Tea Leaves

I'd like to point out that there's a second sin to watch out for here, as well. I personally didn't enjoy Morrowind very much--precisely because of the opposite of what you describe. The trouble wasn't that I was forced to make choices, but that my choices were too unconstrained. There was no sense of tension, and too many options.

A particularly bad version of this that I have seen in American-style RPGs (and I include the BG series in this, although their main plots were quite gripping) is when although there is a "main plot" to be followed, you *must* go through the seemingly not-at-all important side plots or you'll be unable to progress in the main plot. The typical way this manifests (in the BG series as well as Morrowind) is that if you go through the main plot with serious devotion, leaving aside the frivolous-seeming side adventures, you discover that you are quite simply underpowered compared to any enemies you might face. Your equipment isn't good enough, your skills aren't good enough, you're not high enough level, or even you don't have enough experience as a player to know how to work things optimally--and you get eaten alive.

So the important goal is the magician's choice you describe, with an additional variation: a game designer should restrict the player's choices, without making those restrictions obvious--or the game designer should direct the player's choices, without making the direction obvious.

These two behaviors, restriction and direction, aren't mutally exclusive. You can have a mixture of the two. If you have too much of either (or both) you get "railroading", where the player doesn't feel like they have any real control over where things are going. Japanese RPGs typically lean in this direction--you have choices only on the very small scale, until you meet the next plot-hook--and when there are plot points or important items or whatever in a part of the story, you're forced to acquire them. If you don't have enough of at least one of them, you achieve a sort of muddled plotlessness. American RPGs typically lean in this direction--you're left in a wilderness with only a vague idea of where you should go.

The other side of the picture is how visible the control is. In a *good* restrictive Japanese-style RPG, the control is actually not too painful--you're restricted to following the plot, but part of the game is being part of the story. The good design in this genre gives you just enough freedom that you always feel like you can play around doing other things, and you don't chafe too much at your collar because you *really want* to know what happens next. The typical failure of this style of game is one of:

1) You don't have quite enough control of your actions in-between plot points to feel like your progress is your own. Puzzles solve themselves. Combat is not challenging.

2) The plot is not engaging enough for you to care what happens next. In a well-designed Japanese-style RPG, the in-between moments should be fun, but you should always feel the tension urging you on to reach the next objective.

3) The plot is engaging, but makes no sense whatsoever. This shows up both as places where your actions have impacts that are completely unpredictable (you walk onto the red square and suddenly are forced to go to the next plot point even though you haven't really "gotten" there in the story yet. No idea why the red square sent you there), and places where you achieve things that should have an effect but don't (you're standing right next to the person you're supposed to rescue, and nothing happens. In reality, it's because you weren't supposed to be able to get to that spot and there was a bug.) These happen mainly because of complete failures on the designers' part: there should be no way to avoid the railroading and make these things happen (or get to places where nothing happens.)

In a *good* American-style RPG, the control has to be a lot more subtle. It's important for every cause and effect in any plot or subplot to follow naturally. Smart designers do things like make it so that the giant eating the farmer's sheep is there ahead of time, and if you kill it, the conversation with the farmer about it still makes sense. Like: "Well, you could help me out with this problem I have... A giant in my meadow has been eating sheep..." "Oh, you mean that giant? I already killed it!" "Really! Thank you, etc. etc.", and not jumping straight to the farmer thanking you for doing something he never asked you to do.

The failures of American RPGs typically show up as lack of direction rather than poor hiding of the hands. I suspect that the failure of Icewind Dale II is partially because the designers simply aren't used to working this way--they didn't have the time (or money) to come up with a real plot. (And a real heavy plot, like BG and BGII, wasn't really a part of the series.) So they took shortcuts, and relied more on heavy plot direction. But because they were not used to that style of writing, they were unable to make it feel right. It was a mixture of the worst of both worlds.

Anyway, the failures of American-style RPGs:

1) Direction and constraint is so light that you almost forget the main plot even exists. This is pretty rare, nowadays, since BG and friends showed how well it can be done. But Morrowind is a great example of this. The entire world is open to you right from the start, and if you wander off into it and ignore the main plot, it becomes an extremely boring place, because there's no driving force.

2) The plot is strong enough, but the direction is poor. You know what you have to do to proceed, but there hasn't been enough of an indication of what you should do next. If the world is fairly small at the outset (i.e. you're forced to stay within a small town for the first bit), this can be okay--discovery of what's going on can be a lot of fun, as long as the range of possibilities is constrained. But if you're given a wide area to wander without much idea of how to accomplish your next goal, you can get very unhappy if you wander in the wrong direction.

3) The final problem (and the main problem with the BG series) is as I described above: when the progress of the story outstrips the progress of the characters. In both BG and BGII I had multiple occasions where I came up across an enemy in a plot point that was simply too powerful for me to defeat. This is extremely disheartening: the fate of the world is hanging in the balance, and I need to go save some lord's daughter from a rampaging werepoodle just so I'll gain some XP and be able to face the vampires. It breaks the logical flow of the story, and forces the player into a playstyle they don't care for. (If you're pursuing the main plot single-mindedly, you're really not the type of person who wants to go wandering the countryside looking for people to help.)

Sadly, I don't know what advice to give on that last point, except perhaps for game designers to *make sure* to playtest the game in multiple styles of play: Try it out on the folks (or like the folks) who finish every side quest before proceeding, try it out on the folks who storm right through the main story and ignore side quests until they play the game a second time. If you've got it right, neither type of player should be exceptionally overpowered or exceptionally underpowered when they reach the tougher parts of the plot.

Your point about aimlessness is well taken -- like I said, I'm not asking for no plot guidance. I just want the strings to not be quite so visible.

You're also right about the "wandering in to an area you're underpowered for" issue. I'm not sure what the right solution is. Basically, I think the entire D&D-inherited concept of "character level" is suspect here. Like the capital gains tax, there's nothing you can do with a character level system that doesn't break a game's pacing along one dimension or another. Either you ensure that the characters are always facing enemies of roughly the same strength, or you run the risk that they're going to get stomped on like a six year old cheerleader.

Somewhere along the way, people have gotten the idea that a computer RPG must, by definition, be about a big row of numbers that steadily get bigger over the course of a game. I think there are a few games that manage to buck this trend somewhat -- System Shock 2, Deus Ex, and Beyond Good and Evil -- by essentially hybridizing the genre with platform game mechanics. I think that, over time, that's probably the right way to go.

"Somewhere along the way, people have gotten the idea that a computer RPG must, by definition, be about a big row of numbers that steadily get bigger over the course of a game."

That's why my innovative RPG gives you a series of "incompetence ratings" in various skills that you work to reduce to zero.

It's might be interesting to consider the connection between the "side quest/save the farmer's daughter before I can deal with the vampire king" issue here and in novels. Even the worst Tolkien ripoff fantasy that's adding plot purely for the sake of extending the length of a series (by postposing the showdown with the Dread Lord) tends to integrate this kind of thing into the main plot, or at least fake it as important character development. Even bad authors sometimes manage to convery that had Bob the Peasant Hero not saved the farmer's daughter, he wouldn't have been the kind of man who could face the Dark Lord. RPG side quests, on the other hand, often seem either purely mercenary ("I need some gold, and this guy's got some gold!") or make the matter of "preparation" a purely numerical thing, rather than any kind of aspect of even an open-ended story.