A pet project of mine (which I barely have time for, really) is a little MUD engine written in Python. I was thinking about implementing a “multi-handed” system for it; in other words, a way to let players and mobs have more than two anatomical hands. Actually, these “hands” could be anything; a prehensile tail, tentacles, a telekinetic “hand”, etc. In the end I sort of decided against it due to its complexity, but it made me think of something else, namely how players' expectations vary based on what you implement.

I bet several of you have played Quake or one of its successors. They are simple games, where the gameplay for the most part begins and ends at the barrel of whatever gun you happen to be carrying. Yet, these games were and are pretty damn popular. Even discounted such factors as the impressive engine and its moddability, the games evoke fond memories among many FPS enthusiasts. I know it does in my mind.

Now look at Half-Life 2. While made from the same stock as Quake (quite literally; Goldsource—the original Half-Life engine—was based on the Quake 1 source code), it is arguably a much more sophisticated game. At its essence, it's still a run-and-gun game, but it has a well-written story and the occasional thought puzzle that requires more brain than brawn.

Both of these are good games, even if they're not to everyone's tastes. That's fine. Either way, they demonstrate what I want to talk about today: Both games are excellent at meeting their players' expectations.

When you sit down and launch Quake 4 (or any other Quake game), you do so knowing that you will be treated to some simple, uncomplicated shooting action. When you decide to play Half-Life, you know you're in for a more sophisticated FPS, and when you play Baldur's Gate, you do so knowing those dice will be thundering for hours to come as you press onward through the story. And in each case, that's exactly what you get.

So what does this have to do with multi-handing? It struck me as I was thinking about the feature that if I implemented that, then I should probably also implement body parts with separate hitzones. And if I do that, then I should also implement armor that can cover one or more parts of the body, and probably also allowing for layering clothing... and so forth. In short, I would be raising the expectations on the game.

This isn't a problem if you have the competence to pull it off. If you can really implement all those things, you'll have a deeper, richer gameplay, no question about it. However, if you do not, you'll find yourself up the proverbial creek faster than you can say “failure to meet expectations”.

See, once you cross certain lines, you're implying that you're taking your product to another “level” of sophistication, something qualitatively different from a “lesser”, simpler experience, and once you do that, you'd better be prepared to deliver. And if you can't, perhaps you should settle for less. Really!

What I did was decide not to implement this system, and though I might go back and do so one day anyway, I think it was a good choice, for several reasons. By deliberately not implementing every fancy feature you can think of, you stay consistent to the experience you're promising. Yes, this may mean that some people who prefer a richer experience may never end up playing your game... but isn't that better than sorely disappointing them and damaging your reputation? Isn't it better to deliver a solid, simple game that caters to only a small part of the market, than to make a confused game that fails to deliver what it promises and disappoints everyone? I think it is.

To sum things up: It's always better to settle for something simple and do it well, knowing that your audience's expectations are at a realistic level, than to succumb to designer's hubris and raise people's expectations to a level you cannot possibly fulfill. Above all, you have to meet people's expectations, or your game will never be successful (as a side note, this is why I am having no great hopes for Will Wright's anticipated game Spore; at this point, I can't imagine how it could possibly meet the expectations it has built up).