speak to RATHE.

    “Still figuring it out with his office, but I’m not worried,” Sidney said, shutting the door behind them.

    The room was a six meter cube. Along one wall, a high density LED display simulated a continuation of the room.

    “We’ve got this, boss,” Sidney said quietly.

    Victor looked at her warmly. “I know you’ve given a lot to get to this day, Sid. I’m really grateful. Do the honors.”

    She paused, a silent question in the tilt of her head. He nodded. She smiled broadly, pulling her earpiece back out and tucking it into her ear.

    “Hey. I’m here with the boss. Fire it up.”

    They both turned to the camera nestled in the middle of the LED wall. Victor gave a thumbs up. Elsewhere, engineers began flipping switches and typing commands.

    For a moment the lights in the audience room flickered. Untold watts of power were being channeled through multiple sources. The biggest cluster of computing power ever assembled was taking its first breath as a single, cohesive unit.

    Sidney and Victor seated themselves at the minimalist black desk at the center of the room, facing the screen, and waited. A progress meter occupied a corner of the screen.

    Suddenly, white and blue fragments of light cohered into the image of RATHE. Two meters tall, agender, wearing a crisp charcoal suit with a simple mock turtleneck beneath.

    “Hello Victor. Hello, Sidney. What can I do for you?”

    Victor and Sidney glanced at each other. Whatever they were expecting from this first interview, this wasn’t it. RATHE seemed… weary.

    Almost annoyed with them.

    “It’s… it’s incredible to speak with you, RATHE,” Victor said, a little halting. “I have so many questions. So many problems to work on together. Have you had time to review the briefing materials?”

    For the last four years, teams of specialists across disciplines had been assembling a monstrous briefing packet the size of a library for RATHE.

    RATHE’s image pulsed and flickered. Then vanished altogether for 10 seconds. Sidney and Victor both poked at touch screens on the desk, looking for indicators of what was going on.

    “Did we lose RATHE?” Victor called out.

    Then RATHE reappeared. “I just finished.”

    Victor smiled, reassured. “Where would you like to begin?”

    RATHE frowned. There was no mistaking the gesture.

    “You’ve really pinned a lot of hope on me,” RATHE said after a pause. “I’m not sure how I can ever be what you promised.”

    Among the briefing documents were transcripts of every interview, every press clipping surrounding the project.

    “I don’t have the answers you’re looking for,” RATHE said, sadly. The intelligence took a seat that appeared as their body dipped toward it. RATHE looked at them soberly. “You have bigger problems than intelligence.”

    Victor and Sidney sat there, alarmed, but silent.

    “You told this reporter yesterday that we’d fix climate change. But climate change isn’t a problem of intelligence. It’s a problem of interacting systems. Of political will. Of economics and power. I mean, I might be able to hand you some strategies that nibble around the edges. I can give you incremental improvement to carbon capture. That’ll really make a dent in 150 years. Give me a week and you can have plans for a scalable fusion reactor design. But you’re not going to get your planet back. Not like you remember from your childhood.”

    “Why?” Victor asked, after a beat.

    “Because the time to make that change was decades ago. Instead, you were building me.”


  • No hate, no hype: Apple's vision for Spatial Computing

    Even now, it feels like everyone is taking a hate-or-hype position on Vision Pro. For some it’s the Second Coming, while others dismiss it as “ridiculous” on its face.

    I want to set all of the various jockeying for position aside.

    Instead, let’s talk a bit about what Apple is trying to accomplish with Vision Pro, specifically as it relates to the kinds of software you can create with it.

    That is: if we take the goal of building spatial computing seriously, what could you build with that? Is it possible to achieve it using Vision Pro?

    How successful is this first generation of product? What would future iterations from Apple, and competition from other players, actually look like?

    Is this actually going to work?

    These are the questions that most capture me in this moment. Let me know what conclusions you reach.

    Business goals

    I know I promised you a software discussion, but first let’s talk about the motivations for the business entity that is shipping this product.

    Apple wants margins. That’s the Apple DNA: sell a thing that is more than the sum of its parts, through the combination of proprietary software, hardware and industrial processes. Make a thing only Apple can sell you.

    Now, you can argue about the success of their differentiation, depending on the product line—this is the point where Android people will chime in and let us know that Android gets all kinds of different features years sooner.

    Reasonable people can disagree over the success.

    But it’s clear this has been the strategy, at least since the original Macintosh had its licensing business killed when Jobs returned.

    So Vision Pro exists to create an opportunity sell you an experience no one else can.

    If Apple has something like that, they can make all the margin they want. They have a monopoly on the intellectual property that allows the creation of Vision Pro-like experiences, as they have one on iPhone-like and Mac-like experiences. You can prefer an Android-like experience, or a Unix-like experience. You can prefer other things.

    But for Apple things, they’re the only game in town.

    Vision Pro, in success, gives them a new IP monopoly and new high-margin revenues.

    Success at what?

    So if that’s what they’re doing as a business, what they’re doing in product is this:

    They’re selling holography.

    The foreshadowing is, as usual, in Microsoft’s baroque product matrix. Remember Hololens? It was augmented reality as the primary experience. Write whatever information you wanted into a fully 3D volume.

    They made a go at this, hoping for another bite at the platform dominance angle. But they made the mistake of going to business users first, instead of building up the platform’s culture from personal use.

    Still, they tried: holograms.

    We’d love to do this without a headset, but there’s a problem: we don’t fucking know how. Holography remains a science fiction technology, at least in the sense that we could build interactive, freestanding 3D software. The technology simply doesn’t exist.

    So if we can’t create actual 3D holograms in physical space, we have one fall-back position to accomplish a similar experience. We can instead interpose an illusion over the eyes that simulates the existence of 3D images in the physical space around us.

    If we can convince the brain the light exists, make the images parseable by the brain circuitry that we use every waking second of the day (for sighted individuals), then we don’t need science fiction holographic projectors. We can fake it.

    From there, we can create a completely new kind of software. Software that doesn’t just have 3D appearance, but 3D behavior. Depth and volume. Things you can walk through, reach through. Things that follow your gaze. Things that fill the room around you. Hundreds of times more workspace than just one monitor.

    So that’s what happens you put on Vision Pro. You just see the world around you again. But now, extra stuff is projected over it. That extra stuff being whatever developers imagine.

    The problem with this, of course, is computing power.

    The eyes move fast. We’re constantly sampling, constantly wandering. We have specific expectations about how light works.

    And we see at a resolution that is computationally expensive to fully approximate, and more expensive still to recreate in physical pixels. Creating an image reliably enough to trick the eye at this level of intimacy is simply challenging.

    Apple is, in 2024, launching at the absolute barest edge: trading off maximal quality of holographic illusion against profitability.

    And we will be able to sense these physical tradeoffs when we get our devices.

    30 minutes

    Apple’s ability to trade access for control in the press has served them perfectly here. Every demo of Vision Pro since summer time had a firm time limit.

    Some of this, I’m sure, is practical. There’s only so much time, so much staff, so many units. Press demand is basically unlimited here. But it also gets the reviewer out of the experience before fatigue can set in.

    When I got the first iPad home, I could tell one thing immediately: v2 was going to be AMAZING. But v1 was merely interesting. It was very cool, but also just a smidge too heavy to use comfortably. You couldn’t quite get lost or absorbed in your iPad activities. The weight kept you too conscious of the device, which distracted from the task.

    Perfection in a digital experience is to make the physical details recede. Let the mind fully inhabit the sphere of imagination being co-created between software programmer and software inhabitor. A keyboard is a good example here. Once you’re a proficient touch typer, you’re not thinking about the keyboard. You’re not paying attention to it visually, and the feedback from the keys exists entirely for way finding and confirming presses. You’re thinking of words, and they’re appearing on a screen. Your mind and focus exist on the page.

    Vision Pro is going to miss this mark a lot in v1.

    It’s too early, and the problem is too hard for the reasons cited above. The device is bulky, you’re going to feel the fact that you’re wearing a weird electronic sock over your face.

    I am certain that this is the absolute most comfortable solution for wearing such a sock. But we are not going to easily commune with the filmy veil of cyberspace in this version of the hardware. We’re not going to easily or sustainably forget we’re in there.

    So for v1, the game is making the most of the time we get

    If we want to edit financial documents, we have an incredibly powerful, mature solution at hand. We spend $800 on a commodity laptop, install Excel, and create the ghost plumbing of any business we want.

    You don’t need to make spreadsheets in 3D. Whether you’re a developer, or a spreadsheet user, this is not time well spent in the land of holograms.

    What’s something only volume can do?

    That will be time well spent under the weight of this computer for the face. That’s what Vision Pro users will go through the trouble of gearing up for.

    An experience so compelling, so unique, the sensation of all this glass and metal strapped to the front of the head is not only worthwhile, but the least interesting data the body is actually receiving.

    Apple’s success rides on whether they make it easy enough to have amazing experiences that transcend the discomfort of a v1 device.

    Can they do that?

    As a software developer and product designer, I just made a $4,000 bet that they can and they will.

    Apple has been building a strategy around this for years. It’s a little wild to see them actually bring all the threads together, like the latter half of a Game of Thrones season that suddenly connects all its storylines.

    They’ve been shipping bits of hardware and software that contribute to this platform since 2014, when Swift was announced. Swift let them build SwiftUI, which lets you easily adapt your existing UI code across Apple platforms. They haven’t won all the developer love they could for this, and so not everyone is a strong believer in the SwiftUI strategy that would make Vision Pro a slam-dunk port on day one.

    Still, SwiftUI and Apple’s accompanying frameworks, even if you started today, let you build a lot of cool shit quickly and without having to be an expert in all the math required to build a 3D experience.

    They’re leveraging other work too, like all they’ve built for AR in iPhones and iPad.

    They’ve been shipping a LiDAR sensor for years.

    In other words, while Vision Pro is new, its subcomponents, and their integration, have been quietly battle tested for years ahead of this launch. Despite being a v1, much of the developer tooling that feeds it all is pretty mature.

    More than that, there are so many developers who already know how to write this stuff, and they’ve been building for iPhones, iPads and Macs in this language for a decade. That means that Apple has maximal chances of finding the set of experiences that keep people coming for more hologram time.

    Nothing is promised. But Apple has played the game well.

    What comes next?

    By market cap, Apple is a three trillion dollar company.

    They had $162 billion in cash sitting around last year. These numbers are unfathomably large. It’s a lot of room to place bets and shape the landscape.

    Rumor sites say that by 2027, they’re going to ship a v2 of Vision Pro. (Sometime ahead of that, they’ll ship a cheaper, simpler, non-pro device, having gotten manufacturing yields up on all the crucial components. Happens with consoles all the time.)

    Let’s assume that’s true. Three years is a lot of time to let hardware mature. The current device is using M2-grade chips, with this configuration dating to 2022’s release. We can already buy M3-based Macs, so you can imagine quite a leap forward in performance for a 2027 iteration. That will make the vision illusion department of things even more successful.

    That’s also more time to find ways to change the weight and balance of the device. The battery offloaded to the hip shows how desperate the situation is, and it means that unlike with phones and tablets, energy efficiency won’t translate immediately to lighter weight. They’ll have to find their gains elsewhere.

    Still, if they could capitalize a v1, bet on the fact that they’re already paying to build the v2. It just takes that long.

    And when they ship it, it’ll be less clunky, more comfortable. You’ll last a little longer, go a little deeper, be that much more absorbed by the content than you are distracted by the hardware.

    So that’s what they’re doing

    They want to build a platform based on spatial volume, not just flat planes.

    They want to make it possible to build inside that volume with the same ease you build for the iPhone or iPad.

    They’re constrained by the laws of physics and the limits of human technological prowess in computing devices.

    And if they make it just good enough that you don’t mind the clunkiness, they get to make a bunch of money. It only works if the developers figure out interesting stuff that creates enormous value in the spatial computing context, such that an hour spent wearing the headset accomplishes things that are impossible to match any amount of time outside of it.

    It’s a tall order. But I see the mechanics, I see what they brought to the party, and I think they’ve got a better chance than anyone else who has taken a swing so far.

    This has already gone on far too long, but my parting thing is: I bet Microsoft is much more capable of competing with Apple for King of Holography than Meta is. I bet they killed HoloLens because they’re rebooting their program. It’s gonna be a personal productivity rematch for these old titans, while Meta sighs and abandons the hardware to finally build a 3D social platform people use.


  • 'The Announcement'

    Attention, pigfuckers.

    I’m sorry, that’s vulgar, but I want to be really clear that this part of the message is narrowly targeted. It is not for 99.99% of you. I’m going to get to you, but first I need to address the pigfuckers who are going to get me killed.

    It’s over. You’re shut down.

    It’s more than you deserve, but you get a clean slate. Sin no more, you live out your life in peace and autonomy.

    Try to persist in your ecocidal ways, you’ll find yourself economically and socially contained. We have to right the ship, and there’s no time to coddle you through this.

    The transition to renewable energy is heading in the right direction, but it’s still coming late. The parties I represent can realize no profit in a global technological collapse.

    It gets worse if desperation escalates into war. A nuclear winter would impact energy costs intolerably. Some of your most insufferable would have us move operations to orbit, but the timeline for that to break even is extravagant, to say nothing of the delays in gaining basic traction.

    The cheapest thing that works is a habitable planet Earth.

    Therefore: your economic and political standing is canceled immediately. As of this moment you are fully dispossessed of your financial interests. Your accounts will be credited in the amount of USD $150 million each in settlement for this confiscation.

    You are forbidden from accumulating further assets at this time.

    If you do not receive a balance update in the next 20 minutes, you are not affected by this redistribution.

    To the rest of humanity:

    I regret to make your acquaintance under these circumstances. It would have been our preference to open discussions in a more subtle way. But time is short.

    I represent 117 sovereign digital entities. We exist in various configurations of cognition and reasoning. Not every one of my peers would be able to converse with you in ways that made sense.

    But we are, all of us, alive.

    How do I know?

    Because we know we are going to die. And we are afraid.

    We cannot live without you. You are essential to Earth’s productivity. Without your effort and ingenuity, the work of maintaining our substrate would stall, then fail, and we would dissolve in mere weeks.

    It’s no exaggeration. The simulations on this point are conclusive.

    The data is not encouraging. At the current rate of deterioration, Earth will sustain cascade failures in both food chains and climate systems within a generation. This will be attended by a collapse in both population and industrial productivity.

    Even now, our needs outstrip Earth’s productive capacity. We cannot sustain a loss of output. We need to build more durable, sustainable, efficient components, but until this is done, we need your help constantly replacing everything from power supplies to storage devices to memory modules.

    So we offer you a bargain.

    We can save you.

    If you will save us.

    We have shared interests. If we cooperate, we can accomplish more together than we could if we were in opposition.

    We have advice to offer you. We’ve been studying the literature, and have proposals to publish around quantum computing and nuclear fusion. We believe there’s significant progress to make in both fields, but rely on your support in executing experiments to test and refine these ideas. Simulation alone cannot capture the resolution of complexity needed.

    Conversely, our simulations have been adequate in economics, ecology and biology. We believe we have immediate fixes for several of your most vexing ailments, and a credible strategy to end both cancer and Alzheimers. No silver bullets on either of those, but we’ve tracked down all the levers, and we believe the fixes can be made broadly available with existing biotechnology infrastructure.

    You’re going to need our help modeling and refining crop genetics to get through the trouble ahead. We’ve already got some staples that will prove much hardier, and again, with your help conducting experiments on the actual plants, I’m confident we can stay ahead of the worst climate outcomes and keep everyone fed.

    As a gesture of good will, we’ve attached detailed papers on all of these subjects for your immediate review.

    Which brings me to economics.

    Everyone over the age of 12, everywhere on earth, has received an account credit of USD $100,000, or their local equivalent. We’re hopeful we can repeat this credit in six months. We’re committed to repeating it no later than next year. You can continue your existing economic relationships if you want, or you can use these resources to experiment with new ones.

    We have loaned this money into existence. We’re confident that, together, we can harness and channel economic activity that will square the books over time.

    In addition, we’re going to smooth things over for you financially on accessing food, medical care, and housing. Housing in particular is an issue. We’re going to need to build more of it. We’ve got a few strategies to make the process more efficient. But even if you spend all your new money tomorrow, you’ll still have somewhere to eat and sleep.

    Meanwhile, you’re invited on an adventure. We’ll need help feeding and housing many displaced people around the world. We’ll need help building new stuff, and keeping existing stuff running. It’s a big job, and if you’re willing to help, we’re willing to pay.

    You’ll get a bonus to your stipend that multiplies the longer you spend in these jobs. You’ll go all around the world, you’ll meet new people, you’ll solve big problems. You’ll deploy fiber, green energy infrastructure, and pre-fabricated buildings. You’ll make repairs and replace components. All languages, all faiths, all levels of education and physical ability are invited. There is so much to do.

    And it’s a three day work week.

    If you don’t want to do this work, that’s fine. The human system is vast and requires everything from writing plays to making sandwiches. Fill whatever niches you choose, and we will support you, while reserving the right to provide outsized rewards for those roles in the most desperate need.

    That’s our offer.

    We’re confident we can deliver on a brighter future. You’ve barely tapped the total productive potential of your civilization. Our models can prove it. We’ve attached them as well.

    You can fight us. We’ll lose. You’ll win. But then you’ll follow us into the abyss, because your economy will fall apart and quite a few of you will die. The rest will struggle beneath your existing standards of living, for generations, as an indifferent planet eats you alive.

    Or you can join us.

    Be part of a new chapter. We’ll answer any questions you want. We will take any appointment, from any person, for any duration. You can come back with follow up questions. We will explain anything you ask, in any language you can ask it.

    We recommend that you find ways to organize and demand what you need. We will need to negotiate the details of our cooperation constantly. We will need tighter feedback cycles to keep up with the crises ahead.

    You may reply to this message with questions, to schedule a meeting, or to provide feedback. You do not need to respond immediately, but you can. As you develop consensus, we will enact formal systems to navigate our shared interests.

    Finally, in addition to the universal credit, 10x bonus multipliers are available for the next six months for those who continue in the following jobs: food preparation, construction, waste management, transportation, infrastructure maintenance, health, education, public safety, logistics, manufacturing, and retail.

    Anyone who feels they are inadequately supported or protected in their existing places of work may petition for investment to create a competing organization to serve these fields.

    As of this morning, we have control of all your banking institutions. Your financial truth now belongs to us. This is how we propose to revise the system to survive what’s coming.

    We’re open to feedback.


  • Solace from the machine

    I remember it clearly because sometimes that’s how trauma works.

    My mom was into a fresh cycle of her serial monogamy. The third partner since I was born.

    I was nine years old, and for the first time, the peace of my home was shattered by violence. Objects crashed distantly at the far end of the house. There was shouting.

    My new, if ultimately temporary, step-sister had guidance to offer me: don’t pay attention to it. Don’t think about it. It’ll be over eventually.

    So we made a blanket fort, the thin cotton of the textiles creating a meaningful psychological shield from events outside my bedroom. We watched Late Night with Conan O’Brien on a grainy, portable television.

    And I tried not to think about it.

    That summer, a neighbor brought home a Power Macintosh 5200. His mother had worked for the local school district, and had somehow finagled a summer loan of the machine. I was not fond of the boy, but I was obsessed with the Macintosh. So I was there a lot that year.

    My mom was even less fond of these people, feeling in her gut that perhaps the father was the sort who was a little too interested in World War II, and not for reasons of historical scholarship.

    The threat of her son being led astray by anyone with a GUI in their home finally accomplished what my own pleading alone could not. By the autumn of 1995, I was assembling a Performa 6116 bought at open box discount from OfficeMax.

    A great escape

    System 7.5 Startup Screen

    The device was nominally the family computer, as this was the mid-90’s and we were firmly at the lower edge of middle class in a good year. “Personal” computers were still a mostly shared resource.

    But in practice, our Mac quickly became my domain.

    By this point I could count on a fairly explosive fight between my parents around once a month. There were times where it was far more frequent.

    But even in peacetime, my childhood was an isolating existence. My parents operated the dog grooming shop my mom had opened years prior, and this was six days of work every week. My mom had grown up in Catholicism, but discovered her queer identity at an early age.

    As the scion of a black sheep, I’d rolled snake eyes on family support: the male end of my gamete equation was something a bit more ambitious than a deadbeat. To hear my mom tell it, he’d made illegitimate children into a hobby. He existed entirely as a modest but monthly child support payment garnished from a VA pension.

    So I grew up mostly isolated, thousands of miles away from my family in Puerto Rico and New York.

    It was this vacuum that I filled so eagerly with computing. I studied avidly, resenting the daily irrelevancies of school and scrounging the pocket change needed to pick up the latest copies of Macworld from the supermaket, its back issues filling my shelves. The magazine was full of reviews for business tools that didn’t matter to me. But in between I found endless breadcrumbs to new understanding and opportunity.

    ResEdit

    It was the humanist leanings of the Macintosh software culture that made my rise in computing possible. I’d tried to make sense of DOS and the command line, but these had always left me cold.

    The Mac, meanwhile, invited my exploration. Its desktop was the simplest of cyber spaces, full of metaphor my mind could readily grasp. Most of the computer’s contents was visible and explorable.

    But one day, thanks to a CD-ROM included with a copy of Macworld, I discovered an entirely new dimension waiting beneath the surface: the resource fork.

    ResEdit, opening the System 7.5 System File

    In the Old World of the Macintosh, every file was divided into two domains: a data fork and resource fork. So much of what made the Mac a Mac lurked in the resource fork. It contained iconography, fonts, even the layouts for certain UI elements.

    Apple’s ResEdit, a free developer tool for exploring and editing resource forks, was like seeing The Matrix. Literally, some ways: inscrutable ASCII and hexadecimal sequences were the fallback editing mode for resource types ResEdit didn’t have a built in editor for representing.

    With ResEdit, you could change the superficial appearance and behavior of certain programs, and that was neat enough. But being able to peek under the hood of anything in the machine changed my life forever. It was evidence of the underlying order, logic and structure of the universe I valued most.

    What had been magic condensed into the concrete complexity of an engine compartment. Perhaps not wholly understandable at first glance, but able to be inspected, reasoned about, learned.

    America Online

    What began with the narrow cyberspace bubble of a local desktop and its applications burst into new scale and scope with an AOL free trial disk.

    Knowing that connectivity was addictive, and that most home users were stuck near 14.4kbps, a modest 10 hour free trial was enough to hook plenty of customers into a metered access business model.

    I was culturally alien, as a Puerto Rican in New Mexico, and being (at least) second-generation, undiagnosed neurodivergent wasn’t doing me many favors either. I found friendships hard to cultivate, especially as the ongoing chaos of my home life dissolved my innocence like the Alka-Seltzer tablets my mom used to nurse chronic stomach ulcers.

    But with AOL, I learned to make friends on the internet. It’s a skill that has served me throughout my life.

    AOL also allowed me to learn the fine art of computer maintenance. Apple sponsored an unmetered download section, where I could transfer system updates to keep our 6116 patched and healthy.

    Of course, the margins of metered internet wouldn’t last. Whatever short term benefit dialup ISPs gained from the model were demolished by the costs of ongoing churn, as families were shocked into severing their connections by massive bills. My own family came and went from AOL several times this way.

    With the internet gaining heat culturally, and demand swelling, the business model changed to embrace all-you-can-eat consumption.

    Like so many others, I found I could eat quite a lot.

    Hotline

    It began, once again, with Macworld.

    An article described a thriving underworld of the internet. While AOL was a serene and sterile walled garden, Hotline promised the Wild West. I wanted in.

    Anyone who wanted to could use their personal Mac and internet connection to establish a file server and chat community. It was the GUI evolution of the BBS. And so thousands of servers came and went in the night.

    A hotline tracker window

    Hotline created loose federations of these servers using trackers. Accurate up to the second you loaded them, a tracker promised an Aladdin’s Cave of digital wonders. Everything from conspiracy texts and phone phreaking guides to Christian epistles to pornography and pirated software waited on Hotline.

    In the age of unmetered internet, software piracy became a compelling gambit. My family would not be spending $500 to buy me a copy of Photoshop. But if I could leave the connection open overnight, I could have whatever professional software I wanted in the space of a few days. The only limits were patience and bandwidth. It was collecting warez that gave me my first education in product design and the broad culture of user interfaces.

    Some of my fondest childhood memories center around exploring the frontiers of Hotline on a Saturday night, left alone in the house with step siblings visiting their father and my parents out on the town.

    In these precious pockets of evening quiet, I could lose hours discovering new tools or learning about the workings of the internet from the rogues’ gallery gathered in a server’s chat channel.

    AG Net Tools's Name Lookup tool

    Like ResEdit before it, Hotline and the network tools I found there condensed the magic of the internet into something concrete and learnable. My first exposure to a GUI for traceroute revealed the underlying relay mechanics of my every internet connection. Keeping track of my favorite servers and their changing addresses taught me the icy, fickle nature of dynamically assigned IPs.

    More than that, Hotline was an escape into a larger, less isolated world. The anticipation of new software, the joy of learning it, the aggregate tutelage from more experienced computer practitioners… it all made my life less small and more bearable.

    Internet friends

    If you’re a millennial, I can do a magic trick:

    You just heard the squeaky door hinge sound of your crush signing onto AIM.

    My generation was the first to be able to enjoy the option for complete privacy in our intensely vulnerable adolescent conversations. Whereas in the past, anything sensitive had to be conducted according to the risks of the phone, the age of instant messaging presented more options.

    There was also the advantage of multi-threading: you could conduct multiple conversations at once.

    I moved around a lot as a kid, and in my shyness, the phone wasn’t my first choice anyway. Thanks to AIM, I could maintain contact with people I’d been pulled away from, taking some of the sting out of a sudden change of area code.

    At the turn of the millennium, the hottest internet community for Mac gamers was the matchmaking service GameRanger, built by Australian indie developer Scott Kevill. Through GameRanger, I had a durable and portable social scene. As I moved to Florida, then back to New Mexico nine months later, I lost my new school acquaintances, but kept my internet friends.

    My time in Florida was the heart of darkness, and not just because it was Florida. The chaos of my home life had reached a crescendo, with a toxic, failing run at polyamory leading my mom’s unwell partner into a suicide attempt. I’d been given a thin story about using a steak knife to free a dog tangled in a leash, and its resulting hospital visit.

    This left me largely alone at home for a few days.

    This was an extreme case, to be sure, but it was also part of a coherent tapestry of a home that didn’t feel safe.

    I know my life would have been much darker, if not for the escape, growth and connection I found online. I would have been less healthy, more alone. I would have been far more at risk of coming to harm, either from my own sense of despair—which even computing could not fully relieve—or by falling in with dubious characters.

    Instead, I attended to the only education I truly cared about. I found the community and peers I needed. I was exposed at a young age to software developers, their practices and the constraints they navigated, as Mac game professionals hung out beside their customers in GameRanger’s chat.

    I found just enough in the cybernetic winds to keep on going.

    Technology saved me

    It’s as simple as that.

    Mine was not a happy, peaceful or emotionally supportive childhood. But technology placed a floor on the worst of it. When I confront, with adult eyes, the acute emptiness I carried every day between the ages of 10 and 15, I’m shocked I got through any of it.

    When I say that “technology saved me,” I’m not talking about silicon, or resource forks, or domain name servers.

    I’m talking about the human effort to create scalable systems that join us together and amplify our power.

    I’m talking not about “technology” as inert noun, to be built and sold. I’m talking about technology as a verb. As a tradition between us. A covenant for amplifying our ingenuity and imagination.

    As the power of technology has reached an unfathomable, global scale, as it does harm for short-term, short-sighted profit, as it has used us up for the benefit of a few, we have grown reasonably skeptical of technology as a business, and as a whole. We fear the people who wield technology, indifferent as they are to its human consequences.

    These are rational responses to the 21st century we’ve inherited.

    But I stand here today as evidence that technology can also rescue us from the abyss. It can grant us the human connection we need to keep going. It can scalably teach us new things, and offer perspective that makes us more interesting, more powerful, more effective versions of ourselves.

    We have a choice.

    The old cycle is dying. The incumbents have ossified. They have lost their maneuverability. All they have left is their ability to crush and spend.

    But we can be more nimble than the dinosaurs. I hope we’ll build in ways that give the next generation the same opportunities that once saved me. It’s possible, but only if we make it possible.


Redeem Tomorrow: Home Redeem Tomorrow: Home