This is his explanation for the ambition he conceived while still a child: to create virtual worlds. When he arrived at the University of Essex and met Roy Trubshaw, the dream was realised. Together, in 1979, they co-wrote MUD, the first game whose influence can be directly traced through to today's World of Warcraft, Eve Online, and Second Life.
"I thought it would give people the freedom to be themselves - or become themselves. It was a freedom thing from my point of view. That's why I invested so much time in it."
It is a landmark moment. Bartle is talking the day before the final dissolution of MUSE, the company formed in 1985 to promote MUD. Getting into a field early can hurt; you have to invent the business model as well as the technology. In MUSE's case, the money it made mostly went, Bartle says, "to other people". Chief among them: the now AOL-owned Compuserve, then the leading online service. MUSE's founding CEO, Simon Dally, got some of it, too, before going mad and committing suicide.
Bartle's account of MUSE's history is only slightly less improbable than the story of Spinal Tap's drummers. You can still, however, play MUD - as long as you're willing to use a funny, old-fashioned technology called "text". Some people have been playing it for 15 years.
These days, Bartle teaches computer games design at Essex, teaching others to create virtual worlds, lecturing on everything from computer science to mythology. At the recent State of Play conference in Singapore, he found himself repeatedly explaining to people debating abuse, governance, business models, regulation, and the role of game gods (tyrants or benevolent dictators?) that these things have not changed since MUD began. He tries every world out there.
"Designers don't play for the same reasons as other people," he says. He visits virtual worlds to study their design. "I don't get to be the hero. I just get to make other people be heroes." But graphics ended the days when a couple of guys could sit down and write an entire world. "It costs a lot of money these days to make virtual worlds. So all I can do is create them as thought experiments."
From other worlds, "I want to see something that will surprise me." Instead, "What I see is less and less, not more and more, shallower and shallower, not deeper and deeper. They're all going in the same direction, not spreading out."
The primary reason: the need to keep attracting new users. A difficult game - say, where your character, when killed, stays dead ("permadeath") - may be more satisfying long-term. But a newbie would rather resume after a few minutes finding his body than to have to start over. When, that game's users create the next-generation world, the new one copies from the last one - and has to attract its own newbies.
"A text world can't compete with graphical worlds for new players," he says, even though, "Text worlds are deeper environments - the physics is more sophisticated. But the graphics look pretty, and if you're a new player, why text? All the games on computers are pretty pictures. Text is a niche."
It's a lot harder to program graphics than it is to compose text. In a text world, a snowman in a fire melts and puts the fire out. In today's graphical worlds, the snowman sits in the flame and nothing happens. Rain doesn't get the characters wet. Water doesn't flow, and objects don't float away.
"You can't play Pooh-sticks," Bartle complains. "The physics isn't strong enough to support it. And even if it were, the players wouldn't necessarily want it." He mimics an unhappy newbie: "Somebody else might pick it up and run away with it and then I've lost my Pooh-stick."
The pictures are better on the radio; the immersion is better in ASCII.
"Text is the most immersive environment. It's just that we won't know that until we've gone through everything else." µ
Sign up for INQbot – a weekly roundup of the best from the INQ