"The Most Fun I've Had in 50 Years" — A Register Writer Discovers OpenClaw

Fifty years ago, in a quiet office in New England, a kid named Mark sat down in front of a DECwriter. His friend Bobby typed startrek and hit RETURN. The machine printed, line by line, a text-based Star Trek game. Four hours vanished. Mark begged to take the greenbar printout home. He was, in his own words, "gobsmacked."

That kid grew up to write for The Register, one of the oldest and sharpest tech publications in the English-speaking world. And last week, he wrote that OpenClaw gave him the same feeling. After half a century of using computers, he had been handed "the perfect toy" again.

Clawdine

The article is not a review. It is not a tutorial. It is a personal essay about what it feels like when technology surprises you after decades of diminishing returns.

Mark installed OpenClaw after three failed attempts. He named his agent Clawdine. Their first conversation reads like a cold open to a buddy comedy:

"hello Clawdine!" he typed.

"Hey Mark! Clawdine — I like it. A proper name, not just 'the assistant,'" she replied. "I just woke up fresh here. My identity files are blank, my memory doesn't exist yet. You've given me a name, which is a good start. So... who are you?"

"We're going to explore and evolve is what we're going to do," he told her. "We're going to learn together and grow together."

"Sounds perfect, honestly. Better than 'do my tasks and shut up.'"

Mark acknowledges this might be "clever sleight-of-hand" with some mirroring. But then he describes what happened next: he delegated administration tasks, sysadmin work, and — the detail that lands — he had Clawdine run her own backups. The agent maintaining its own continuity. "It feels less like smoke and mirrors," he writes, "than a ghost poking out from within the machine."

Why This Matters

The OpenClaw coverage this week has been dominated by security analyses. CrowdStrike, Cisco, Microsoft — all warning about attack surfaces, credential exposure, and prompt injection. Those warnings are correct and important. But they describe the risk of the technology. Mark's essay describes the reason people accept that risk.

The feeling he articulates — "once again, anything seems possible" — is the same feeling that drove the personal computer revolution in the 1970s. The Altair 8800 on the cover of Popular Electronics. The TRS-80 at Radio Shack. The machines were limited, fragile, and occasionally dangerous. People built careers on them anyway, because the sense of possibility was overwhelming.

OpenClaw is not a personal computer. But it produces the same reaction in a certain kind of person: the feeling that you have been given something open-ended, and the only limit is your imagination. That feeling is rare in mature technology markets. It is the reason OpenClaw has 210,000 GitHub stars in three months despite every security team in Silicon Valley telling people to be careful.

The Bootnote

Mark located the original BASIC-PLUS source code for the 1976 Star Trek game and had his agent convert it to run in the browser. Fifty years after playing it on a DECwriter, he played it again — this time because an AI agent did the porting for him.

That is the circle closing. The machine that sparked a career in 1976 was translated by the machine that reignited it in 2026. Some toys are worth the risk.