I'd like to talk a bit about a design concept that doesn't exist yet, as far as I know. It's vaguely related to my last essay, so you'll have a better idea of what I'm talking about if you read that first.
The concept is about restarting challenges in a specific way. I don't know what to call it, exactly - let's call it "reincarnation", unless someone has a better idea. As in "restart, reload, retry, or reincarnate".
In many western RPGs, you get to make your own character(s).
Further character development is functionally on rails. While you could technically give your character some weird new skill, that isn't a very effective use of your points.
Eastern RPGs take the opposite stance, giving you rote characters and allowing you to heavily specialize them over the course of the game.
Basically, these two approaches seem like the Big Two in virtually every kind of game. Decisions are either planning or execution. If your game is heavy on planning, then your important decisions are made outside the "heat of play" - what stats your character should have, which people to bring, what program to run. If your game is heavy on execution, the important decisions are made during play, small adjustments to your course that change the permanent state of the game. Such as what spells to get, or whether to help the biker gang or the nuns.
It's not quite that simple, because (A) it's an axis, not a toggle, and (B) "heat of play" is not a very exact thing to say. For example, buying and equipping armor is execution-level when you are thinking about the "combat stats" game, but it's planning-level when you are thinking about the "combat" game. Actually, the "combat stats" game is the planning level for the combat game.
It's also a little complicated because games offer so many un-choices. For example, classic RPG combat is not full of decisions. The decision as to whether to attack, use magic, block, etc... it's a fake decision. One of them is obviously right in any given situation. Even when a decision is a decision, it's not really important: a few minutes after combat is over, you'll be pretty much fully recovered again. So it's kind of vague.
BUT... overall, when you look at a game from any given angle, you can see that some parts of the game are about planning, and some parts are about execution. That's all there is, right? Planning and execution.
Well, sort of. I mentioned that a game often has "layers". One layer is frequently "planning" for another layer, although the edges might be blurry.
For example, making a character in a western RPG is definitely planning for the rest of the game.
Aaaaaand there's the glitch.
Most games think of this kind of thing in terms of "the rest of the game". Even execution-level decisions frequently affect you for the rest of the game. But "the rest of the game" is a very long time! You're asking the player to make a skill-based decision that will affect the rest of the game... when he has the least skill!
Many players will replay the first five or ten hours of a game over and over, making new characters, trying new approaches. I've played Daggerfall, Morrowind, and Oblivion for over a hundred hours EACH, but I've never even gotten halfway through the main plot because I tend to play the first ten hours dozens of times instead.
Even when you are executing rather than planning, it is frequently in terms of "the rest of the game". When you level up a Final Fantasy character and give them the "summon monkey" spell, that's what they have. You can't change it, you can't take back those points and re-allocate them.
That is, in many ways, part of the charm. If your character is overspecialized in fire magic, he'll have a great time exploring the North Pole, but the marshlands will bury him without a problem. Dealing with weaknesses and strengths is a big part of the game: if you make a stealthy character, you'll have a hard time doing combat even when combat would be easier than stealth.
But this means that the game needs to contain a LOT of polish and deep content for every aspect, because someone who chooses stealth needs to have as much to do as someone who chooses combat! Since the choice is forever, there needs to be alternate content forever. If it ever gets boring, the player will get upset because his character has become boring... and he can't change it.
Let's talk about building houses.
In The Sims, you build a house for your little simmies. Basically, once a house is built, it's prohibitively expensive to significantly change the house (unless you're a cheating bastard). You'll change it here and there, maybe even buy a new house once or twice, but overall, when you build a house, your architectural choices last close to forever.
On the other hand, in Dungeon Keeper and Evil Genius, you build a "house". Okay, a lair. But it serves fundamentally the same purpose, right? Keep your goons happy. Plus, well, killing anything that comes in, preferably in a hilarious fashion.
But these kinds of games do something very unusual. Something that RPGs don't do. They... well, their house-building reincarnates.
Every time you beat a level, your lair is gone. The next level has a new layout, and you have to build a new lair.
The lair-building is very complicated, the very definition of "easy to learn, hard to master". Even after dozens or hundreds of bases, you're never settled into a rut, because each level presents a new set of restrictions. This level has a lot of lava, that level is going to be attacked by dragons, the third level has lawyers that tax you based on your monster levels...
This creates a game that is a lot of fun, even when it isn't perfect. It's like learning to walk, and then being asked to navigate a wide variety of obstacle courses in any way you please.
I choose the term "reincarnation" because this is not simply restarting the game! Every new level is improved not only by your own prior experiences, but also by the new abilities that the game gives you. Now you have access to a new kind of trap, now you can build a helipad...
In its way, it's a bit like playing a good FPS or platformer: every level builds on the previous levels, even though you aren't really stuck with the decisions you made in those previous levels. However, this is not an execution-level thing. This is a planning thing.
When you build a lair, you are building a system that the game will toss challenges at.
In an RPG, the equivalent would be if the game contained 20 five-hour adventures, and you created a new character at the beginning of each... but you got access to different options at each new adventure. So, in the beginning maybe you can be a warrior or a thief. The second level, you learn to be a cleric and a barbarian. Third level, ranger and mage... third level's the ranged challenge, you see.
Obviously, the levels wouldn't be completely separate. The easy way would be to make each new adventure the next generation in the same family. Anything that happens to the village, any treasures you bring home, any NPCs you help or hinder affect the next generation of adventurer.
This "reincarnation" system of design allows you to design a core game, but then allow players to explore it in a way that encourages them to learn and try many different approaches. It allows you to get the best planning-style play... over and over and over!
Fundamentally, this kind of design is the same as classic level-based design... except that you focus on the planning elements rather than the execution elements that are classically central. People have done this kind of thing before, especially in in tactical games. But, as far as I know, nobody's really thought about it explicitly.
What do you think?