Interestingly, though, in many ways game designers have, in many ways, encouraged this type of sandbox-style play. Hidden objects, puzzles, easter eggs, and the “testing” of how the game responds to different, unanticipated actions, all seem to encourage this type of play, and in many games it’s still encouraged. That’s not to say, of course, that as an author you can’t discourage it quickly, nor is it a bad thing to do so.

To me, it relates back to one of the central ideas I’m trying to forward: that games should try to be more about people and their emotions, motivations, and relationships, rather than about objects. Games in general (including IF, although to a lesser extent lately) have for so long been about objects — at least with respect to puzzles and puzzle-centric games — that I think players have been programmed to think in those terms.

As someone working on a 3D project closely related to IF, it’s interesting how an open graphical environment really encourages this. For instance, in text, you can pretty easily get away with describing a room without going into intricate detail about every last item in the room — a good example from Vespers is the kitchen. And by doing so, you effectively steer the player away from the things that don’t matter and towards those that do. But when you design a kitchen with graphics in 3D, you sort of have to include those little details — pots and pans, utensils, plates, and so on — or else you don’t do a very good job portraying the environment. But seeing all of those objects just seems to trigger an object-oriented reponse in players that steers them towards TAKE, SEARCH, and OPEN.

There are ways to deal with that, too, but in general I think it all comes down to design. The author sets the boundaries, and players either respond to those boundaries or they don’t. The question is how much the author wants to allow for.

I’m not for a moment advocating a railroaded design. I’ve complained about games like Photopia at length in the past. I want to allow as wide a scope of reasonable action as possible. The emphasis here, though, is on “reasonable.” For practical AND aesthetic reasons, I think this is really the only way to go.

I think another point of this is highlighted by your approach to “never allow the interactions described.” That’s one approach, but sometimes it’s hard for players to accept a limitation like that. Alternatively, characters in the game could be designed to react in certain ways to the actions the player performs, although that introduces increasing levels of complexity to the code. That’s something I’ll address in a later post.

We’ve learned through playing many old-school games that this is the way we we HAVE to play to get anywhere. I don’t think IF (or any other form of interactive narrative) has to be bound by those rules, however, and I’m happy to say that many recent games, notably those from Emily and from Eric Eve, have been able to steer their players away from that sort of thing and introduce NPC interactions that feel more fluid and natural.

As an author working on a game of my own, I would simply never allow the interactions you’ve described in the situation you’ve described. I don’t think it’s too much to ask my player to accept the premise and situation of the story she is in, and to behave in a reasonable manner. If my player was paying a friendly visit to a neighbor and tried to start rummaging through the cupboards in front of him, I would simply tell her that she’s not going to do that sort of thing. I think most good-faith players would accept this and realize that they should be focusing on the conversation and generally behaving like a reasonable human being. In other words, she will shift her frame of reference from “old school adventure game” to “(more) realistic interactive story.” In a another situation — say, the neighbor does or says something that makes her suspect him of a crime, and she breaks into his house in the middle of the night to investigate — the explore and loot approach would be natural and thus allowed.

As far as conversation: yes, we have a long, long way to go, but just introducing a TADS 3 style conversation node or two can go a long way toward making it seem more like a real conversation between two people, as opposed to a series of transactions between the player and an information vending machine. Authors should try to give their NPCs some personality and life. Maybe when you ask a sensitive question, they ask one back instead of immediately answering. “Why should I tell you?” If you ask them the same thing twice, they raise their eyebrows and say they just told you that. Etc, etc.

I agree completely — and this is actually a big part of an upcoming post (soon, I hope).

Of course, “If I’m writing a game that involves a lot of character nuance” presupposes that you’re going to spend a good deal of effort creating that nuance, and the game will likely be more richly developed than this. My example here is obviously something of a caricature of PCs in IF; actually, I had intended originally to imply that this wasn’t necessarily a representation of IF specifically, but rather games in general. As I suggested in my prior post, I think IF in general has done better than most game forms at developing character depth, an certainly your work is a noteworthy example of this.

George — I’ve tried the demo version of Graveyard but have yet to try the full version, or any of their other works. I’m eager to check out The Path, though. I’ve heard a lot about their stuff, and I think it might represent some interesting steps forward in this area.