And yeah, Return to Zork was just nuts, I couldn’t play through a minute of the game without a walkthrough. It was so full of random solutions! But I don’t remember the GUI as being contextual. Did it change depending on what you were interacting with?

I also confess to enjoying the contextual multi-verb interface in Return to Zork (a sample in the article’s topmost image), although it could indeed be overwhelming sometimes. A great number of the available action options have led to either the player’s death or destroying important items and objects in the game (You could kill characters too).

A multi-verb system is not bad in and of itself. It’s up to the designer to make proper use of it. And if you’re not going to make proper use of it, then redesign the interface to fit your gameplay!

With the “Slug/push/fondle” example, it would be a brilliant use of the contextual verb-coin interface if they all had different outcomes. You need to get through a door on a bridge, a guard is standing in front of it. You can slug him, but since he’s way bigger than you, he slugs back and you stumble away defeated. Bad choice. You can push him, but he barely budges. You can fondle him, which throws him off balance, he backs up a couple steps to the edge of the bridge. The, push him again, and splash! You can then walk through the door.

I would love to see a contextual multi-verb system implemented effectively. It would mean, however, that you’d have to include lots of possible verbs with meaningful results, and give the player some freedom in how to solve the game. Lot’s of work!

But I also agree with what Vince says. You need to design your game really well if you want a complex interface to pay off for the player. Especially as complexity usually comes together with a certain amount of tedium.

I’ll also add that one of the things that I’ve received most positive feedback on (and am most happy with myself) in the (so far) only adventure game I’ve made, is that I added actions that didn’t fit into the standard verb coins that the game use.

With the one-click system, there’s no telling what your character will attempt to do with some objects, until after clicking on them (or mouse over in some cases). Pick up? Activate? Move? This isn’t really mitigated at all by the sierra system, since all you have is the ambiguous hand. In this sense a context sensitive menu/verb coin might be the best compromise of efficiency vs. versatility, and it would eliminate the need for “I can’t talk to THAT, it’s a cup” type messages. You could have “Look at Guy, Talk to Guy, Slug Guy, Push Guy, Fondle Guy” in one menu and only “Look at Beautiful Sunset” in another. Though since basically every option would have a meaningful response, it would once again encourage the “try everything” strategy. For better and worse.

I love this article, though. When I first started attempting to make games, I always wanted to make a custom GUI, that would be aesthetically appropriate and “cool,” but I never really thought about it in terms of usability, redundancy, etc. I’m a big advocate for choosing an interface that encourages the playstyle you’re trying to create.

An illustration that might help people who are so used to clunky Lucas Arts and Sierra style interfaces that they don’t notice them anymore. If you’ve played many JRPGs, you’ve probably noticed there are two major ways that you interact with the creature/objects on the map. Your Final Fantasies, and virtually everything else uses the one-tap system. Walk up to a guy, tap action button, meaningful interaction occurs. But then you’ve got your Dragon Quests, Earthbounds, and other games inspired by these, where every time you hit the action key, a menu pops up and you have to choose “Talk, Examine, Pick-Up, Stats, Items” blah blah. To me this has always seemed like an overly clunky and intrusive mechanic, especially when there are multiple verbs to choose from (I think the old Dragon Quests even had one for using stairs). Yet we accept this functionality in Avdenture Games all the time, without thinking about it. Granted, the world-interaction gameplay in an RPG is usually much more straightforward than that of an adventure game (a lot of them play almost exactly like a keyboard-driven adventure, but the focus is different), but it just shows how, if you aren’t going to make good use of a complex interface, it is only going to get in the way of the people trying to play your game.