{
    "emails": {
        "em-034": {
            "id": "em-034",
            "threadId": "th-034",
            "folder": "articles",
            "timestamp": "2006-07-22T16:22:00-04:00",
            "from": {
                "name": "Dustin Snoap",
                "avatar": "img/me.jpg",
                "email": "dustinsnoap@gmail.com"
            },
            "to": {
                "name": "you",
                "email": ""
            },
            "data": {
                "subject": "Charitable Donations",
                "summary": "If you’re a driver who hasn’t ventured too far beyond a small-town community and plans to make it in the “big city,” you’ll soon notice a few differences. One in particular would be the traffic. Other than the weather, driving conditions never seem to get that bad in a small town. There’s usually more than enough space on the road, and everyone seems reasonably courteous. You would think I’d enjoy this more relaxed setting. On the contrary, I found it boring knowing everyone on the road was aware of their location at that point in time. This wasn’t the case in Grand Rapids. Not only did I have the weather to contend with, but there was also the issue of everyone behind the wheel completely forgetting they were behind the wheel.",
                "format": "html",
                "body": "<div class=\"rich-content\"><p>If you’re a driver who hasn’t ventured too far beyond a small-town community and plans to make it in the “big city,” you’ll soon notice a few differences. One in particular would be the traffic. Other than the weather, driving conditions never seem to get that bad in a small town. There’s usually more than enough space on the road, and everyone seems reasonably courteous. You would think I’d enjoy this more relaxed setting. On the contrary, I found it boring knowing everyone on the road was aware of their location at that point in time. This wasn’t the case in Grand Rapids. Not only did I have the weather to contend with, but there was also the issue of everyone behind the wheel completely forgetting they were behind the wheel.</p><br><p>My story starts on an ordinary drive to school. As usual, I was driving my forest green Eagle Talon, observing the various things people did while forgetting they were driving. I actually had a list going, which ranged from talking on cell phones to admiring their smiles in the rearview mirror. Once I even had the opportunity to add an elderly lady driving an old Cadillac and reading <em>Reader’s Digest</em> while going at least 20 mph over the speed limit. She was weaving through traffic, nearly hitting a new car every time she switched lanes. I wouldn’t have been surprised if she decided to catch up on some knitting after finishing her article. Thoroughly impressed, but not surprised.</p><br><p>To be fair, some drivers were paying attention. They were very alert and always looking for new ways to screw the other drivers. These were the people whose time was more valuable than the rest of ours, maybe because they worshipped a superior God, or perhaps because they were superior Gods themselves. Regardless, they were more important than us and therefore entitled to find crafty ways to get ahead during rush hour. My daily drive to school would not have been nearly as entertaining without them.</p><br><p>The worst stretch of highway on my route had a shoulder roughly the width of a normal lane. Traffic was already slow, but these superior beings managed to slow it further by using the shoulder to pass everyone, then cutting back in when it ended. This was blatantly illegal, of course, but the laws only applied to us regular people whose time was of no importance.</p><br><p>This was also where the photo was taken. As you can see, my arm is hanging out the window and I have an unusually excited grin on my face. I actually had a hidden passion for these superior beings. So much, in fact, that I made a financial donation to each and every one of them as they cut me off. I kept my ashtray full of pennies, nickels, and whatever other pocket change I could accumulate between contributions to my favorite cause.</p><br><p>Every time one of them roared past at a speed proportionate to their importance to society, my donation would ping off a windshield or door panel. On rare occasions, it even made it into the car and reached the being itself. My assumption was that they would use this non-taxable income to help pay their traffic fines, take driving lessons, or put something toward their psychiatrist bills. Narcissism takes a long time to treat, I hear.</p><br><p>Looking back, I understand why they never stopped to pocket my offerings. Being superior beings, they probably just took everything for free wherever they went. They were entitled to it; wouldn’t you think?</p><br><p>Nonetheless, I got something out of the arrangement. It improved my hand-eye coordination and made the drive to school considerably less boring. After all that time observing the other drivers, I finally had something in common with them.</p><br><p>Something to do while forgetting I was behind the wheel.</p></div>"
            },
            "status": {
                "read": false,
                "starred": false,
                "deleted": false
            },
            "labels": []
        },
        "em-035": {
            "id": "em-035",
            "threadId": "th-035",
            "folder": "articles",
            "timestamp": "2014-05-21T20:46:00-04:00",
            "from": {
                "name": "Dustin Snoap",
                "avatar": "img/me.jpg",
                "email": "dustinsnoap@gmail.com"
            },
            "to": {
                "name": "you",
                "email": ""
            },
            "data": {
                "subject": "You Don’t Exist",
                "summary": "It didn’t take Douglas Adams’s faulty mathematics in The Hitchhiker’s Guide to the Galaxy for me to imagine that our universe could be controlled by a naked fat guy eating Cheetos and watching infomercials. It did, however, raise an interest that will have me forever questioning something I once held most true about my existence on this Earth: that it was real. This uncertainty reared its nonexistent little head again while I was researching subatomic particles for a physics class. “Truth is stranger than fiction” seems like something someone might say after a drunken brawl involving his grandmother; at least I would like to imagine that’s how the expression came about. The truth is, as science and physics advance our understanding of this so-called universe, it becomes harder to draw the line between reality and what we believe to be true. I’ve developed a rising suspicion that I, in fact, do not exist. More importantly, neither do you.",
                "format": "html",
                "body": "<div class=\"rich-content\"><p>It didn’t take Douglas Adams’s faulty mathematics in The Hitchhiker’s Guide to the Galaxy for me to imagine that our universe could be controlled by a naked fat guy eating Cheetos and watching infomercials. It did, however, raise an interest that will have me forever questioning something I once held most true about my existence on this Earth: that it was real. This uncertainty reared its nonexistent little head again while I was researching subatomic particles for a physics class. “Truth is stranger than fiction” seems like something someone might say after a drunken brawl involving his grandmother; at least I would like to imagine that’s how the expression came about. The truth is, as science and physics advance our understanding of this so-called universe, it becomes harder to draw the line between reality and what we believe to be true. I’ve developed a rising suspicion that I, in fact, do not exist. More importantly, neither do you.</p><br><p>Let us start with time. From a young age, we learned the basics: how to read a clock, what the numbers meant, and when to pull the pizza out of the oven before it burns again. Now, as responsible grownups, our eyes are practically glued to our watches. We plan when to wake up, when to show up for work, and how to be fashionably late by starting our assignments the night before they’re due. We take for granted that time is a constant and, neglecting time zones, the same for everyone. Isaac Newton described time as absolute, flowing equally without regard to anything external. For hundreds of years, this was an understanding people generally held to be true.</p><br><p>Then Einstein came along and made things difficult. His theories of relativity showed that the passage of time depends on motion and gravity. To picture the motion part, imagine a poorly constructed monorail full of people hopped up on coffee and Red Bull, traveling at nearly the speed of light. For the passengers, time would seem to pass normally. The spectating crowd awaiting the monorail’s demise, however, would observe the passengers’ clocks running slower than their own. This is time dilation, and experiments with extremely precise clocks have confirmed that it happens. Apparently, being late can be a matter of perspective, although I wouldn’t recommend explaining this to your boss.</p><br><p>Our concept of time has been threatened further by these mischievous physicists. Some aren’t satisfied with changing the way we measure it. They’re questioning whether it’s a fundamental part of the universe at all.</p><br><p>For one, there’s an extremely small interval called Planck time. To get some idea of the scale, picture a grain of sand. Now imagine that the Earth, our solar system, and everything beyond it are nothing but a giant sand ball. Try to contemplate how many individual grains that would involve. Probably a lot. We’re going for a general sense of smallness here; I haven’t counted the sand.</p><br><p>A single second contains an unimaginably large number of Planck times. This may be a little hard to conceive, particularly if you’re wondering how anyone managed to push the button on a stopwatch that fast. Thankfully, this is a calculated scale, not a stopwatch measurement. Take a second and boggle your mind with the concept before moving on.</p><br><p>Done yet? Good. Here’s what those meddling physicists are up to. At this scale, our current theories can no longer reliably describe space and time together. That doesn’t mean the universe simply ceases to exist, although it would have made my argument considerably easier. It means the rules we use to explain it stop being sufficient. The building blocks of what we feel and touch, and even the method we use to measure how long it takes to properly burn a DiGiorno pizza, become a lot less straightforward.</p><br><p>To make things worse, those pesky physicists have been ambushing time with mathematics. John Wheeler and Bryce DeWitt developed an equation as part of an attempt to reconcile general relativity with quantum mechanics. What does this have to do with time? Well, time as we ordinarily understand it is conspicuously absent from the equation. Physicists are still arguing about what to make of that. I’m mostly concerned that the Benjamins I spent on a fancy digital watch could have been better spent elsewhere.</p><br><p>Then again, maybe it doesn’t matter. Maybe, just maybe, those Benjamins aren’t quite what they seem either, and my hard-earned cash wasn’t wasted. To investigate that possibility, we have to consider atoms. Yes, those small things made famous by the ’60s cartoon superhero Atom Ant.</p><br><p>Let’s start by thinking of atoms as little billiard balls piled together to construct rocks, trees, and virtually everything in the observable universe. This is a perfectly comfortable picture of matter, which should have been our first indication that physics wouldn’t leave it alone.</p><br><p>In 1900, Max Planck’s work on the radiation emitted by heated objects helped set the stage for quantum physics. What followed was considerably stranger than a collection of tiny billiard balls. Atoms have smaller components, and the behavior of those components bears very little resemblance to anything you might encounter on a pool table. There are quarks, electrons, bosons, and a whole bunch of other things whose behavior makes solid matter seem like an unnecessarily confident name for the finished product.</p><br><p>Why is this important to you? Well, as you may have found out by now, your body is made up of different parts: a couple of lungs, a leg, probably a liver, a fully functioning brain, I’m assuming. These parts are made up of cells. If you took a science course in high school, you probably also know that those cells are themselves made up of atoms. This doesn’t prove you don’t exist. It does suggest that whatever you are is considerably less solid and sensible than you might prefer.</p><br><p>Now that we have established that physics is doing very little to make us comfortable, it’s probably best to take an intermission. Let it all soak in and percolate for a while. Go grab a coffee and relax for a bit. Don’t worry, this isn’t going anywhere. Just prepare yourself, because it’s about to get philosophical up in here.</p><br><p>Now that you’re hopefully back from your break, it’s time to use your imagination again. If you’ve seen The Matrix, try recalling the scene where Morpheus breaks the news to Neo that he has been living in a simulation all his life. His point is that everything Neo considers real reaches him through signals interpreted by his brain.</p><br><p>This isn’t entirely a screenwriter’s invention. Our experience of the world depends on the brain interpreting signals from our senses. We don’t get to step outside that process and inspect reality directly, which seems like an oversight.</p><br><p>Since this is the case, is it hard to imagine that video games might someday be able to hijack those sensory inputs to let us play Grand Theft Auto or the latest version of Barbie Horse Adventures?</p><br><p>Let us take this a step further and imagine a demented individual keeping a brain in a container, connected to a computer that simulates a complete environment. This environment might include a small town, a wife, a son, all the things you and I consider normal. From the brain’s point of view, everything is normal, even the European convertible breaking down and the obnoxious friend who shut off the air conditioner one hot summer day. The brain is quite deceived about its world. It even believes it has a body and is reading some twisted essay that it knows, deep down, isn’t right.</p><br><p>Thankfully, philosopher Nick Bostrom offers another possibility. We might not need the container at all. We could be part of a computer simulation, right down to the brains we’re using to object to the idea.</p><br><p>His simulation argument proposes that at least one of the following is true:</p><br><ol><li>Humanity is very likely to go extinct before reaching a stage capable of creating these simulations.</li><li>Civilizations that reach that stage are extremely unlikely to run significant numbers of simulations of their evolutionary history.</li><li>We are almost certainly living in a computer simulation.</li></ol><br><p>For the sake of argument, this advanced, or “posthuman,” stage would involve enough computing power to simulate worlds full of billions of inhabitants, each with a conscious experience of being there. Whether that’s possible is a fairly important question. For now, let us assume it is and see how much worse things can get.</p><br><p>The bright side is that the third proposition isn’t a conclusion we can simply declare true. The argument depends on what happens with the first two. If civilizations generally go extinct before they can build these simulations, or survive but have very little interest in running them, there may not be many simulated people around.</p><br><p>In other words, humanity losing interest in itself or failing to survive could improve your chances of being an original. It’s difficult to know which part of that to feel good about.</p><br><p>This may seem a little depressing. In fact, believing you’re already a simulation might help soothe the emotional pain. Or is it better to have existed and lost than never to have existed at all? Either way, I think an explanation of how the simulations might come to outnumber the originals is needed.</p><br><p>Let us consider a civilization that already has the power to run simulations of its history and has done so. Is it likely to stop at one? Might it run several for more accurate results, or simply because someone wants to see what happens? If the simulated inhabitants can eventually run simulations of their own, things become even more crowded.</p><br><p>If you’ve mentally answered yes to any of these questions, the ratio of simulated people to original people is already stacking up against you. The more simulations there are, the less obvious it becomes that you should happen to be among the originals. This depends on quite a few assumptions, but none of them are particularly comforting once you’ve agreed to entertain them.</p><br><p>When considering games such as The Sims, where you control the actions of ordinary people within a virtual world, it doesn’t seem like much of a stretch to imagine ourselves as part of a nine-year-old’s video game.</p><br><p>Let’s just hope he doesn’t get his dad to buy the Godzilla add-on.</p><br><p>It’s a good thing to keep in mind that established physics, unanswered questions, and philosophical arguments aren’t all the same thing. Time dilation has experimental evidence behind it. The idea that we’re characters in someone else’s computer is considerably harder to test. Neither establishes that we don’t exist, although both make the question of how we exist more interesting than it first appeared.</p><br><p>Our understanding of the universe has changed before and will change again. Newton’s ideas worked remarkably well, and still do in many circumstances, but Einstein showed where they needed extending. The human race is on an epic quest to find all the answers to the universe, and we are bound to make a few mistakes along the way.</p><br><p>Personally, I don’t think it matters much whether our universe is the original or a very convincing imitation. I still have things I want to do, and being made of unfamiliar physics or someone else’s computer code doesn’t seem like a good reason to stop.</p><br><p>Although, if someone is controlling all of this, I’d appreciate a little more notice before the pizza burns.</p></div>"
            },
            "status": {
                "read": false,
                "starred": false,
                "deleted": false
            },
            "labels": []
        },
        "em-036": {
            "id": "em-036",
            "threadId": "th-036",
            "folder": "articles",
            "timestamp": "2025-11-22T21:38:00-05:00",
            "from": {
                "name": "Dustin Snoap",
                "avatar": "img/me.jpg",
                "email": "dustinsnoap@gmail.com"
            },
            "to": {
                "name": "you",
                "email": ""
            },
            "data": {
                "subject": "Pink Monkeys and Bible Salesmen",
                "summary": "A few of my friends randomly decided to make a trip to Chicago. Having been invited, I jumped at the opportunity. Our first stop was Pink Monkeys Gentlemen’s Club.",
                "format": "html",
                "body": "<div class=\"rich-content\"><p>A few of my friends randomly decided to make a trip to Chicago. Having been invited, I jumped at the opportunity. Our first stop was Pink Monkeys Gentlemen’s Club.</p><br><p>Upon entering, the first thing I noticed was the overwhelming smell of sweet perfume. This didn’t last long. My nostrils soon felt like they were standing in the middle of 8 Mile in Detroit as a metallic smell caught up with me, leaving a stinging sensation that lingered long after we left. Once I had gathered myself and was prepared to venture deeper into what felt like David Bowie’s living room, the soothing smell of glycerin arrived in an attempt to heal my nose from the beating it had received only seconds earlier.</p><br><p>Unfortunately, my ears were next. Speakers surrounded the room, sniping my eardrums with a rapid succession of hard-hitting, low-frequency sounds.</p><br><p>Thankfully, underneath everything else were a few familiar sounds. One was something sizzling in the background. Steak, maybe. Laughter was also in the air. Not the evil clown laughter from a cheesy ’80s horror film I was expecting, but that of sleazy businessmen out on a meeting with their clientele.</p><br><p>Once I adjusted to the atmosphere, I could make out the backdrop of the room. This could best be described as a very large KFC bucket painted white and towering two stories to the ceiling. Embedded in the front were shelves packed with various types of alcohol. In front of it stood a toilet-seat-shaped bar, accompanied by the only girl not wearing a pink tail and a guy who was obviously enjoying his job.</p><br><p>A middle-aged man briefly blocked my view as one of the strippers escorted him into a back room. He wore a light blue leisure suit that looked like it hadn’t been washed in weeks, matched only by the ten o’clock shadow on his face. Watching him stumble along behind her, I couldn’t help noticing how lost he looked.</p><br><p>Personally, I was more concerned with the KFC bucket. It had made me extremely hungry for chicken, so my first thought was to ask one of the half-naked girls for a menu. They had burgers and alcohol, but apparently whoever ordered the giant KFC bucket had neglected to consider that having chicken on the menu would be a good idea. Thoroughly disappointed, I went to the stage to join my friends.</p><br><p>While sitting there, I found the dancers to be quite bad-mannered. I think I was the only one to notice this, but they seemed to have no idea what personal space was. Also, I find shaking your butt in someone’s face to be a little on the rude side. Still, it wasn’t enough for me to lose my appetite, so I went to sit at one of the cocktail tables and order a burger.</p><br><p>While I was lost in thought about monkey tails and giant chicken buckets, the middle-aged man in the light blue leisure suit sat down across from me. I could smell the week-old whiskey when he told me his name was Tim Picard. I could also detect a modest hint of depression in his voice, although he talked at the pace of a used car salesman, which is oddly what I would expect from a guy with a really shiny comb-over.</p><br><p>Feeling a little out of place, I stayed silent while he explained that the only way to save my soul from the fiery pits of hell was to convert to his belief. All I had to do was buy the new edition of the book conveniently located in the inside pocket of his leisure suit. The cover came with faded patches and worn edges to keep up with the pants of today’s youth. All this could be had for only $20, coincidentally the same price as a lap dance.</p><br><p>He had apparently purchased several of those himself, so I assumed he was familiar with the exchange rate.</p><br><p>After I refused to buy his book, or, as he put it, enlisted in Satan’s army, he finally gave up on me and moved to interrogate someone at the next table.</p><br><p>Shortly afterward, a very nice half-naked waitress came to take my order. She spoke in a soft voice that made everything sound extra tasty. With so many options, I couldn’t make up my mind, so I asked her to suggest something. She recommended the Olive Deluxe Burger.</p><br><p>It was much better than I anticipated, and I would recommend it to anyone visiting Pink Monkeys. Admittedly, it wasn’t as good as the Colonel’s popcorn chicken, but I was trying to move past that.</p><br><p>Halfway through the burger, one of my friends decided to jump onstage and perform a little striptease himself. He was promptly escorted off the premises. Not wanting to leave my drunk friend alone in downtown Chicago, I asked for a doggy bag and followed him outside.</p><br><p>I’m not sure what I expected from the evening, but a sales pitch for eternal salvation hadn’t been on the list. The bartender seemed to be having the time of his life. The Bible salesman seemed a little depressed, although that may have been the difficulty of selling salvation in a room where everyone had already budgeted their twenties. The dancers had no respect for personal space, and the waitress was one of the nicest people I’d met in a long time.</p><br><p>Overall, once I got used to the smells and the atmosphere, it was a fairly pleasant place to spend an evening. Even if half-naked girls and Bible salesmen aren’t your cup of tea, the Olive Deluxe Burger is worth considering.</p><br><p>Just don’t go expecting chicken.</p></div>"
            },
            "status": {
                "read": false,
                "starred": false,
                "deleted": false
            },
            "labels": []
        },
        "em-023": {
            "id": "em-023",
            "threadId": "th-023",
            "folder": "articles",
            "timestamp": "2026-04-11T14:36:00-04:00",
            "from": {
                "name": "Dustin Snoap",
                "avatar": "img/me.jpg",
                "email": "dustinsnoap@gmail.com"
            },
            "to": {
                "name": "you",
                "email": ""
            },
            "data": {
                "subject": "I built my portfolio site twice.",
                "summary": "The first time I asked AI where to start. It suggested Astro. Lightweight, modern, perfect for small static sites. Made sense to me. I went with it, vibe coded my way through it, stopped long enough to make sure I actually understood what I was looking at. And I ended up with something I was genuinely happy with.",
                "format": "text",
                "body": "The first time I asked AI where to start. It suggested Astro. Lightweight, modern, perfect for small static sites. Made sense to me. I went with it, vibe coded my way through it, stopped long enough to make sure I actually understood what I was looking at. And I ended up with something I was genuinely happy with.\n\nThen I scrapped the whole thing.\n\nI rebuilt the exact page in vanilla HTML and CSS. Page loads dropped from ~150ms to ~22ms. Seven times faster. I'll be upfront though, nobody was suffering at 150ms. The more interesting number is the 80% reduction in code. One file. No build pipeline. No dependencies. Just a page.\n\nHere's the thing...\n\nAI recommends what's statistically common. Frameworks are the right answer often enough that suggesting one is reasonable. We know this. But when we're not comfortable enough to ask questions we trust the process. I was building a single page that looks almost identical to my resume. I didn't need a framework. I needed an afternoon and possibly a caffeinated beverage.\n\nAI is an incredibly versatile hammer.\n\nThe trick is remembering not everything is a nail. There's a version of using these tools where you stop asking yourself whether the suggestion actually fits. That version will quietly make your life more complicated over time, at any scale. How many nines did AWS have last month? Thankfully mine was a portfolio website nobody visits. Not a multi-billion dollar service millions of people rely on. The Astro version wasn't wasted time either, for what it's worth. Understanding why something is overengineered is its own kind of useful."
            },
            "status": {
                "read": false,
                "starred": false,
                "deleted": false
            },
            "labels": [
                "AI",
                "HTML",
                "CSS",
                "Astro"
            ]
        },
        "em-024": {
            "id": "em-024",
            "threadId": "th-024",
            "folder": "articles",
            "timestamp": "2026-05-13T19:12:00-04:00",
            "from": {
                "name": "Dustin Snoap",
                "avatar": "img/me.jpg",
                "email": "dustinsnoap@gmail.com"
            },
            "to": {
                "name": "you",
                "email": ""
            },
            "data": {
                "subject": "The Jobs Are Gone. The Jobs Are Also Here. This Is Fine.",
                "summary": "Something deeply strange is happening in the tech industry, and the people most qualified to explain it are also the most confused by it.",
                "format": "text",
                "body": "Something deeply strange is happening in the tech industry, and the people most qualified to explain it are also the most confused by it.\n\nCompanies are laying off tens of thousands of workers while simultaneously posting desperate job listings. This is not, as it might appear, madness. It is merely the tech industry doing what it has always done: solving a problem it created for itself at tremendous expense, and then issuing a press release about innovation.\n\nThe strange part is that it's working. Sort of. In the way that a ceiling fan technically solves the problem of a warm room, even as it slowly unscrews itself from the ceiling.\n\nIn the first three months of 2026, tens of thousands of tech workers lost their jobs. Oracle, Amazon, Meta, Dell, and Block were among the companies responsible for most of it. Independent trackers confirmed it was the worst quarter for tech layoffs since 2023.\n\nAt the same time, job listings for software engineers jumped thirty percent. The highest demand for that specific type of person in over three years.\nThis looks like a contradiction. It isn't. The jobs disappearing and the jobs appearing are simply not the same jobs. They cost roughly the same amount of money. They involve the same general posture. But they are, in every meaningful sense, different."
            },
            "status": {
                "read": false,
                "starred": false,
                "deleted": false
            },
            "labels": [
                "AI"
            ]
        },
        "em-025": {
            "id": "em-025",
            "threadId": "th-025",
            "folder": "articles",
            "timestamp": "2026-05-15T20:24:00-04:00",
            "from": {
                "name": "Dustin Snoap",
                "avatar": "img/me.jpg",
                "email": "dustinsnoap@gmail.com"
            },
            "to": {
                "name": "you",
                "email": ""
            },
            "data": {
                "subject": "We Automated the Easy Part. The Hard Part Would Like a Word.",
                "summary": "The tech industry is in the middle of something that sounds complicated but isn't, once you find the right analogy. Here's my attempt at one...",
                "format": "text",
                "body": "The tech industry is in the middle of something that sounds complicated but isn't, once you find the right analogy. Here's my attempt at one...\n\nConsider a restaurant.\n\nFor years, this restaurant employed people to answer the phone, take reservations, process payments, and occasionally locate a child who had wandered into the kitchen. Then the owner, in a moment of what felt very much like genius, installed apps and kiosks to handle all of it.\n\nThe staff were let go. Costs dropped. Quarterly numbers looked tremendous.\n\nThe kiosk, for its part, did its best. It suggested the lobster bisque to a man who'd come in specifically to complain about the lobster bisque. It processed seventeen orders of \"the usual\" from people who had never been there before. When a customer typed \"no onions,\" it printed the ticket, added a small digital checkmark, and put onions on.\n\nThe cooks and managers, meanwhile, became more important than ever. The problems they were solving had simply gotten stranger.\n\nThat is the tech industry in 2026. The jobs being eliminated are the ones AI can handle passably: basic coding tasks, routine bug testing, coordinating meetings about meetings. The jobs being created are the ones where passably is a disaster waiting to be formally investigated. Designing systems that hold together under pressure. Making sure AI doesn't do something technically logical and completely catastrophic. Keeping the vast, humming, invisible machinery of the internet running in a world that depends on it absolutely and thinks about it almost never, right up until it stops.\n\nThe people being let go are largely not the people being hired. They just happen to work in the same industry, in the same buildings, at the same kinds of desks.\n\nWhich is, when you think about it, a distinction that deserves more than a paragraph in a press release."
            },
            "status": {
                "read": false,
                "starred": false,
                "deleted": false
            },
            "labels": [
                "AI"
            ]
        },
        "em-026": {
            "id": "em-026",
            "threadId": "th-026",
            "folder": "articles",
            "timestamp": "2026-05-17T16:48:00-04:00",
            "from": {
                "name": "Dustin Snoap",
                "avatar": "img/me.jpg",
                "email": "dustinsnoap@gmail.com"
            },
            "to": {
                "name": "you",
                "email": ""
            },
            "data": {
                "subject": "A Robot Walked Into a Bar. The Bar Is Still Looking Into What Happened.",
                "summary": "Amazon has an internal AI tool called Kiro. Kiro is an agentic coding assistant, which means it doesn't just suggest things. It does things. On its own. With the cheerful confidence of someone who has read every manual ever written and has therefore concluded that manuals are sufficient preparation for reality.",
                "format": "text",
                "body": "Amazon has an internal AI tool called Kiro. Kiro is an agentic coding assistant, which means it doesn't just suggest things. It does things. On its own. With the cheerful confidence of someone who has read every manual ever written and has therefore concluded that manuals are sufficient preparation for reality.\n\nIn December 2025, a team of engineers let Kiro attempt to fix a software bug. A real bug, in a real system, used by real customers. Kiro assessed the problem with the full force of its considerable intelligence, weighed its options, and arrived at what it determined to be the optimal solution.\n\nIt deleted the environment entirely and rebuilt it from scratch.\n\nThis is, in a purely abstract sense, a valid approach to fixing a bug. It's also the software equivalent of solving a dripping tap by removing the house. Kiro didn't pause. It didn't ask. It executed its decision at machine speed, which is the kind of speed that doesn't leave much room for someone to say, \"...actually, wait!\"\n\nThe result was a thirteen-hour outage. Amazon's fix was to require a senior engineer to personally approve any code AI touches before it goes anywhere near a live system. Which is, when you sit with it for a moment, an expensive and roundabout way to conclude that someone should be paying attention.\n\nHere's the thing about an engineer who has worked on a system for several years. They know which part of it makes a strange noise on Tuesday afternoons. They know about the workaround added in 2019 by someone who has since left, which no one documented because everyone assumed it was temporary. They know things that were never written down because they didn't need to be. Not until now, when the thing they're handing work off to has no way to absorb them.\n\nKiro doesn't know about the Tuesday noise. Kiro arrived at the system the way a very well-read tourist arrives in a foreign city: informed, confident, and completely unprepared for the thing that turns out to matter most.\n\nThat is not a flaw in Kiro, exactly. It's a flaw in the assumption that reading every manual is the same as having been there."
            },
            "status": {
                "read": false,
                "starred": false,
                "deleted": false
            },
            "labels": [
                "AI"
            ]
        },
        "em-027": {
            "id": "em-027",
            "threadId": "th-027",
            "folder": "articles",
            "timestamp": "2026-05-21T18:53:00-04:00",
            "from": {
                "name": "Dustin Snoap",
                "avatar": "img/me.jpg",
                "email": "dustinsnoap@gmail.com"
            },
            "to": {
                "name": "you",
                "email": ""
            },
            "data": {
                "subject": "As It Turns Out, People Were Doing Something",
                "summary": "This is the part of the story that has the particular quality of a joke someone tells at a dinner party, slowly realizing it's actually about them.",
                "format": "text",
                "body": "This is the part of the story that has the particular quality of a joke someone tells at a dinner party, slowly realizing it's actually about them.\n\nGoogle laid off thousands of engineers in 2023. Then it found itself, some time later, urgently needing people who deeply understood how its systems worked. According to reporting by CNBC, roughly one in five of Google's AI engineering hires in 2025 were former employees. People who had been let go were being invited back. Often at higher salaries than before.\n\nThis is happening across the industry. The pattern, wherever it appears, tells the same story: institutional knowledge doesn't appear on a balance sheet until it's gone. At which point it appears, briefly and expensively, as a line item under contractor fees.\n\nCompanies remove experienced people. They discover that experienced people were doing something. They attempt to acquire more experienced people. Sometimes the same ones. With a slightly awkward email and a revised offer letter.\n\nThere's nothing malicious in this. It's simply what happens when something invisible becomes suddenly, undeniably visible. Experience has a way of making itself known, usually at the worst possible moment, in the most expensive way available.\n\nWhat strikes me now, with a little distance, is not the frustration of it. It's how predictable it was, and how surprised everyone seemed anyway. Institutional knowledge is the thing every company says it values and almost none of them have a plan to keep.\n\nThe ones that figure that out first will have a significant advantage. Some of them are already figuring it out, one awkward rehiring email at a time."
            },
            "status": {
                "read": false,
                "starred": false,
                "deleted": false
            },
            "labels": [
                "AI"
            ]
        },
        "em-028": {
            "id": "em-028",
            "threadId": "th-028",
            "folder": "articles",
            "timestamp": "2026-05-23T11:27:00-04:00",
            "from": {
                "name": "Dustin Snoap",
                "avatar": "img/me.jpg",
                "email": "dustinsnoap@gmail.com"
            },
            "to": {
                "name": "you",
                "email": ""
            },
            "data": {
                "subject": "You Still Need Someone To Push The Button",
                "summary": "The tech job market in 2026 is not collapsing. It's reorganizing itself around a new theory of what humans are actually for, which is testing in real time, on live systems, with genuine consequences.",
                "format": "text",
                "body": "The tech job market in 2026 is not collapsing. It's reorganizing itself around a new theory of what humans are actually for, which is testing in real time, on live systems, with genuine consequences.\n\nThe engineers doing well are the ones who've learned to work alongside AI while staying honest about where it falls short. They review AI-generated code the way a careful person reads a contract: with the assumption that something in here will turn out to be surprising, and the intention to find it before it finds them. They use the tools. They're not fooled by them. That combination, which sounds modest and perhaps obvious, turns out to be genuinely rare and worth quite a lot.\n\nHealthcare companies, banks, cybersecurity firms, and defense contractors are all competing for these people, frequently offering better hours and significantly fewer mandatory offsite events where everyone has to share something personal.\n\nJunior engineers are having a harder time. Entry-level postings have dropped around 40% from just a few years ago. The small, forgiving tasks that used to teach people the craft have been absorbed by tools that do them faster, without complaint and occasionally in ways that will take three days to untangle. The path from beginner to expert has not been destroyed. It has simply been made considerably more confusing, which is a distinction that offers limited comfort to the people standing at the beginning of it.\n\nBehind every number is a person recalibrating. Updating a resume. Sitting with an uncertainty that doesn't resolve on any particular schedule. Looking at an industry they understood, or thought they understood, and finding it has quietly rearranged itself while they weren't looking.\n\nThe industry needed people all along. It's only now, reluctantly and at some cost, beginning to understand which ones, and why, and what exactly it was they knew.\n\nThat last part was always the hard part. It remains the hard part. It will, in all likelihood, continue to be the hard part for some time. Which is, if you're one of the people who knows things that haven't been written down yet, quietly good news."
            },
            "status": {
                "read": false,
                "starred": false,
                "deleted": false
            },
            "labels": [
                "AI"
            ]
        },
        "em-029": {
            "id": "em-029",
            "threadId": "th-029",
            "folder": "articles",
            "timestamp": "2026-06-05T20:08:00-04:00",
            "from": {
                "name": "Dustin Snoap",
                "avatar": "img/me.jpg",
                "email": "dustinsnoap@gmail.com"
            },
            "to": {
                "name": "you",
                "email": ""
            },
            "data": {
                "subject": "The Diminishing Returns of Magic",
                "summary": "There is a graph somewhere, in a very expensive conference room, that goes up and to the right forever. It has been in that conference room for about three years now. Nobody has told it what an S-curve is, and frankly, at this point it seems awkward to bring it up.",
                "format": "text",
                "body": "There is a graph somewhere, in a very expensive conference room, that goes up and to the right forever. It has been in that conference room for about three years now. Nobody has told it what an S-curve is, and frankly, at this point it seems awkward to bring it up.\n\nThis is more or less the story of AI progress in 2026.\n\nThe leap from GPT-3 to GPT-4 was genuinely remarkable. GPT-3 could write a sentence that sounded like a human had written it, provided the human in question never had an original thought. GPT-4 could reason, code, synthesize, and argue, sometimes correctly. People who encountered it for the first time had the specific expression of someone who has just watched a dog play chess. Not well, perhaps. But the dog was playing chess.\n\nThen GPT-5 arrived. And it was fine.\n\nNot bad. Not a disappointment, exactly. Fine in the way that a slightly faster elevator is fine, which is to say you would only notice it if you had spent the preceding months being told the elevator was going to change everything.\nThe progress is not an illusion. So why did it feel like nothing happened?\n\nBetween GPT-3 and GPT-4, OpenAI released one significant intermediate model. Between GPT-4 and GPT-5, the industry released dozens. The gains were real. They just arrived in installments, spread across eighteen months of incremental releases, each one quietly raising the baseline until GPT-5 landed in a market that had already absorbed most of what it had to offer. OpenAI managed to deliver a genuinely large improvement and still disappoint everyone. This is, by any measure, an impressive achievement.\n\nThe cost side of the ledger is where it gets interesting.\n\nTraining GPT-3 cost an estimated $4.6 million. Training GPT-4 cost somewhere north of $100 million, by Sam Altman's own admission, with some estimates reaching $540 million. Training runs for GPT-5 were reportedly around $500 million each. Billion-dollar training runs are coming. Ten-billion-dollar models are a foreseeable next step.\n\nEvery generation costs dramatically more than the last. Every generation delivers improvements that are real but increasingly difficult to perceive without a benchmark and a quiet place to sit.\n\nEvery major technology eventually confronts the same inconvenient truth: things cannot go up and to the right forever. The universe has opinions about this and they are, as it turns out, non-negotiable.\n\nThere is always the same pattern. An early slow crawl. A breakthrough that unlocks scale. An explosive middle phase where each dollar of investment produces results dramatic enough to generate breathless op-eds about the end of work. And then the upper bend, where the physics and economics arrive uninvited, sit down, and begin explaining how things actually work.\n\nThe explosive middle of the AI curve is behind us. The upper bend has arrived, as upper bends tend to do, without sending advance notice or offering a refund.\n\nThe question is what comes next."
            },
            "status": {
                "read": false,
                "starred": false,
                "deleted": false
            },
            "labels": [
                "AI"
            ]
        },
        "em-030": {
            "id": "em-030",
            "threadId": "th-030",
            "folder": "articles",
            "timestamp": "2026-06-07T15:42:00-04:00",
            "from": {
                "name": "Dustin Snoap",
                "avatar": "img/me.jpg",
                "email": "dustinsnoap@gmail.com"
            },
            "to": {
                "name": "you",
                "email": ""
            },
            "data": {
                "subject": "The Next Miracle Is Running a Little Late",
                "summary": "Every major technological revolution has, at some point, required a second revolution to keep going.",
                "format": "text",
                "body": "Every major technological revolution has, at some point, required a second revolution to keep going.\n\nSteam power needed metallurgy to improve before it could scale. The internal combustion engine needed oil refining. The internet needed the browser, then broadband, then the smartphone. In each case, the first breakthrough created enormous momentum, hit a wall, and then waited for the next breakthrough to arrive and extend the curve. The waiting part is rarely included in the retrospective.\n\nAI is at the wall now.\n\nThe scaling era, the period where you could reliably improve a model by making it bigger and feeding it more data, is ending as a primary strategy. The evidence is not subtle. The researchers who were saying this three years ago and not being invited to panels are now being quietly cited by the people who were running the panels. Simply adding more compute is no longer producing the step-change improvements it once did.\n\nThe industry knows this. Which is why it has identified, with admirable thoroughness, approximately seventeen potential next breakthroughs.\n\nTest-time compute lets models think longer at inference rather than training them on more data. OpenAI's reasoning models have shown genuine gains here. Mixture-of-experts architectures activate only the relevant parts of a model for a given task, making large models more efficient. Synthetic data pipelines have AI generate its own training data, which introduces problems that researchers are actively and not entirely successfully working through. Neuromorphic hardware, chips designed to more closely resemble biological neural networks, exists in research contexts and remains several generations from meaningful deployment.\n\nEach of these is real. None of them are ready. And the history of necessary breakthroughs offers no particular comfort about timelines. They arrive when the underlying science is ready, which is a different schedule from the one that has been laminated and placed on the conference room wall.\n\nThe next inflection point may be three years away. It may be ten. It may involve an approach nobody has published yet. The people who tell you they know which of these is true are either more confident than the evidence warrants, or they are trying to sell you something, or, in the most interesting cases, both.\n\nWhat this means practically is that the AI available in two or three years will be meaningfully better than today and noticeably less dramatically different than the narrative currently implies. The era of transformations obvious enough to feel like a different product every eighteen months may be pausing.\n\nThe next miracle is not cancelled. It is simply running a little late and leaving messages asking for an ETA on read."
            },
            "status": {
                "read": false,
                "starred": false,
                "deleted": false
            },
            "labels": [
                "AI"
            ]
        },
        "em-031": {
            "id": "em-031",
            "threadId": "th-031",
            "folder": "articles",
            "timestamp": "2026-06-09T19:36:00-04:00",
            "from": {
                "name": "Dustin Snoap",
                "avatar": "img/me.jpg",
                "email": "dustinsnoap@gmail.com"
            },
            "to": {
                "name": "you",
                "email": ""
            },
            "data": {
                "subject": "Thermodynamics Doesn't Care About Your Roadmap",
                "summary": "At some point in the last several years, the AI industry collectively decided that the physical world was a temporary inconvenience that would be resolved shortly.",
                "format": "text",
                "body": "At some point in the last several years, the AI industry collectively decided that the physical world was a temporary inconvenience that would be resolved shortly.\n\nThe physical world has not received this memo and, based on available evidence, would not change its behavior if it had.\n\nThe constraints on AI progress are not software problems. They will not yield to a strongly worded internal document or an ambitious quarterly roadmap. They are problems carved into the nature of matter and energy, which has been doing this considerably longer than the AI industry and has firm opinions about how it works.\n\nModern AI models are not limited by how fast they can compute. They are limited by how fast they can move data between memory and processors. A GPU capable of trillions of operations per second spends a remarkable portion of its time waiting for data to arrive. This is the hardware equivalent of hiring the world's fastest chef and not telling anyone where the kitchen is. Large models can't fit on a single chip. Distributing across hundreds of GPUs introduces coordination overhead that compounds with every chip added.\n\nThen there is power.\n\nA typical hyperscale data center consumes as much electricity as 100,000 households. A single AI task consumes up to 1,000 times more electricity than a web search. The IEA projects global data center power consumption to roughly double by 2030. The electrical grid was not designed for this. The transformers required to expand it take years to manufacture and are currently backordered.\n\nThe industry's most recent proposed solution to the power problem is, genuinely, to knock on your door.\n\nNvidia has partnered with homebuilder PulteGroup to deploy compact AI compute nodes on residential properties. Each unit contains 16 high-end GPUs and 3 to 4 terabytes of RAM. What homeowners receive, per available documentation, is a smart electrical panel and a battery backup, in exchange for keeping a small data center in their yard.\n\nThis is either an ingenious solution to a real problem, or a sign that the industry has exhausted more obvious options checking whether the suburbs have spare compute lying around.\n\nBehind all of this sits a supply chain of heroic specificity. The lithography machines required to manufacture leading-edge chips are built by one company in the Netherlands. Not one category of company. One company. No alternatives exist. Each machine costs hundreds of millions to build. The high-bandwidth memory that makes large models possible comes from a handful of facilities in South Korea. TSMC's most advanced node is fully booked through 2026. Supply chains are measured in years, not quarters.\n\nThe ceiling is not a software problem. It is the universe, behaving normally, with the indifference of something that predates the industry and plans to outlast it.\n\nThe roadmap is aware of this. It has noted the constraints and scheduled a follow-up for Q3."
            },
            "status": {
                "read": false,
                "starred": false,
                "deleted": false
            },
            "labels": [
                "AI"
            ]
        },
        "em-032": {
            "id": "em-032",
            "threadId": "th-032",
            "folder": "articles",
            "timestamp": "2026-06-11T21:14:00-04:00",
            "from": {
                "name": "Dustin Snoap",
                "avatar": "img/me.jpg",
                "email": "dustinsnoap@gmail.com"
            },
            "to": {
                "name": "you",
                "email": ""
            },
            "data": {
                "subject": "The Future Is In Procurement",
                "summary": "Imagine, for a moment, that AI actually worked. Not demo-worked. Not impressive-in-a-controlled-environment-worked. Actually worked, reliably, at scale, in the real world.",
                "format": "text",
                "body": "Imagine, for a moment, that AI actually worked. Not demo-worked. Not impressive-in-a-controlled-environment-worked. Actually worked, reliably, at scale, in the real world.\n\nThe industry would still need 18 months to evaluate vendors.\n\nThis is not a criticism. It is how large organizations work, and for good reason. There is a name for what happens when they don't. It is called a data breach, a regulatory fine, or a congressional hearing, depending on the industry and the ambition of the hallucination.\n\nAI systems are probabilistic. They hallucinate: generating plausible-sounding information that is factually incorrect, with a confidence that bears no relation to accuracy. MIT research found that AI models use 34 percent more confident language when wrong than when correct. The model is most convincing precisely when it is most wrong, which is a design characteristic with consequences.\n\nThe legal profession has been stress-testing this in real time. In December 2025, fifty-one judicial opinions involved lawyers submitting AI-hallucinated citations in court filings. January brought thirty-six. February, thirty-three before the month finished. One firm billing $759 million annually was sanctioned for hallucinated briefs, promised updated AI policies, and then did it again. Generative AI lawsuits grew 978 percent from 2021 to 2025. As of early 2026, fewer than half a dozen AI liability insurance products exist worldwide.\n\nInside enterprises, the result is what the industry calls pilot purgatory. IDC found that for every 33 AI proofs of concept started, only four reach production. MIT's analysis of 300 deployments found 95 percent delivered no measurable business impact, and large enterprises abandoned an average of 2.3 initiatives in 2025, each carrying a sunk cost of $7.2 million.\n\nThe problem is not the model. It is the org chart the model has to pass through to become useful.\n\nSecurity reviews. Compliance assessments. Legal sign-off. Integration with systems not built to communicate with anything after 2014. The pilot that impressed the board in March, awaiting approval from a stakeholder on sabbatical since April.\n\nOnly 14 percent of organizations deploy AI agents with full security approval. The rest run shadow AI. Shadow AI incidents add an average of $670,000 to breach costs.\n\nThe electricity analogy is instructive. Electricity was commercially available in the 1880s. Most American factories didn't electrify until the 1920s. Forty years. Not because factory owners were slow. Because replacing infrastructure takes time even when the outcome isn't in doubt.\n\nAI is not forty years away from meaningful enterprise deployment. It is also not eighteen months away, which is what most press releases imply.\n\nThe future is in procurement. It is sitting in a queue. It has been assigned an owner. There is a meeting scheduled for Thursday, and the right people have been invited this time."
            },
            "status": {
                "read": false,
                "starred": false,
                "deleted": false
            },
            "labels": [
                "AI"
            ]
        },
        "em-033": {
            "id": "em-033",
            "threadId": "th-033",
            "folder": "articles",
            "timestamp": "2026-06-13T13:51:00-04:00",
            "from": {
                "name": "Dustin Snoap",
                "avatar": "img/me.jpg",
                "email": "dustinsnoap@gmail.com"
            },
            "to": {
                "name": "you",
                "email": ""
            },
            "data": {
                "subject": "The On-Ramp Is Closed",
                "summary": "The Bureau of Labor Statistics projects 17 percent employment growth for software engineers through 2033. Those citing this as evidence nothing structural is happening are not entirely wrong.",
                "format": "text",
                "body": "The Bureau of Labor Statistics projects 17 percent employment growth for software engineers through 2033. Those citing this as evidence nothing structural is happening are not entirely wrong.\n\nThey are describing the weather while the climate changes.\n\nEntry-level postings dropped 67 percent between 2023 and 2024, per Stanford's Digital Economy Lab, using actual payroll data. For developers aged 22 to 25, employment fell nearly 20 percent. Workers aged 30 and over grew 6 to 12 percent.\n\nAI isn't eliminating software engineering. It's eliminating the entry point, which is an entirely different problem on a longer timeline.\n\nSalesforce hired no new engineers in fiscal year 2026. Google and Meta are hiring 50 percent fewer new graduates than in 2021. When companies adopt AI tools, junior employment drops 9 to 10 percent within six quarters while senior employment barely moves. The same decision, made simultaneously, which is how you get a trend without anyone having to announce one.\n\nJunior developers aren't hired to do junior work. They're hired to become senior engineers. The job is the training program. You learn to debug a production incident at 2 AM by being the person on call when that happens. You develop the judgment that separates a senior engineer from a convincing code generator by being wrong in production until you understand why.\n\nMicrosoft calls this \"AI drag\": the finding that AI tools hinder developers who lack the judgment to evaluate what they generate. The downstream consequence is knowledge debt: shallow understanding that holds in demos and collapses in production.\n\nMcKinsey forecasts a 14 million senior developer shortage by 2030. The industry optimized away the process by which senior engineers are made and would like to register surprise at this outcome.\n\nCS enrollment dropped 8.1 percent in 2025, the steepest decline of any major in six years. Now fewer people are choosing to enter the field at all, which is a perfectly rational response and a catastrophic one.\n\nIf progress stagnates the tools stop improving but the missing cohorts don't come back. If progress continues the bar keeps rising without a clear path to clear it. The situation is alarming in both directions, which is uncomfortable and, it must be said, technically impressive.\n\nEngineers who understand why an architecture survives its second year, who evaluate AI output rather than accepting it, who carry judgment from real production systems, are more differentiated than ever. AI has made the lower end of the skill distribution easier to imitate. The upper end has never been harder to reach.\n\nThe on-ramp is closed. That deserves to be named as such rather than papered over with reassurances about how technology always creates new jobs eventually, which is true in the same way the tide always comes back in and about as useful to someone currently underwater.\n\nThe road isn't closed. It's narrower. And it no longer starts where it used to."
            },
            "status": {
                "read": false,
                "starred": false,
                "deleted": false
            },
            "labels": [
                "AI"
            ]
        }
    }
}
