Skool Daze: The Sandbox Nobody Called a Sandbox

The word “sandbox” arrived in games criticism about twenty years after the thing it describes. By the time we had the vocabulary — persistent world, systemic design, emergent behaviour, NPC schedules — the discussion had attached itself to open-world games with budgets, and everyone agreed the lineage ran through Ultima and Deus Ex and eventually Grand Theft Auto.

Skool Daze came out in 1984, cost a few quid on cassette, and did most of it already. David Reidy built a school with a working timetable, filled it with named people who had somewhere to be, and gave the player a goal that the school itself has no opinion about. Then he let go.

The school does not care about you

This is the part that still surprises. You are Eric. Your report card is in the headmaster’s safe and it says things about you that you would rather your parents did not read. That is entirely your problem, and the school has a day to get through regardless.

Mr Wacker is the headmaster and he has a day to run. Mr Rockitt, Mr Withit and Mr Creak have lessons to teach, and they walk to the rooms where those lessons happen because the timetable says so. Angelface is a bully who behaves like one. Einstein is a swot who will report you to a teacher for anything he sees, which makes him the most consequential character in the game and also the one you spend the most time trying to lose in a corridor.

None of these people are waiting for you. They are not standing at their posts in idle animations until the player triggers them. They are going about a day that would happen identically if Eric stayed in bed. Bells ring, lessons start, playtime happens, and the population of Skool moves through the building on business of its own.

Forty years of games have taught us to read a populated environment as a set of props arranged for our benefit. Skool Daze predates that assumption. The building is running a simulation and you are a guest in it.

Trouble as a currency

The mechanics are almost entirely mischief. You have a catapult. You can hit people. You can write on blackboards. You can sit down in a lesson, or fail to, and both have consequences. The combination lock on the safe is spread across the four teachers, and getting it out of them requires causing the sort of trouble that a schoolboy causes.

The failure state is where the design gets clever. Misbehave and a teacher gives you lines. Lines accumulate. Enough of them and you are expelled, and the game ends. So the resource you spend to make progress is exactly the resource that kills you, and the entire game is a negotiation with that fact.

That is a beautiful loop, and it is more sophisticated than most of what the eighties managed. Compare it to a health bar, which is a resource you lose by being bad at the game. Lines are a resource you lose by playing the game at all — the only way to keep your record clean is to do nothing, and doing nothing loses. Every action worth taking costs you something you cannot afford much of.

And crucially, the accounting depends on being seen. A catapult pellet fired in an empty corridor costs nothing. The same pellet fired where Einstein is standing costs you lines, because he will find a teacher and tell. So the game is about surveillance and timing, and the timetable is the thing you exploit, because a timetable tells you where everyone will be.

Read that back and see what it is. You have a goal, a hostile schedule, an information economy, guards with patrol routes and a snitch. Somebody built a stealth game in 1984 and set it in a comprehensive.

Emergence out of very little

The engine underneath is small. Each character has a place to be, a route to get there, and a short list of reactions. Reidy simulates attendance rather than personality, and lets the collisions between attendance and Eric’s behaviour produce the stories.

That is the same trick Little Computer People pulled off a year later from the opposite direction — a single character with a routine, in a house, doing things whether or not you look. Both games are proof that the perceived aliveness of a simulated person is almost entirely a function of whether they have an agenda. The fidelity is irrelevant. Skool Daze’s people are a dozen pixels and a walk cycle, and they feel like people because Mr Creak is going somewhere.

What comes out of this is the good stuff: the emergent anecdote. You set up an elaborate plan involving a specific teacher in a specific room at a specific time, and it collapses because Angelface picked a fight in the corridor and delayed you by eight seconds, so now the bell has gone and everyone is somewhere else. Nobody wrote that failure. The schedule wrote it.

Every immersive sim since has been trying to manufacture that sentence. Hypnospace Outlaw gets there by making a whole fake internet that exists regardless of your attention. Skool Daze gets there with a timetable and about forty kilobytes.

The goal is yours, which is the radical bit

Here is the design decision that separates Skool Daze from everything around it on the shelf in 1984. The game states an objective and then declines to have a plan for it.

A platformer of the period tells you the objective and hands you the route: this screen, then that screen, then the boss. The objective and the method are the same object. Skool Daze splits them. The safe is over there, the combination is inside four men who are walking around, and the sequence in which you get it out of them is a question the game never answers because it does not have an answer. The answer is whatever the timetable and your nerve permit today.

That is the definition of the thing we now spend a great deal of money on. Hitman is this. Deus Ex is this. The reason it works at cassette scale is that Reidy kept the state small enough that the player can hold it in their head — a handful of rooms, a handful of people, a clock. Modern immersive sims have to fight legibility problems that a 1984 school does not have, because a 1984 school fits on a single sheet of graph paper.

The comparison worth making is with Head Over Heels, which arrived three years later with a far more sophisticated engine and is a strictly harder game — and which is, structurally, a set of authored puzzles. Ritman decided what each room wants. Reidy decided what each person wants and then went home. Both approaches produce excellent games; only one of them produces a different game on Tuesday.

The parts that have not survived

Being honest about it: Skool Daze is opaque in a way that a 1984 game could get away with and a modern one could not. The route to the safe combination is discovered by experimentation that borders on brute force, and there is a generation of players who solved it because a friend at school told them, which was the era’s actual documentation system. The manual helped less than it should have.

It is also fiddly. Eric’s controls are of their time, the collision is approximate, and a plan can fail because a jump did not register rather than because you were wrong. That distinction matters in a game whose entire appeal is that consequences are legible.

And the C64 conversion, which arrived after the Spectrum original, has the usual problem of a game designed around monochrome legibility acquiring colour it did not need. The Spectrum version is the reference.

Why it matters

The received history of emergent design credits the people who could afford to say what they were doing. Skool Daze had no vocabulary for itself, no design document anyone quotes, and a marketing pitch about being a naughty schoolboy. It sold well, spawned a sequel in Back to Skool the following year, and settled into being remembered as a charming British curio.

The charm is real and it is also camouflage. Underneath is a systems game with scheduled agents, an information economy, a self-limiting resource loop and a world that runs without you — shipped for the price of a paperback, eight years before the phrase “emergent gameplay” got its first outing in a press release.

It runs in any Spectrum emulator, and there is a well-known modern disassembly-and-remake project that has documented the original’s logic in enormous detail if you want to see the machinery. Give it a lunchtime. Then watch Mr Creak walk to his lesson without looking at you once, and count how many games since have had the confidence to let a character do that.