I don’t mean it never happens. But my rule of thumb is, think about what happens if the player finds the path he least wants to. Because it’s even odds that will happen.
I like this idea, however. Try and shift the odds. I don’t recall a game that’s tried that approach.
As Emily said, it doesn’t matter why you close off solutions. If it’s a cause the player won’t revisit (initial character selection, finishing the previous chapter, whatever) then it’s a one-solution puzzle. Which is fine, of course; just be aware of what you’re building.
Speaking as just a player I agree (with zarf). When it comes to puzzle solutions, or even just plain obstacles that aren’t complex enough to be deemed puzzles, the road oft taken is almost always the road first stumbled onto. I can think of only one instance where I skipped an obvious way to push the story forward because I didn’t like where it would lead: the elephant harness in Bronze. It’s important to note that by that time in the game, the game had already firmly and clearly established itself as a multiple-paths, multiple-solutions game, and the description of said harness was clearly foreshadow-y. I was sure in my hunch that there would be another way.
Most I-F are way too opaque to support that. And if Bronze had tossed me a “Belle isn’t heavy enough to make a difference” in response to my JUMPing onto something it would have reduced my belief in my ability to choose, and hence, I would’ve thought twice about choosing to skip taking that elephant harness.
This is one class of variant-solution puzzle I’d like to implement, where you’re actually gating parts of the map between one character and another — or you’re changing the method for getting to that part of the map. If you’re playing the Russian agent, you can get into the Russian embassy through the front door; if you’re the Chinese agent, you can access the lobby openly, but getting to the back rooms requires subterfuge (climbing through a window, let’s say). Neither of them can get into the janitor’s closet without the key, so that’s a puzzle common to both characters.
Depending on who you’re playing, one solution is obvious and natural, and another not so. For the most part, I’m not talking about closing off out-of-character solutions. If a puzzle has multiple solutions worth implementing, err on the side of leaving them all in (the Russian agent can also climb in the back window, but he doesn’t have to).
This is what I mean about allowing the PC to be the hero of his choosing. If he wants to play the dark, mysterious stranger that everybody’s afraid of, fine; if he wants to play the noble, honest, forthright friend; fine. It’s more about changing the way the NPCs of the world react to you, than about changing the physics of puzzles. The golden key still opens the golden cage; it’s how you get that key that varies. The mysterious stranger can get it through trade, theft, misdirection, and intimidation; the honest friend can get it through trade, theft, misdirection, and camaraderie.
Agreed; useful but tricky, and not to be abused. If the character is well-defined and the few restrictions intuitive, this shouldn’t be too much of a problem, because the player won’t find himself trying those things (the noble knight won’t execute his captives, but he doesn’t have to; the barbarian can execute prisoners if he chooses, but it isn’t necessary).
I am going to try that in one game, but that has a very different structure. You play six suspects being interviewed about a murder, and you play their alibis in flashback. Depending on how you play the first alibi, it sets up certain conditions and facts about the murder — the time of the murder, the time the body was discovered, the murder weapon, and so on. For instance:
You start by playing the housemaid; one task is to bring the morning newspaper to the master of the house. If you meet up with him before you fetch the newspaper, he slaps your bottom (and you can react to that in a number of ways); if you fetch the newspaper and give it to him, there’s a lead story that incriminates one of the other suspects; if you fetch the newspaper but withhold it, it incriminates you, under the assumption that you’re trying to hide a specific story from him).
If I did it this way, I would rather close off avenues that reflect ways the player doesn’t choose. If you shoot a guy in the back, people henceforth wouldn’t respond when you try to play nicey-nicey. The player is telling me which paths he prefers; why close off the avenues he seems to like the most?
This is primarily what I mean by character choice. It describes things to the player through a filter of his choosing, especially NPC reactions. Some people take it personally when the NPCs treat the PC badly — as if the player is a fool, or dishonest, or whatever.
Exploring the world with different characters doesn’t necessarily mean the same situation/game, though. Think about multi-viewpoint novels, where you get slices of different POVs. (The most recent example I’ve read is the A Song of Ice and Fire series, but there are lots out there.) That can be a way of focusing attention on a given scene in a way that’s character focused. This way, you’re still revealing different viewpoints, but without necessitating replay to get them.
As a player, I find that far more compelling than a choice of who to play through a particular scenario with. For one thing, it gives you more story to play with.
Switching characters mid-story is a potential downside - I generally dislike that sort of thing, but it can be done well, and it can be very compelling. You can even structure it in such a way that it’s got a layer of artificiality on it. I think the usual device is Chaucer’s - you have a bunch of people come together and tell stories, and each story is from that character’s POV, and maybe you wrap the stories up in a little extra something just to give a start/ending to the book.
Yeah, that was a stupid thing for me to say. I guess when I was playing Backup, after killing someone for the first time i restarted in order to find a way through without killing anyone. But that was partly because I’d already half found it.
I guess what Metamorphoses did was something like this? I vaguely recall Emily saying that, depending on how you played, the game might force you to repeat a command more or less times. That’s kind of inverted, though (and I’m not overly fond of being forced to repeat a command several times before the parser accepts it).
Well, it needn’t be one solution. There could be several solutions, one or more of which may be closed off; and hopefully it’s apparent why your previous choice has been closed off. “Jimmy the Rat says, ‘You’re the crooked cop who shot Billy Bags, ain’tcha? I ain’t believin’ nothing you say about how you’ll go easy on me.’ He clams up until his mouthpiece shows up and tells you you got no reason to hold him any more.”
This isn’t so different from a structure I’ve seen elsewhere – in “A New Life” I bought something that wasn’t the lamp, and then I got to a dark room and realized that if I had the lamp I could solve that puzzle easily. Then on subsequent playthroughs I bought the lamp. (Still haven’t finished that game, but combining the hints, the walkthrough, and some stuff I figured out for myself I think I’ll be able to do it next time.) I think we’re talking about doing something similar, except with choices that have to do with your character rather than choices about what to take with you. It’d have to be a game that encourages replay, though, which might just be a game in which it’s easy to get stuck. (But then you wouldn’t choose the character you wanted, you’d choose the character you could get unstuck with.)
Adding, one possible mechanism that might avoid the pitfall of the generic story – one you can play through with any kind of character – would be that of alliances. I think Emily and Ron have talked about this as a possible substitute for the simplistic good/evil karma meter you get in some games. If certain kind of actions alienate some factions of NPCs and endear yourself to others, you could find your choices guiding you down different paths – steal the Russian ambassador’s car at the beginning, and later you’ll find yourself having to avoid the Russian guards, but the Americans will help you. Of course it’d be a lot of work to program this well.
The deal was that, if you did enough puzzle solutions that involved determination rather than cleverness – that is, breaking things, sacrificing things of value to you, risking minor pain – then future similar actions would go through with less hesitation. There was always a puzzle solution that did not require repeated actions.
But I’ve still come around to the idea that this is usually not a fair thing to do to the player, unless you are very clear in your parser response that you’re merely checking to see whether the player is sure, and that repetition might work.
After reading this discussion I’ve thought of some games (non IF) that are good examples of some of the discussed options.
Postal2, in which the PC is someone with history, background, skills, etc., but that doesn’t mean the player itself can’t move around levels in different ways: you can kill everyone, just kill the “bad guys” or don’t kill anyone - if you have patience for it, which the game promptly tests. But you don’t have to choose a “bad, good or ugly” guy at the beginning: you choose what you choose when you choose it.
An IF game could be set the same way, so that the player fills the blanks of the PC with his/her own traits - or just fill the blanks with the traits of anybody else - while playing.
Another example that poped into mind was the graphic adventure “I have no mouth, and I must scream”. In it you play four or five PCs (you choose each one in turns), each with a different background and personality, but you can make moral choices with any of them that go against their traits. This can result in NPCs regarding you as a bad person, as a good person, as a predictable person or as a person who is trying to change their ways.
But, of course, we’re talking about games, and different options lead to different gaming experiences, so the first thing to do should be to decide what gaming experience does the author wants to give the player. I believe any of the options discussed before has already seen good, solid and interesting games, either by forcing the player to play with a very strong-will character, or a game where the PC is supposed to be the player itself.
Someone wrote something about an IntroComp example, and it made me think of another entry with a very interesting effect regarding the PC and the player itself: Memento Moratori, in which the player does not play the PC, it rather tries to manipulate him. So, so nice.