It was someone’s LLM-wrapper startup: upload your résumé, the LLM will interview you, and it’ll craft new bullet points. I dropped $40 on it because I thought it might be worth it, and I think it was, though not necessarily in the way intended. The bullet points were okay. But I spent several hours in total thinking about the questions it asked and crafted some pretty heavy responses (which I wish I’d kept), and I think that was pretty priceless.

For whatever reason, I’ve found it far easier to focus lately, and to do the kind of hellish introspection that I think can really benefit me… but has historically come only in the deepest fits of depression. It’s different now. It’s like I can just open my mouth, reach down my throat, and pull my guts out.

The roughest part about today was thinking about my first job. My first job as a software engineer, that is. It started in 2012. I was 31, just out of college with my bachelor’s degree in Computer Science. I didn’t know anything about anything. And spoiler alert, it went awfully. I think it went about as badly as it could possibly have gone.

So I’m sitting there, trying to respond to the LLM’s questions about my bullshit bullet points that I’d added to my résumé to try and make those four years count for something. To count for anything. And I just couldn’t.

The effect was magnified because I worked back in time.

I’ve been at Agile Six for almost five and a half years now, and it’s been mostly very good. I’ve actually achieved things. I’ve finished projects. I’ve delivered, day after day, week after week, year after year. I’ve gained a (somewhat shaky) confidence in myself. It used to be that I was so anxious that I found it impossible to be useful in a meeting, whether there were two other people or thirty-two. Other people seek me out to answer questions on a great variety of things. Sometimes I can answer them!

And so, five and five-twelfths years into working at Agile Six, I feel like I could go to a great number of different places and succeed. It might be rough at the beginning. It might remain a challenge as long as I stay there. I might not be a staff or principal engineer, either when I start or when I finish. I might not give talks. But I believe that I would succeed.

I don’t know if I would have succeeded at Agile Six without the experience I had at Aviation International News. I joined a team of fairly junior guys, guys who were just figuring it out the best they could. They were underpaid. I started out at AIN making $72,500 a year. That was 1099, so I paid both sides of the income taxes there. That was January, 2016, and I thought it was an unfathomable amount of money at the time.

What was more valuable, though, was that I had genuine relationships there. Outside of my wife and my son, my two best friends in the world are the two guys I worked with closely at AIN, Mike Giaimo and Ryan Koch. They were polite, decent, and honest, and I learned to feel safe with them. They took interest in what I was doing, and I took interest in what they were doing. I learned a lot from both of them. And it was the first time since I was in the Navy that I felt like I was working as part of a genuine team.

And that helped me flourish as an engineer. I made my share of mistakes there. I wandered into my share of boondoggles. I wasted time and money. But I learned, and I helped my coworkers improve, and they helped me improve. I left, and they left shortly after, to bigger and better paying roles, and I’m immensely proud of that - I feel like I helped them aim higher, and I know they helped me to do the same.

Because when I started there, I was in the absolute pits.

I won’t mention names here because this wasn’t a positive experience for anyone involved.

I was a 1099 employee for a small company based in Germany. They made a desktop document database app, a specialized web browser, a couple other things. I was paid $42,500 a year to start. It was January of 2012. I was 31 years old, and I had just gotten my bachelor’s in CS.

It was the only job I applied to. It was the only interview I had. When they offered, I accepted, and I was happy with that $42K.

I didn’t have anything like real guidance. Never paired with anyone. Never had anything more social than the shared phpBB and a conference call every two weeks on Tuesday that I had to walk down the street to get reception for. I was told, hey, make a mobile web browser on iOS. So I started writing code, and if it compiled I celebrated, and if it didn’t compile I tried to get it to compile. I didn’t know about automated testing. When I built something, I did the same manual steps every time. I worked off a checklist. I didn’t have a Git repository.

I was aware immediately that there was a massive mount of internal tension in the company. Rob, the other North American engineer (he was in Canada), was running behind on delivering his synchronization plugin. It had been under development for a few months and was still very buggy and prone to crashing. Then Rob announced he was leaving, to go work at GitHub, and I was told that the browser was on hold and I was taking over development of Rob’s sync plugin.

This was intensely frightening to me. I wasn’t prepared for this. Rob’s code used manual memory management and a lot of techniques that I was new to, e.g. futures. It was incredibly complicated code. It had no unit tests or integration tests.

So Rob and I started working together, him onboarding me. This is basically the only mentoring relationship I had there. He taught me how to use git-flow, which at the time was kind of the established way of working with Git. He taught me how to package builds. He apologized over and over again for the fact that there was no test coverage. And each day, we would package up a beta build of the plugin for people to test. And the next day, we would log in to find the same complaints: crashes, data loss. Rob’s end date was coming closer and closer. This was, I think, about a month or two in. My first job.

I started hyperventilating a lot before work and before those dreaded biweekly phone calls. Bitterness and coldness occasionally becoming open anger. Sometimes I screamed into a pillow that I kept next to my desk. I kept working because I couldn’t start over - I had no place to begin.

I told Rob that I wasn’t prepared for it. I told everyone I wasn’t prepared for it. I wasn’t ready, however much I had learned in the past few weeks. Then Rob left, and the sync system still wasn’t released, and it was all my problem.

I don’t know how I got that damned thing out the door. I did not at any point understand it well enough to work effectively on it. I never implemented all of the unit tests or integration tests that were needed, even if I’d known how to do that. At some point, it just seemed like it worked well enough. It stopped losing people’s data. It didn’t crash. It had a massive number of other problems, but those wouldn’t all come up at once.

Finally, in, I think, September or October of that year… or was it the next year? that it came out. Big fanfare at first.

Then big problems in the ticket system. Each day I’d come in and there would be complaints. I’d try to sort them out as best I could. Sometimes I could help. Sometimes I couldn’t. Sometimes people were just furious, and I couldn’t blame them. I don’t think we ever lost data, but I sure ruined some people’s days. (If you ran my Sync software back then, I apologize. You didn’t deserve that.)

The only thing I could think of to do was to replace it with something. Right? Build a second system. Sync 2. Something that would solve all of the problems of the first system. Something that I could consider a victory.

The problem was that I hadn’t learned anything. I was still trying to develop a sync system that hung off the flagship application like a tick. I was never allowed to see the source code to the actual application. Heaven forbid. I hadn’t learned anything about project management. They didn’t use tickets. There wasn’t any kind of project development methodology in evidence anywhere in the company. The CTO and other programmer didn’t see the value of unit tests. There was never any idea of taking problems and decomposing them into smaller problems. There was no mentoring. Just… why aren’t you done yet?

Surprisingly, Sync 2 was not a success. I never got it out the door at all. I had a semi-working, semi-functional sync system by the end, but it was slow and used up tons of memory. I received a call after I posted that demo saying, essentially, that the CTO had developed a working sync technology and they no longer had need for my services. I was fired.

There were contributing factors - we were developing the sync system with the idea of a cloud service hosted on AWS in addition to Box.com, Dropbox, WebDAV, and peer-to-peer syncing. The contractors who developed the system did a good job, but there was no real server, and it was up to my client application to maintain the integrity of the cloud system. I don’t know if that mattered, really, but it was another thing that made developing that technology harder.

That was four years. The first four years of my career. I was so broken and defeated by the end that I didn’t even know if I wanted to be a programmer anymore. I looked at other jobs. I remember being on the verge of tears when I was rejected for a non-engineering role (it was because I lived in Mississippi, and the company was not set up for payroll in that state… it really felt like the universe grinding me into dust with its heel).

I just don’t know how to treat those four years, or how to think about them.

So I think it’s worth thinking about what I would change and in what order, to find the path of least resistance.

First is that I need to go to bed sooner and wake up at a reasonable hour. That enables pretty much everything else. My ideal schedule notwithstanding, right now I need to work more in the 7-to-3 range right now for familial compatibility.

When I’m waking up earlier in the morning, that frees up, say, two hours. Let’s say, 6AM, which is probably as early as I’ll ever care to go. Second could be a half-hour of low-impact exercise (e.g. neighborhood walk or rowing).

Third could be a decent breakfast, if I choose to eat one. A few mornings like that and I bet I’ll have a better appetite. And this is an excuse to read the headlines, but only actual event summaries, no commentary. AP, Reuters, BBC, etc. No social media.

Fourth is painting miniatures again. It’s a good use of the time between when I finish work and when Sara gets home. If John has homework, I’ll help him. If not, I’ll paint miniatures. This is a good time for language practice.

Fifth is going back to MathAcademy. This might mean resetting some progress to do it right. Don’t have a daily goal, don’t pay attention to the competition “feature”. Turn it off if possible. Learn what you can each day. Stop when bored or frustrated. This might be a good lunchtime activity.

Sixth is going back to LeetCode. I’ve done a decent amount on there but have gotten out of the habit. As absurd as coding interviews are, I do believe there’s genuine intrinsic and extrinsic value to practicing LeetCode questions. This is something I can do in the evenings with my wife, since that’s an alternation of stimulation/concentration and break time where I get up, get a drink of water, use the bathroom, etc. I can just use her to cue me on when to take a break.

Seventh is aiming the majority of the remaining time, whatever possible, at reading books. Big picture overview books are actually what I want right now, not specifics, nothing too dry. I need to build a large number of basic vocabularies very quickly. I need to be conversant across many different subjects, learn their broad strokes. Then as I drill in, prior experience will help me keep up with the information flow more easily.

If I do just those seven things, which I believe I could adopt in seven consecutive weeks, then I think I would be well on my way to living the kind of life I want to live. It addresses my health, my learning, and my ambition.

I want to achieve intellectual power. I want to translate that into financial power and social power.

I want to be someone who is well-known in the circles I respect. Not as an amusing character. Not a concerning wacko. But as a serious person, an critical thinker, a creative artist, a solid engineer, an adventurous scientist, a self-made man. Someone who led an undistinguished life for 45 or 50 years and flourished over the course of the next twenty or thirty.

I don’t want to lecture. I don’t want to teach or do TED Talks. I hadn’t considered a YouTube channel; a daily journal could actually be incredibly good for me. It would be embarrassing to document myself as I really am. Truly humiliating. It is not nearly as humiliating as it would be to stop making videos that show progress. It would indicate clearly where I gave up. Where I decided to settle. And the fact that it’s public, that it’s raw, even if nobody else watches it, that I know they could is absolutely mortifying.

I guess that’s somewhat similar to this blog.

Tear down the wall, right?

Just practice radical honesty.

I would like to be a part of conversations. A part of teams working on challenging problems. A contributor to science. I want to discuss ideas like real-world obstacles to artificial general intelligence, being fully conversant in all necessary details. I want to be able to read and visualize the equations in scientific papers. I want to build an ever-expanding web of knowledge that keeps abreast of all new developments and their implications.

I would like to be physically strong. I love the belief (from Plato?) that it is a sin to grow old without having experienced the beauty and power of one’s body. I have at times approached this. I would like to improve to the degree possible without causing further damage. I would like to run again, which means that I need to take very good care of my knees and hips. I need to work out multiple times a day, mostly very lightly, to minimize damage.

I would like to paint miniatures. I don’t know if I care about keeping them around. I might paint them and throw them out. I like making a little dude, I guess, something that tricks the eye a little bit, that makes something fantastic real. I think that’s why I’m so impressed with people who get skin really well, or who use colors in imaginative and shocking ways, etc.

Some of the areas that attract my interest are artificial intelligence, artificial life, psychedelics, culture and religion and history, ethics and aesthetics, language, biology… maybe any subject that is of a higher tier of abstraction than geology, perhaps. I don’t have a substantial amount of interest in physics or chemistry or astronomy.

I want to be informed generally. Movies, music, science, culture. I want to be a breezy conversationalist. I want to ask people their opinions about things, and watch how they react, and learn from that. I want to understand people across the spectra of humanity. And people like talking about themselves. They also clearly indicate to me their degree of openness to me by doing so.

I want to be the one in charge in an interview. Any interview. I will find answers to my questions, and I will answer yours if it suits my purposes. I don’t intend to be blown off again. That doesn’t mean randomly spraying applications; it means process and precision and investment. This doesn’t mean coldness or meanness, but cordiality at worst and genuine warmth at best.

Always be looking for a job, never accepting an offer. Stay sharp, evolve, learn what they’re looking for, learn where their shortcomings are, learn the contradictions present in the organization, learn the issues with the process and exploit them ruthlessly. Always leave them wanting more. I have no reason to move, yet.

Learn a musical instrument. Maybe one we’ve tried already. Develop a new relationship with the instrument. Play first, music second. Let curiosity about technique guide learning. Don’t think in terms of notation but the sounds and the feelings. Practice daily, but without a plan.

Be fresh and sharp, like a man half my age… those with whom I’m competing. Use my wisdom. Use my patience. Use my breadth and depth. Dress better. Get better haircuts. Bathe more carefully. Brush my teeth better.

Instead of reading one book at a time, read twenty. A little of each a day. Perhaps a paragraph or two, or until you find something worth pondering. Let that go to the background. Next book, learn something, explain or describe or relate it. Repeat.

$DEITY willing, this is my mid-life crisis. But it might be my 2/3-life crisis, or 7/8, or more.

In the past couple of posts, I’ve focused on what I want to do, what I want my life to be like, what I’d like to work on. I’d like to focus now on what I don’t want my life to be like.

I don’t want my wife or son to resent, dislike, fear, or hate me. I don’t want to hurt them. I want to set them up for success as much as possible, and to limit the maximum amount of harm that they can experience. I don’t want to depend on them.

I don’t want to live in fear, either from political forces, financial troubles, health troubles, relationship issues, etc.

I don’t want to die alone.

I don’t want to die with more significant regrets. I already have regrets. Massive regrets. I have not lived my life the way I would have liked. That’s not to say my life has gone poorly; I’ve been granted immense good fortune at every turn. My behavior has been poor. My self-regulation has been poor. At nearly every junction, large or small, I’ve made a suboptimal choice.

I don’t want my death to be a burden. I don’t want to leave unpaid bills. I don’t want to leave uncertainty.

I don’t want to leave a mess. I don’t want people to go through my things and think “God, what is all this junk?” or “Aww, that’s cute, he wanted to get into $HOBBY but never got around to it” or “Uh-huh, The Art of Computer Programming, never opened.” I don’t want to leave behind things that I bought but never used.

I don’t want to be bored. Just another guy on the couch, doomscrolling.

I don’t want to be fat. I don’t like it. I don’t want to breath heavily when I go up the stairs. I don’t want to struggle to get in and out of cars.

I don’t want to be crippled by rheumatoid arthritis. I don’t want to let it imprison or defeat me. I don’t want to let it hold me immobile. I don’t want it to take what freedom of movement I have.

I have to face the fact that I only have one life, and it’s nearing its end. My wife only has one life. I want her to enjoy it. My son only has one life. I want him to live well. The best thing I can do for my wife is enjoy being with her, and make her enjoy being with me. The best thing I can do for my son is show him how to live well.

I don’t know why I’m restless. But pondering that, I realize I don’t know if I’m restless. Or maybe I’ve just been trying to escape a long project, the kind that has traumatized me in the past.

My first job was remote. I was just out of college - but keep in mind that because of various issues, I was 31 by the time I got my bachelor’s degree. I spent my twenties in and out of college, the Navy, unemployed, and finally, blissfully, back into college.

I’m trying to think back and remember why precisely I chose to be a software engineer. As I recall, I was considering a few different options. One of them was being an architect. I’ve never had any interest in architecture, so that might help you to understand that I am not a born computer scientist. I didn’t program much as a teenager - not much beyond scripts and QBASIC, and much of that didn’t work. I occasionally wrote in Inform 6.

I graduated high school thinking about being a musician and a poet and an actor. I went to college, went to another college, dropped out and joined the Navy, went back to college, went to a different college, dropped out, went to a different college and graduated. Along the way I considered music theory and composition, teaching high school band, creative writing, philosophy, English Literature, psychology, and sociology. Maybe others that I can’t recall now.

Never once, until my last trip through college, did computer science seriously cross my mind. Nor mathematics.

I think the fact is that I thought I was “good with computers” and that I might turn that into a way of making money. I was also interested in creating a software system capable of planning and writing the novels that I wanted to write. My final semester of college (this is how naïve I was), I applied for a job with the tiny company that made the product that most closely resembled what I had wanted to build for myself, a desktop document database system. It was the only job I applied for. I got the job.

(I say a software system, not an app, not an application, because I was aware that what I wanted might require a suite of applications, or an ecosystem, something closer to Microsoft Office than Microsoft Word. Or that it was likely to have a content server, a map server, photo galleries with inspiration, etc etc etc. It wasn’t just a personal database.)

But what do I want now? I suppose I enjoy taking imaginary systems and making them real. I don’t always know how to do that - I think I’m weak in DSA…

(I actually gotta get back to doing LeetCode, because regardless of how stupid I think interviewing with it is, and how unlike the day-to-day work of software engineering it is, it’s still a lovely way to have a problem with defined objectives, clear requirements, and rapid feedback cycles.)

…but I enjoy the process just the same.

I was intoxicated at first by infrastructure engineering for the same reason that I was intoxicated at first with large language models: it seemed like I could do so much more. I could spin up an entire computer instead of a text window. I could specify an entire computer network and all of its services. It reflected a lot of the work I had done with MediaWiki, Drupal, etc, to manage my ideas (when I wasn’t looking at desktop software).

Today, I feel strongly that I should stop using LLMs to do things. At work, whatever, I’m mandated. But in my free time? No. I must not. I’ve noticed the effects of it. I’ve noticed that I use it to commit and push code now, instead of writing commit messages myself (admittedly, I’ve rarely been conscientious about this).

The industry is going to go the way it’s going to go. That’s fine. If I’m an employee, I’ll do what I’m paid to do, and how they want it done, and try to do so to the best of my ability. But I don’t think I’m serving my best interests, or my employer’s best interests, by using it outside of work to replace my own efforts.

But what do I want to write? I suppose I see everything as mathematics now. Which might seem weird, given I’m not traditionally strong in mathematics. I passed calculus… on my third or fourth attempt. I’ve never been good at doing the work with mathematics. I’ve never put in the effort for very long.

And I guess I do see everything as writing, too. So writing mathematics?

I had a painful experience a year ago, when I applied for a job and was rejected… at the CEO approval stage. And only once I’d gotten really excited about that job. The company had been formed on the premise of a scientific paper, about 107 pages with citations, discussing the combination of simulations and artificial intelligence and how useful they are in the development of new scientific and technical knowledge. And I wanted to help build these systems, so much so that I was crushed when they rejected me.

This became, I guess, a part of the story that I tell myself. It became a bit of a mission to “show them” that they’d missed out on someone, a person that had potential along these lines. I bought a lot of books about machine learning and MLOps and read a couple of them. I bought a lot of books about complexity science and read a couple of them. But then I bought a lot of books about organizational dynamics and social psychology and read a couple of them. And then I bought a few books about cybernetics and have started reading…

There’s a deep question left unresolved in my mind: how good could I be at mathematics? Mathematics, the language underlying seemingly all of reality. The thing that I’d hardly passed a class on after the fourth grade. I get discouraged when it gets hard. A really smart person, I think, finds math easy. They get the Calculus homework and they run through it, from beginning to end. They always remember the Chain Rule.

I don’t remember Calculus II. I remember something about series. Literally that’s it. I know that I learned about integrals in Calculus I. And yet I got a B in that class. I took Discrete after that, and passed, though I remember not putting in much effort and having trouble following the ideas. I took a Statistics class - the kind that requires calculus - and it was about the hardest thing I’d ever done. I sat in the first row, attended every class, and did each homework assignment three times: when it was assigned, when we had a unit test, and again before the final. If I recall correctly, I got a C+. I remember nothing else from that class. I’m not positive I earned that grade.

So I’m anxious about math. It’s tender for me. I think it’s incredibly important. But I’m afraid to put too much energy into it, lest it turn out that I have good tools and significant privilege and still just can’t learn anything more complicated than differential rules.

I need to go back to MathAcademy. And probably repeat all that I’ve done of Mathematical Foundations II, and possibly Mathematical Foundations I. That seems crazy, and it would mean redoing months of work, but I think I’ve half-assed it in places. I put it off, I criticized the tools, I half-assed it, and then… what do you know, it got hard. I stopped learning as quickly. It was less fun. And so I’d put it off, criticize the tools, and half-ass it more.

But there is no royal road to Geometry, and mathematics is the most important thing I could be learning right now. I need to master mathematics, up to and including things like diff eq, lattice theory, topology, etc. I need to go as far as math can go and possibly push it further. I need to do that to be able to express rigorously the things I imagine.

But beyond that, I want to understand mathematics. It’s exactly the kind of thinking that I want to be doing. It’s the kind of thinking I would feel privileged to be able to do. It would be even more incredible to be paid well for it.

My job rejection led me, indirectly, to the Santa Fe Institute, and complexity science. I think that interests me. I like the ideas of symmetries of behavior, patterns that hold true at multiple scales. I like it for the same reason I like LeetCode; it’s a beautiful idea. I like concision, symmetry, minimalism. I would love to have ideas like those behind ant colony algorithms, boids, etc.

I’d like to observe nature and see the mathematics behind it. The calculus of the way a tree blows in the breeze. The cellular automata of a snail shell. And find new phenomena and new mathematics behind it. I think it’s everywhere, right? So there must be value in describing its expression, right?

So returning to the books I wanted to write, and the software I wanted to write to help me write those books… what might that look like, now? I suppose what I wanted was something that could express every possible aspect of every possible story. Some encoding that could allow me to move laterally between genre, plot, theme, characterization, etc etc etc. A mathematics of story, story being anything that could happen in reality plus countless things that could not happen in reality plus metadata like authorial intent that we have no good reason to believe exists in reality. So a mathematics of intention and influence, perhaps.

Or if we view a story not as a simulacrum, but just as an arrangement of words composed to provoke a speculation, then the mathematics of provoking speculation? And intention and influence?

But then, we find that coming close to psychology, to the mind - a story that the human body seems to tell itself, that there’s a self with qualities and characteristics and a mortgage and rheumatoid arthritis and a wife and a kid and a blog.

The mind and what it tells itself. I guess a mathematics of consciousness.

That’s actually hilarious. Ludicrous. Ridiculous. The thing I want to solve is literally the hardest problem that any human would ever have to solve. We have multi-billion-dollar industries based on tiny, tiny portions of this problem. And there are as many problems in consciousness as grains of sand in all the beaches in the world. We don’t know much about consciousness collectively, and almost none of us knows anything individually.

I guess that’s what I would want to have done. Since I can’t just wave a wand, then I have to think further about what I want to do. In other words, how I work toward this overarching goal that I may be intellectually or chronologically incapable of solving. Perhaps I can treat that journey as the ends in itself.

What goals might contribute toward this? Well, I think perhaps I would like to get a doctorate in mathematics. That implies a mastery of the standard curricula of mathematics, up through real analysis and beyond. I don’t really know at what point they say “you can stop now, where do you want to go?” And I am not against the idea of studying more mathematics. Afraid, yes. Against, no.

I don’t think we’ve found the answer yet. Not that I’d know if we had. But regardless, I have a great amount of work to do in this area. I have books on complexity science, organizational dynamics, artificial intelligence, language, mythopoetics, mythology, religion, history, fiction, design, management, manufacturing, etc. Each of these has a tremendous amount to teach me. Each of these should be a goal. I should, as I’ve suggested, rotate through books. Parallelize, not serialize. Mark my place in each and drop it when I tire. The goal should be to stimulate thought, and if I’m bored or sleepy then my thought isn’t being stimulated and I need to find something else.

If I could do what I wanted for the rest of my life, what would I do each day?

What would I do on the last day of my life? I would like to see the ocean, I think. I would like to watch the waves crash on the rocks near Bodega Bay. I would like to smell the air and feel the bite of the cold wind or the heat of the summer sun. I would like to have my wife beside me.

What would I do the night before? I would love to eat an Italian dinner with my wife, and make love with her. If we’re still able to move at that point.

What things would I do, for all days between this one and that one?

I’d like to wake up in the light. I’d like to take a walk and see the world in the morning. I’d like to have a small breakfast, just a couple of things. I would like to read a magazine. Something new, current, to inform me and to give me ideas. I’d like to swim in a pool, perhaps, or ride a bike, and think about these things.

I’d like lunch to be a filling thing. More about comfort and pleasure, but eaten slowly. I would love to have someone to talk to, either my wife or son or someone else. Maybe I could meet a new person each day and have them over for something. I can prepare the food, serve it, have a conversation, and clean up.

Then one of two things occurs to me - I might read a book or watch a movie, or I might work. Either way, the area between lunch and the work day ends is really my peak time of day, I think. And I think that peak actually continues until fairly late, maybe 10PM or so. Probably I should be working from 2PM to 10PM or something like that if I am working a full-time job. If we’re talking about what I would choose to do, I suppose I might work from 2PM to 6PM. Either way, I see myself working until I’m no longer able to. I don’t mind that. But I do count homelabbing and side projects as work too, FWIW - anything related to my core information technology/computer science career.

Either way, we get to dinner. I would love to have dinner prepared for me and my wife. I think I’m more interested in cooking lunches and snacks and things like that than I am with cooking dinners. My wife and I inevitably get different things when eating dinner together, unless it’s pizza, and even then we still sometimes get different things. Within that, my wife and I have different preferences for a given dish - I add salt and spice liberally, she never adds anything - so cooking something that pleases us both is difficult. All this is to say that if I’m cooking for dinner, it’s likely for myself. I don’t know if I’ll ever have enough confidence in my cooking to invite someone over for dinner, but maybe. It’s a nice idea, and one worth working toward.

I like a fire and a cozy room and a movie or a book. Or a laptop and work to do. Or a miniature to paint and music to listen to. It really doesn’t matter either way, I think I can be content with that. Then a shower and off to bed.

I’d like to go to plays more often, and movies. I’d like to use the library more. I’d like to try new restaurants when they open. I’d like to go to different places, all comfortable, and live there for a few months, and meet people and eat food and work and read. I’d like to travel all over Europe, while it’s still here. I’d love to hear and learn more languages.

With the exception of Europe, it’s striking that this is all very achievable, very doable, very possible, and completely different from how I live my life now.

I have a funny relationship with writing, y’all. I wanted to be a writer (and a rock star, and a poet) in high school. I wrote a decent amount, some on typewriter, some on computer, some in notebooks. Not a tremendous amount, and none of it was any good, and unfortunately my inner critic silenced my inner artist instead of spurring them to greatness.

I think there are a number of poisonous things that a young artist can believe, and unfortunately I believed a great many of them. I believed that brilliance didn’t need to be developed; that one was born brilliant, and their abilities were largely static. This sounds perhaps a bit absurd. I don’t really know how to explain it other than to say that I didn’t really understand the concept of practice, because no one I ever knew practiced anything.

I believed that artists had to be wholly original to be legitimate. Artists whose influences were too visible in their product were compromised. There was no room for artistic conversation. I was in competition with every other writer who had ever lived. I must be the most sensitive, the most discerning, the most eloquent, the most devious, the most polished, the most amusing, the most thrilling, the most exciting, the most prolific.

I would be the writer who had been around the block, published in every small college journal and men’s magazine, whose books were printed on cheap brown paper and were thin and easily digestible. I would be the writer who blew into the literary scene as an enfant terrible, who tore through penthouses in Manhattan and clubs in Ibiza, and then became a recluse, publishing only intermittently, peering out suspiciously. I would be the writer taught in universities, in literature classes, alongside Steinbeck and Hemingway and O’Connor and Morrison. I would be the writer at red carpet events for my movies.

My visions grew grander still. I would write a fictional encyclopedia, a fictional bible, create an entire world. A valley of so many miles by so many miles, three towns, rivers, farms, streets, roads, neighborhoods, mines, museums, geology, ancient history, post-Columbian history, modern history, contemporary characters. Ongoing drama between the mayor’s family and (one of) the main character’s. Shadowy pasts. A psychedelic apocalypse. People transcending the limitations of humanity.

There’s an interesting dynamic where an idea becomes an obsession. I think creative people are in an economic or thermodynamic relationship with their ideas. You invest some time and effort into an idea, and the idea creates inspiration, I think. It inspirits us. And we can sing about the idea and about all the ornaments we can hang on the branches of that idea. It becomes a goose that lays golden eggs, a seed for a crystal.

But an idea can become malignant. It can take the time and effort you give it, and demand only more. You aren’t hanging ideas on a tree anymore. You’re just moving existing ones around, sometimes losing them, losing bits of each. The more you handle them, the more you see your fingerprints in the paint, the more smudges there are, the more of them you drop and break. Eventually you’re left with nothing but a vision of your creative spirit as a vast, empty desert, salt and wind and dust.

I was so bitter when I used ChatGPT the first time. I had spent so many hours staring at a computer screen, and that counts 1) trying to learn things, 2) trying to get the computer to do things, 3) trying to write things. Suddenly I was having a conversation with a token predicting machine that seemed to know everything I’d ever heard of and in far greater detail, that could spit out a few hundred lines of code in seconds that would take me hours to write, and that could write pages of prose that would take me days to write.

But that was already 13-14 years after I gave up on writing, though.

It was a cycle of drafting, iterating on the basic concept, without ever actually writing much prose. Trying to find a perfect structure, something crystalline and geometrical, something that revealed the order of the universe. To align things, to have a golden path that I could walk to literary greatness. With the perfect structure, my masterpiece was a fait accompli. Writing any given chapter would be like filling in the blanks. I couldn’t possibly fuck it up with my weak prose.

Each time I rethought everything, I got less and less out. The idea had shot its wad, as my grandmother would say.