#adom

The Berlin Interpretation of Roguelikes is a genre definition that was made at a roguelike dev conference in 2008. It was a handful of people who were inspired by Rogue to make their own procedural death labyrinths. This was just before Spelunky. It was before The Binding of Isaac, it was before FTL, and it was before Risk of Rain. In other words, this interpretation was made right before the definition of a roguelike radically changed. Before these genre fusions, roguelikes were mostly games like NetHack and Angband, games that were directly inspired by this RPG from 1980. They were roguelikes, because they were "like Rogue."

But let's talk about what this interpretation is actually doing. It is not, in fact, a set of rigid guidelines of what features a game needs to be part of the genre. It is a set of high and low value factors, that a game may not have all of, but the more it has, the more it is, well, Rogue-like. There isn't a game that purely embodies each of these factors, because there's doesn't have to be. Even Rogue isn't fully faithful to this interpretation. It's just that if your game has enough of these factors, then it's clearly working in the same design space, and part of the broader conversation. That's all it was doing.

So is it still relevant? Well, kind of, but not really. Sure, permadeath and procedural generation are still important to people, but for it to be turn-based and grid-based aren't necessary anymore. And I guess for the later polls, either people got tired of me posting these in their tags every day, or they just don't have an opinion one way or the other, because their vote counts didn't even hit the triple digits. It seems like complexity, resource management, and exploration/discovery are still important, while non-modal gameplay and hack-and-slash aren't as much. But I hesitate to give these results the same kind of credence because of how many people from the earlier polls didn't vote on these ones.

It seems like, these days, the roguelike isn't really about its base mechanics. It's a structural genre. You can graft any base mechanics onto it, as long as you start a new run after you die, and each run is made unique by randomness. It seems pretty straightforward.

Okay, what about Mystery Dungeon?

The Mystery Dungeon games these days are mostly known for being Pokemon spinoffs, but the series goes back to the 90s with games like Shiren the Wanderer. These games are directly inspired by the original Rogue, complete with grid-based movement, item discovery, and heavy resource management. It's more "like Rogue" than a lot of modern roguelikes. But because it's structured more like a Japanese RPG, with checkpoints, persistent progression, and a full-blown story, it would probably nowadays be called a "roguelite." Even though it precedes that term by decades!

I'm honestly really surprised that Mystery Dungeon doesn't come up more in discussions about roguelikes. Like, it's a really obvious point of contention, and one that's worth talking about. But only one person in my replies actually mentioned Mystery Dungeon in their thoughts about permadeath. I suspect it's because most of the roguelikes people are invested in these days are indie PC games from the West, while Mystery Dungeon is a Japanese series on console and handheld. But I can't prove it.

Speaking of roguelites though, we have to talk about that term. I don't think that "roguelite" is an inherently elitist term, or that it's useless as a label. it clearly serves some purpose in conversation, even if I wouldn't use it myself. I don't want to prescribe my definitions of terms from on high. I want to understand how people are using these terms in daily practice.

And it seems like it comes down to two definitions. One is the Berliner standpoint. It's less common, but it sees some use. Basically, if you're not making a grid-based dungeon crawler, you're not really making a game "like Rogue," are you? Genre fusions like Risk of Rain and Slay the Spire play completely differently from the roguelikes of old, so it feels weird to just call them roguelikes. This is treating it as a mechanical genre, defining roguelikes by dungeon-crawling in the same way we define a shooter by shooting.

The other definition is a broader one, defined by two key features: procgen and permadeath. This is the most common one in modern conversations about roguelikes, because it takes it as a given that these genre fusions are what roguelikes Are. By this framing, a roguelite is any procgen game that features persistent upgrades. Something like Rogue Legacy, where your progression isn't just you as a player, but your character growing more powerful. Permadeath and procgen serve a very specific purpose in conjunction with each other, and there is a sense where either of these factors being diluted misses the entire point. But plenty of people don't put stock into the like/lite distinction, because they don't see it as mattering. No definition here is wrong, they're just all operating under different beliefs.

(sigh) Look, genre is a lot like gender. In fact, in some languages, they're the same word. It's vague, it's arbitrary, and it doesn't make sense if you squint too hard. But it's important to people, either to describe what they're doing or to understand what others are doing. There's never going to be a strict, clear definition of roguelike that perfectly covers all cases, because that's just not how genre works. Roguelikes are what we mean when we talk about roguelikes. It's very easy to poke fun at a genre label like this, but it's a lot harder to understand what it's doing, and what it means to people. That's the conversation that really matters here.