I’ve written before about my computing origin story, and it’s still true. These boxes of sand, they can change the world, but only with the right programming. On its own, a computer isn’t even a fancy calculator. With software, it can be whatever it needs to be.

Who wouldn’t want to write software that could change someone’s life for the better? It’s still the dream.

I moved into management when I remembered that programming was a means to an end. I had become enamoured with programming as its own craft, but I knew I was never going to be that one programmer that you’d give an impossible problem to, only to have them come back two weeks later with a performant solution and a conference talk ready for you to help them edit. My interests are too varied for that.

I realised that I would be better placed to make computers do amazing things as a translator. As the engineer in the team who could speak Engineer, but also Product, Test, Metrics, Sales, Marketing, Economics, HR, Psychology, Productivity, Science, Design, Philosophy, and Pragmatism. I was the person who could go into a room of disagreement and come out with a plan that everyone got behind, not because I wrote it, but because I could help people cut away the distractions and focus on what they were all there for in the first place: getting things done.

I love aligning people, understanding their motivations and pointing them in the right direction. Watching as their enthusiasm for the work grows as we build and ship. Hearing their excitement when the thing succeeds, or helping them rediscover their grit when the thing fails. It’s a huge part of who I am.

But now? Now I have a legion of tireless engineers at my disposal. They produce pretty good code, exactly to spec, and their enthusiasm is boundless. They work all day, every day, and will never need an encouraging word or a frank conversation that makes us both feel rough for a few days afterwards. And not just engineers, but designers, product managers, security experts, sales and marketing teams, finance departments, operations. All there, ready for me to point them at the problem and to go get it.

And I stand here, with this limitless potential, paralysed.

See, every new disruptive technology has this effect: it shifts things in ways that are both obvious and subtle. The obvious change that these agents bring is that the cost of production – of simply producing things – rapidly drops towards zero. That’s obvious, and scary if you’re in the business of producing things in order to pay the mortgage.

But the subtle effect is that everyone gets to level up.

What does that even mean? Well, for those that haven’t made the transition before, it can feel utterly terrifying. It can make these coding agents seem new and weird and alien and like nothing we’ve ever seen before. And we get the fear. We try to find new words for them, and build out new tools, and treat them like new problems to solve.

But I think that’s missing the point. These are not new problems to solve, they’re just problems from the level above where we’re each used to sitting. Engineers don’t have to invent new ways to figure out how to use these agent swarms, they just now get to think like managers: what’s in the way of these swarms delivering what they need to? What’s unclear? How do I maintain quality when I can’t review all the code? How do I keep track of more work than I can store in my head? How do I think ahead to what problems we might be creating in the process of building?

These are manager problems, and we already have solutions, in practices, processes, and tooling.

And what about my problems. My paralysis? See, I’m very used to building a shared understanding of a problem, then working to produce a deliverable plan that everyone gets excited about. I’m not used to pulling those problems out of thin air (at least not the product problems).

My comfort zone is walking into the room, helping people figure out what problems they need to solve to win, and then rallying the troops to solve those problems with excitement and as much fun as possible.

But now I need to move to that level. I get to be the room with the problems.

Not everything has to be a long meditation on creativity or what it means to be productive while maintaining a healthy balance between what you need to do and what you want to do. Sometimes, I’m allowed to just point you at stuff I like.

Like You are listening to, a wonderful only-the-internet-could-do-this old-school mashup website combining a curated, chilled playlist (rotates regularly) with live public feeds from air traffic control, shipping forecasts, transit authorities. Hauntingly beautiful and a reminder that even on the internet, stuff is happening right now that may never happen again.

Identity is weird. I’m a writer, but saying that casts this weird spell on people. I think the spell comes from a good place, a place of compassion and curiosity.

But it’s a deadly spell.

“So… how’s the writing going?”

Immediately I feel a fraud. How is the writing going? I haven’t published to the blog for ages, haven’t sat down to actually put an essay together for weeks.

No, months.

Who says I’m a writer?

“Oh, you know. It’s going”

The deflation is palpable. They’re rooting for me, they want to hear about the progress on the novel, or the collection of short stories, or the self-help book, or the… something. Anything.

How can you be a writer without something to show for it?

And yes, that’s a point of view, but what they’re really looking for is whether my writing has become work yet. Am I cranking out words every day? Am I doing the reps? Am I putting in the hours? Do I have projects and goals? Am I going to do NaNoWriMo this year?

What’s the plan?

And there is no plan. Writing is not a job, for me. Making it a job will make it lesser. Neither is it a hobby.

It’s just a thing that makes me feel at home.

Sometimes in my life, I don’t feel at home. I feel washed away, adrift on someone else’s idea of what my life should be. In some seasons, that’s necessary, a journey away from home that brings with it some worthwhile reward, or some unquantifiable progress in some part of my life.

In others, it’s just exile. A meaningless drift away from what I hold dear. A drift of inches and degrees, where I’m suddenly in the middle of the ocean, lost, before I notice I’ve left the shore.

How does one become lost at sea? Two ways, slowly then suddenly.

Writing can sometimes be a lifeline in those drifts, a way to set sail for home, to remember the route, to follow the compass and spy land. But sometimes I’m so far from home that the writing itself feels alien and unfamiliar.

And there is nothing I’d rather be doing less in those moments than writing.

Does that mean I’m no longer a writer? Do we stop breathing between inhale and exhale?

Writing is home. Sometimes it can bring me home, but sometimes I have to fight to return before the writing can come.

And so I fight.

And then, I write.

I don’t think back on my early years with a computer anywhere near enough. Sam Henri Gold’s excellent article on the MacBook Neo made me need to do just that.

My mum got me a Commodore VIC-20 from the local Fine Fare supermarket when I was probably five. We were a Tomorrow’s World household and watched Star Trek repeats over dinner. She knew what was up.

Computers were going to eat the world, and her kid was going to be on the right end of the fork.

I spent hours on that machine. Days, weeks, months, and years. It came with a manual on how to program stuff in BASIC. We couldn’t afford a tape drive with it (legend has it that Mum actually bought the Vic-20 at a massive discount as they were clearing stock and nobody was buying them) so we only had a few cartridges with games.

Of course I played games. An asteroids clone, a space invaders clone, and a game of chess that my dad ended up getting into an argument with.

True story: He came down from spending an hour playing the computer at chess absolutely fuming. The thing cheated! He was educated enough to know that computers aren’t sentient, they can’t “cheat”, but there was obviously a bug! It had made an illegal move and, worse, it had subsequently beaten Dad.

This led him down a rabbit hole where he learned (and then taught me, lest I be caught out by a sneaky silicon opponent in future) about an obscure move called en passant, whereby a pawn can take a piece by moving past it diagonally, but only on a few particular squares on the board.

We were learning.

A few years later I got a hand-me-down Sinclair ZX Spectrum with a grand total of 48k of RAM. This thing had a tape drive, it had a joystick interface, and it had a carrier bag full of games. It also had a dodgy powerpack and, as it turned out, a broken audio cable that would disconnect randomly, often in the final ten seconds of a ten minute game load. Fun.

I had graduated to yelling at the computer now, following in my father’s footsteps.

The machine also came with four hefty volumes of INPUT magazine. Holy crap were those some insanely cool magazines. No internet, no disks, code was stored on tape and you crossed your fingers that they would load. The most reliable way to distribute code, to learn, was to buy magazines with actual code listings in them, and type them in.

And pray you hadn’t made a typo in the 1,000 lines you’d just entered.

You’d hit run and… it worked! I recruited whichever of Mum or Dad seemed to have most reserves of patience left to read out line after line of code to me as I typed. I learned to run the code every so often, to verify that what we’d done so far was correct before moving on. I learned what subroutines were, how to pass variables.

Mostly I learned to ensure my code lines were numbered in multiples of ten so that I could add lines between them later if needed.

I discovered that I could create sprites and assign them to letter keys on the keyboard. I could animate them to move across the screen. I made pretend games that drove cars around the screen, racing each other with no controls.

I got my dad to bring home reams of graph paper from his work (why they had graph paper I have no idea, he was a civil servant in the DHSS), but he brought it home by the box. I spent hours, days drawing out images on the graph paper then writing code to set up little 8x8 grids to print to the screen.

I was taking my art and putting it on the television and it was amazing. I was nine. For my tenth birthday I remember getting a card from an aunt that had a BASIC computer program in it. I read it and saw that it was making sounds, could see the rough sequence and “up and down-ness” of the notes, and guessed it was going to play Happy Birthday. I was right.

My parents were both impressed and thought I might be a witch.

The following year they had scrimped and saved and bought me an Atari ST. I was intimidated. This felt like a proper computer. It was certainly the first computer I owned that came in a box that looked new. Or actually a box at all.

This wasn’t something I could program easily. The previous machines presented essentially a REPL as the primary interface: type in lines of BASIC code to make the computer do things. The ST was different. It had a windowing system, a mouse, icons! You launched programs to make it do things, and then quit them to return to the desktop, instead of just pulling the power code when you were done.

It was new. It was exciting. It ran games, for sure (I lost so many months to Buggy Boy and Operation Wolf), but also art packages, music trackers (TCB Tracker for life), samplers, desktop publishing software, vector art creators… so much stuff.

It was my main computer for eight years. Eight years. I upgraded the RAM, and even ended up hacking together a hard drive that just about mostly worked most of the time.

My favourite program, though, was Protext. Looking back, it was “just” a text editor, but it transformed my computer into a place for writing. A place where I could draft my stories and my reports for school. A place where I could put down my thoughts in a way that stripped them back to the purest essence. Devoid of thinking about presentation, there were no fonts to pick from, no text sizes, no headers, nothing. Just me, a screen, my fingers, and my ideas.

No wonder I still use Neovim.

I’d write in Protext, then save the file out and import it to Timeworks for getting ready to print out for handing in to school. I’d add in pictures I’d made, charts I’d compiled, and place the stapled output neatly on a pile alongside everyone else’s handwritten efforts, lousy with Tipp-ex and smudged ink.

I suspect people hated me. But I loved it. I loved being able to use this box of wonders to create. And that desire fed the games I played. I graduated from Buggy Boy and Operation Wolf to Frontier, a game where the story happened as much outside the game as inside it. I was Commander Harwood, rogue smuggler and bounty hunter, occasional arms dealer but also sympathetic to the explorers of the cosmos, happy to ferry them to remote star systems. For a price. Alongside the mechanics of the game (fly to spaceport, pick up jobs and cargo, map a course to the next star system, fly to spaceport, complete jobs, fight off pirates), I kept a diary of Harwood’s thoughts as he travelled the galaxy. What was he doing on those long flights? At one point I had him playing a game of chess with a far off friend, correspondence chess during long haul flights. I’d track the moves on a real board next to the computer, and then on breaks I’d set up the board in Colossus Chess X and get the computer to make the next move.

Harwood was a pretty lousy player, but occasionally he’d get lucky: I’d set the computer skill to -1.

After I finished school and went to university I bought myself a PC, then a Mac, and, well, here I am. As I got access to more and more finance, I found myself able to choose whatever computer best fit what I wanted to do. I was able to experiment with what fit, buying something, trying it out for a year, then exchanging it for something else.

As a kid, I worked with what I had. And frankly, I think I had more fun.

It’s easy to read this as a rose-tinted view of the shift. This stuff used to be fun, now it’s work. Nostalgia is a hell of a drug.

But I don’t think that’s it at all. Or at least not the interesting part.

Desktop publishing on the ST was slow. Frontier ran at maybe 10 frames a second on a good day with empty space in front of you. Those floppy drives could take minutes to save or load a long essay. Creating graphics “larger” than the display (essential for print work to not look blocky) would make the computer so hot you could hear the plastic case clicking and creaking as it warped. It was objectively horrible, and I was well aware of it at the time.

But I never saw the solution as I need a new machine. I might have read the magazines and wondered what an Atari TT or a Falcon might have been capable of, but they were pipe dreams and the magazines knew it. And really, I never actually wanted a new machine. I wanted to explore every nook and cranny of the box I had in front of me, to see what I could still shake loose from it.

Reading Sam’s article, there I was. A part of me I’d forgotten about. The part of me that didn’t give a shit about the computer’s limitations, as fixed as granite as they were against my meagre budget. But I didn’t care. All I cared about was making it do what I wanted, create the output I could see in my head. And I didn’t care if it was slow, or messy, or risked burning a literal hole in my desk by leaving it running overnight. All I cared about was making the thing do the thing.

And it was glorious.

These days, I often think of computers as tools, but I used to see them as treasure boxes.

Treasure boxes are buried, hidden. They require effort to unearth. Who knows if they’re even there, or if they’ll be empty when you finally haul them up. But you dig anyway.

You dig because digging is fun, the hunt is fun, and sometimes…

Sometimes you get the gold.

My writing process is typically thus:

1. I will bang out a stream of consciousness first draft, getting everything that I think I want to say out of my head and into text;
2. I will then read over that first draft and realise it’s not right. The fact is that while I may have had an idea of what I wanted to say, the piece iteslf has no fucking clue what it’s about. In fact, it’s garbage. I suck and should never write again;
3. I must now leave it for a day. Or a week. Or a month. And then, if I feel up to it, I will come back;
4. Usually I select-all-and-delete, then try and write a plan to actually figure out what the piece was really about;
5. That “plan” ends up being the final draft

The mistake I often make, which leads to the worst pieces, is to think I can jump directly to 4, skipping the first three stages on the way. Just write out what I want to say. How hard can it be?

But the point is that the writing doesn’t happen in stage 4. It happens in stage 3. And you can’t skip to stage 3, since it’s a reaction to the first two stages.

When you jump to 4, you end up in 2 for eternity.

You can’t plan this stuff, you can only write it.