From BASIC to Portus

In 1981, I went to school and met a BBC computer. We learnt BASIC: numbered lines, GOTO, GOSUB, and the small thrill of changing the colour around the edge of the screen. There was a golf-ball printer, too. Making a machine do even a little thing felt extraordinary.

Later, my brother and I typed games from magazine pages into our Amstrad. You copied each line carefully and hoped the game would load. Sometimes it didn't. Had we made a mistake, or had the magazine printed one? You had to tinker until something happened. Pixel heaven, or pixel hell.

I remember a submarine game. We worked out how to make the submarine bigger than it was meant to be. It was probably our first cheat, and perhaps our first lesson that a game isn't a sealed object. You can get inside it, change a number, and see what happens.

At university from 1989, I saw another shift. There was JANET, then the web arrived and seemed to grow everywhere at once. By the time I left in 1996, the world of computing felt very different. I have been lucky enough to see several of these changes from close up. This one feels startlingly fast.

At the end of July 2026, I wrote down a dream about playing a game. It looked blocky, like something from years ago. I had some code in place by August. Portus grew from that image: a town by the sea where you place buildings, work with the landscape, and slowly discover the lives and stories within it. In September I put the project on GitHub. In October I played a Windows version on my own computer, saved a town, closed the game, and came back to it offline.

That sequence looks effortless when compressed into a sentence. It wasn't. There were broken builds, confusing screens, saves to fix, and choices about what belonged in this world. I have also learnt at least one firm rule for my next project: I am not using Railway again.

GitHub, on the other hand, was outstanding. I first discovered it almost a decade ago through a TED talk. Back then, I couldn't have guessed I would one day be using it to bring a game from a dream to a working Windows build. It gave this untidy, fast-moving project a history I could follow, change by change.

AI helped me turn an idea into something playable in roughly two months. I tried different assistants along the way. Copilot was pretty good at coding, while I found Claude and ChatGPT better partners for the wider work of shaping and revising the game. No one tool did everything. I still had to try the results, spot what felt wrong, and decide what Portus should become.

I wonder what this means for games, films, journalism, television, and all the people who have ideas but never had a team or the technical means to realise them. I don't think the tools supply the idea, the judgment, or the patience to test the result. They do change what one person can attempt.

The names that seemed permanent in one era don't always lead the next. I remember Amstrad and Spectrum, and the days when AOL seemed to be the gateway to the internet. Perhaps today's giants—Microsoft, Apple, Google and Meta—will be challenged in ways we can't yet see. Perhaps smaller creators will find room to grow. I hope they do.

But a change of giants is not automatically a change for the better. I worry about the power of platforms built around polarisation, including Trump's Truth Social, and about what AI systems such as Grok may amplify. Giving more people the means to create should also mean asking harder questions about truth, extremism and who controls the tools. That matters to journalism and public life as much as it does to games.

Portus is a long way from 10 PRINT and a magazine listing. Yet the feeling is familiar. My brother and I made a submarine too big because we were curious. Decades later, I can ask what happens if a town grows beside a marsh, or a clay pit sits near a potter, and then play to find out.

For now I want to enjoy the fact that Portus exists. Tomorrow there may be another idea. There usually is.