A world that keeps running when you log out, full of people you did not choose
| Foundational title | Ultima Online (Origin Systems, 1997) |
| Defining title | EverQuest (989 Studios, 1999) |
| Core principles | Persistence, simultaneous players, character progression, social interdependence |
| Key influence | MUDs — text-based multi-user dungeons running on university networks |
The massively multiplayer online role-playing game took the persistent worlds of MUDs and gave them graphics, then handed them to tens of thousands of simultaneous strangers. Ultima Online proved the commercial case in 1997, EverQuest turned it into an obsession, and the genre spent the next decade discovering that the hardest problem in world design is not building the world — it is the people you let into it.
Persistence is the whole idea. In a single-player RPG, the world exists only while you are looking at it; close the game and it politely freezes. In an MMORPG it does not. The dragon you failed to kill is still there. So is the player who killed it while you were asleep, wearing its head.
That one property changes every design decision downstream. Loot must be contested, because everyone wants it and it cannot simply be handed to each player individually. Progression must be long, because a world that everyone finishes is a world nobody stays in. And the game cannot be balanced solely around what is fun for one person, because it is not being played by one person — it is being played by thousands, simultaneously, all pursuing their own interests inside the same simulation.
Origin's ambitions for Ultima Online were extraordinary and largely defeated by its own players. The most famous casualty was the Virtual Ecology — a genuine simulated food chain in which plants grew, herbivores ate them, and carnivores ate the herbivores. Players killed everything on sight, faster than the system could regenerate any level of it, the herbivores were exterminated, the carnivores starved, and the entire model collapsed before launch.
The lesson generalises far beyond wildlife, and every subsequent MMO absorbed it. A persistent world containing thousands of rational actors will have any exploitable system exhausted immediately — not by griefers but by ordinary players behaving sensibly. Which is why wildlife in every MMO since simply respawns on a timer in a fixed location, in open defiance of the fiction. It is a deliberate, load-bearing lie, and Ultima Online is the reason everyone tells it.
The MMORPG is the genre in which the commercial model most nakedly shapes the design. A subscription is charged per month, which means the game's financial interest is not that you enjoy it but that you keep playing it — and those two goals overlap only partially.
Hence the grind: content calibrated to take a very long time, progression curves that lengthen as you climb them, and rewards gated behind repetition rather than skill. Hence, too, the social architecture, which is the most effective retention mechanism ever devised. A player will quit a game they have tired of. A player will not so easily quit a guild of forty people who are relying on them to show up on Thursday.
Every MMORPG eventually discovers the same thing: the systems are the pretext, and the people are the product. The economies that emerge, the guilds, the reputations, the rivalries, the elaborate social hierarchies built around who can kill what — none of it is authored, and all of it is why anyone stays.
This is the genre's great achievement and its permanent liability. It can produce experiences no designer could script, because thousands of humans in a shared persistent world will produce stories nobody planned. It can also produce harassment, exploitation, and economic behaviour of a viciousness that regularly surprises the people who built the place. An MMORPG is not really a game with players in it. It is a society with a game attached, and every serious problem in the genre is a social problem wearing a technical costume.