---

---

Words Instead of Numbers

By 1992, every RPG in existence asked the same question the same way. How strong is this character? Seventeen. How smart? Twelve. What does twelve mean? Check the chart.

Steffan O’Sullivan looked at thirty years of tabletop design — the charts, the modifiers, the cross-referencing, the mental arithmetic — and asked a question nobody else had thought to formalize: what if the character sheet just told you?

Not a number. A word. Fair. Good. Great. Superb.

The answer became FUDGE — the Freeform Universal Do-it-yourself Gaming Engine — and it rewired how an entire branch of the tabletop hobby thinks about resolution mechanics. The Adjective Ladder didn’t just simplify RPGs. It redefined the interface between the game and the player, replacing numerical abstraction with semantic clarity. And then O’Sullivan gave it away for free, a decade before the rest of the industry figured out that open-source game design was even possible.

---

The GURPS Years

O’Sullivan entered the industry through Steve Jackson Games in the late 1980s, writing supplements for GURPS — the Generic Universal RolePlaying System. His output was distinguished by deep research and a willingness to inhabit unusual genres. GURPS Swashbucklers (1990) covered history from Elizabeth I to Napoleon with enough fencing detail that historical fencers would later cite it. GURPS Bunnies & Burrows (1992) adapted B. Dennis Sustare’s beloved 1976 original into the GURPS framework. Two bestiaries drew on genuine zoological knowledge.

He was good at it. Swashbucklers is frequently named among the best GURPS supplements ever published. The historical fencing community considers its combat chapter a genuinely useful reference.

But something was shifting. The deeper O’Sullivan went into the machinery of simulation, the more he noticed what the machinery was costing. Every modifier required a lookup. Every combat round required arithmetic. The system was powerful, elegant in its comprehensiveness — and it kept interrupting the story to ask you to do math. Working on a planned GURPS Faerie supplement, he found the system too rigid to handle anything as nebulous as Faerie. Building non-human races, he discovered the point-buy system didn’t scale well at extreme values.

O’Sullivan called this “System Intrusion.” The moment when players stop role-playing to calculate.

He would spend the next decade trying to eliminate it.

---

The Usenet Proposition

On November 17, 1992, O’Sullivan posted a proposal to the rec.games.design newsgroup. He’d watched two previous collaborative RPG projects dissolve into flame wars and chaos. This one would be different. He wanted community input. He would make all the final decisions himself.

The post outlined a game system that worked differently from anything then available. Instead of numerical scales, it used an ordered hierarchy of adjectives. Instead of d20 linear probability or percentile spread, it used custom dice that produced a tight bell curve centered on “no change.” Instead of a closed, commercial product, it would be developed collaboratively and released for free.

He called it SLUG — Simple Laid-back Universal Game. The community feedback refined it. The acronym changed. By 1993, it had become FUDGE: Freeform Universal Do-it-yourself Gaming Engine. The “Donated” in the original expansion acknowledged the community contributions. But the core architecture — the ladder, the dice, the modular design philosophy — was O’Sullivan’s.

The text of the 1995 ruleset was written entirely by O’Sullivan. He retained the copyright, later transferred to Grey Ghost Press in 2004. The collaboration was real. The authorship was clear.

This was also, quietly, a publishing revolution. O’Sullivan released FUDGE as a free system on the internet nearly a decade before Wizards of the Coast introduced the Open Game License in 2000. Fred Hicks, co-creator of Fate, later called FUDGE “the first example of a ‘crowd-sourced design’ RPG I can think of” and “seriously ahead of its time.” O’Sullivan treated a game system not as proprietary product but as common infrastructure. The hobby hadn’t seen anything like it.

---

The Ladder

The Adjective Ladder is the invention that matters.

Seven levels. Terrible, Poor, Mediocre, Fair, Good, Great, Superb. Each maps to a numerical modifier (−3 through +3), but the numbers are background math — the player-facing interface is the word.

Before O’Sullivan, a character with Strength 14 in one system meant something completely different from Strength 14 in another. You needed system-specific fluency to interpret what any number implied about actual capability. The Adjective Ladder eliminated this. A “Good” swordsman is a good swordsman. A “Superb” lockpick is superb. The word IS the information. No chart. No reference. No cognitive load.

The mathematical backbone is the 4dF system — four custom six-sided dice, each showing a plus, a minus, or a blank on two faces. O’Sullivan and collaborators Andy Skinner and Reimer Behrends tested over a hundred dice techniques before settling on the format. He’d spent more than a year using 2d6 — one positive, one negative — before spotting Koplow Games’ six-sided dice printed with plus and minus signs. Remove one of each face. Add two blanks. Roll four. The result ranges from −4 to +4, following a sharp binomial distribution: 62% of rolls land within one step of your base rank. Extreme swings occur at 1.23% frequency. Your character is, statistically, who they say they are.

This was a direct counter-innovation to the d20, where a first-level wizard and a twentieth-level fighter have the same 5% chance of rolling a natural 20 on any given swing. O’Sullivan built a “competence-first” engine. The system respects what you’ve already established about your character.

Earlier games had used descriptive terms — Tunnels & Trolls, Prince Valiant. O’Sullivan was the first to lock them into a mathematically consistent resolution engine where the adjectives weren’t flavor text. They were the operating system. Today Fudge Dice are manufactured by multiple companies worldwide, sold in every game store that carries dice, and used in systems that have nothing to do with FUDGE — because the probability curve is that good.

---

The Architecture

What makes FUDGE remarkable as engineering — not just as invention — is the universal resolution engine.

In most RPGs of the early 1990s, combat used one ruleset, social interaction used another, and magic used a third. FUDGE applied the Adjective Ladder and 4dF to every possible interaction. Picking a lock, debating a king, dueling a rival — same mechanic, same dice, same cognitive pathway. The GM’s overhead dropped to almost nothing. Learn one system, run everything.

The modularity is the second architectural achievement. FUDGE was designed as a toolkit, not a finished game. GMs could select which subsystems to use — gifts, faults, superpowers, different magic systems — and snap them into the framework without breaking the core math. This was deliberate. O’Sullivan built FUDGE to be incomplete, so that every table could complete it differently. RPG historian Stu Horvath, writing for MIT Press, called FUDGE “simultaneously a system of rules and a treatise on game design theory.”

The structural weaknesses are real. The “steep curve” problem means that even a +1 modifier represents a massive statistical advantage — bonus stacking threatens balance quickly. The original 1995 rules lacked a standardized skill list, leading to attribute bloat in poorly run campaigns. The toolkit nature means FUDGE requires a competent GM to assemble into a playable game. Rick Swan observed that FUDGE is “about as appropriate for novices as calculus is for preschoolers.”

O’Sullivan addressed these over time. The Princess Bride RPG added rigid character templates. The 10th Anniversary Edition expanded the ladder beyond Superb with “Legendary” levels. The architecture evolved. The foundation never changed.

---

Sherpa

In 1997, O’Sullivan published a game called Sherpa. It was designed to be played while hiking.

No dice. No character sheet. Character stats fit on a business card. Full rules on a 3×5 index card. The randomizer was the centiseconds digit on a digital watch.

This wasn’t a gimmick. It was the logical extreme of O’Sullivan’s design philosophy: if system intrusion is the enemy, build a game where the system is so minimal it can operate while you’re climbing a mountain. Sherpa tested the lower bound of how little mechanical infrastructure a role-playing experience actually needs.

The answer was: very little.

---

The Princess Bride

Twenty-six years after FUDGE first appeared on Usenet, O’Sullivan published The Princess Bride Role-Playing Game (2019) through Toy Vault, successfully funded on Kickstarter.

It’s the synthesis. Everything he learned across three decades of design, applied to a single licensed property.

The core is still FUDGE — Adjective Ladder, 4dF resolution, modular subsystems. But O’Sullivan layered on narrative refinements that reflected two decades of evolution in the broader hobby. Player-facing mechanics, where the GM rarely rolls dice and players roll against NPCs’ static ladder ranks. Meta-currency refined into “Grandpa, Wait!” points for players and “Life is Pain” points for GMs — a narrative economy that lets players intervene in the story’s flow.

The swordplay subsystem deserves specific mention. Based on the historical fencing schools referenced in the film — Agrippa, Capo Ferro, Thibault — it adds a tactical rock-paper-scissors layer to melee combat while maintaining the core FUDGE resolution. Complex subsystem built on a simple foundation without breaking the foundation’s logic. That’s the architectural test, and it passes. Twenty-seven years after FUDGE, O’Sullivan was still refining the same philosophy: trust the players, trust the GM, get the rules out of the way.

---

The Propagation Chain

The lineage from FUDGE to Fate is among the most thoroughly documented in RPG history.

Fred Hicks and Rob Donoghue were active FUDGE players in college during the mid-1990s. Hicks ran a Fudge-based Amber campaign in 1999. Their collaboration produced Fate — originally an acronym for “Fudge Adventures in Tabletop Entertainment.” Asked to confirm Fate was based on the FUDGE engine, Hicks replied simply: “You recall correctly.” He later explained the motivation: FUDGE “didn’t give me a lot of good pre-built parts to work with. That’s what led to creating Fate with Rob Donoghue: the desire for a more plug-and-play experience with Fudge.”

Fate inherited four core FUDGE mechanics: Fudge Dice, the Adjective Ladder, skills rated on that ladder, and Fudge Points (which became Fate Points). Its signature addition — Aspects, free-form descriptive traits that function bidirectionally — grew directly from O’Sullivan’s Gifts and Faults system. The creative pipeline is clean: O’Sullivan invented the foundation, Donoghue and Hicks built the house.

The Fate Core Kickstarter asked for $3,000 and raised $433,365 — 14,445% of its goal. The system has won multiple Origins Awards and ENnie Gold Awards. Games powered by Fate include Spirit of the Century, The Dresden Files RPG, Diaspora, Atomic Robo, and dozens more. Beyond Fate, Grey Ghost Press published a dedicated FUDGE ecosystem: the Deryni Adventure Game (Origins nominated), Terra Incognita (Origins nominated), and Gamemastering Secrets (Origins Award winner). The Fudge Factor magazine ran fourteen issues.

O’Sullivan deliberately left FUDGE unfinished — a kit, not a cathedral. Its greatest commercial success came when someone else took the kit and built exactly the pre-assembled game he’d refused to make.

---

The Footnote That Wasn’t Small

O’Sullivan was one of five co-plaintiffs in Steve Jackson Games v. United States Secret Service (1993), a landmark digital privacy case. He was reportedly logged into Steve Jackson Games’ Illuminati BBS at the exact moment the Secret Service unplugged it during a March 1, 1990 raid. His private email was read and deleted without a warrant. Judge Sam Sparks ruled the Secret Service violated the Privacy Protection Act and the Electronic Communications Privacy Act.

A game designer from New Hampshire, connected to a BBS about tabletop games, became part of the legal precedent that shaped American digital privacy law. The case contributed directly to the founding of the Electronic Frontier Foundation. It tells you something about the era, and about the kind of communities O’Sullivan moved through before the internet was mainstream.

---

The Honest Assessment

O’Sullivan’s career follows a clean three-act arc: learn the rules, break the rules, rewrite the rules.

Act one was GURPS — mastering the high-crunch simulationist tradition. Act two was FUDGE — stripping it down to semantic essentials. Act three was The Princess Bride — synthesizing minimalism with narrative sophistication. Each phase built on the last. The craft deepened visibly.

The output is modest in volume — roughly a dozen distinct publications across thirty-five years, plus six RPGGeek 24-hour contest wins that show a designer still experimenting with extreme constraints decades after FUDGE. Game design was always a side pursuit, not his primary livelihood. But the ratio of quality to quantity is striking. Nearly everything he published contributed something to the design conversation.

Here is what the numbers say.

---

The Scoring Case

---

The Hidden Pattern

O’Sullivan solved the problem nobody else was trying to solve.

In the early 1990s, every designer was trying to build a better simulation — more realistic combat, more accurate skill checks, more granular character creation. O’Sullivan asked a different question: what if the problem isn’t the simulation? What if the problem is the interface?

The GURPS years taught him that the math works. The 3d6 bell curve is elegant. The point-buy system is fair. The problem isn’t the engine — it’s the dashboard. Players don’t need to know their Strength is 14. They need to know they’re Strong.

Every system O’Sullivan built after 1992 was an interface redesign. FUDGE replaced the numerical dashboard with a semantic one. Sherpa reduced the interface to a wristwatch. The Princess Bride RPG made the interface narrative. Different expressions of the same insight: the rules should be transparent to the experience, not obstacles within it.

The open-source decision follows the same logic. If the system is infrastructure, not product, then restricting access to it is an interface problem too. Remove the paywall. Let everyone build on it. The infrastructure improves when it’s shared.

O’Sullivan is the inverse of the Greg Porter case study in this project’s methodology. Porter built a brilliant system nobody copied — Architecture 7, quality without propagation. O’Sullivan built a deliberately incomplete system designed to be copied, and it was. The architect who wouldn’t build the cathedral drew the blueprints everyone else used.

---

What Remains

The Adjective Ladder, now the operating system for the entire Fate ecosystem — Dresden Files, Spirit of the Century, Atomic Robo, hundreds of third-party implementations.

The 4dF dice, a physical standard manufactured worldwide, used in systems that have nothing to do with FUDGE because the probability curve is that good.

The proof — delivered a decade before the Open Game License — that a game system released for free can generate more design innovation than one locked behind a copyright wall.

And The Princess Bride RPG, published twenty-six years after FUDGE, showing that O’Sullivan wasn’t done. That the same designer who stripped role-playing down to seven adjectives in 1993 could build a tactically rich swordplay system on that foundation in 2019 without betraying a single principle.

He didn’t create industry infrastructure on the scale of D&D or GURPS. He didn’t produce a vast solo-authored catalog. He didn’t make game design his primary career.

What he did was redesign the conversation between a game and its players. He proved that the interface matters as much as the engine. And he gave it away.

Total: 28 points. Year: 1992.

---

Total: 28 points. Year: 1992.