Limited parser

I’m looking for information about games that use a limited parser.

I’m working on something right now (“The Messenger”) that has a very limited action set: move, examine, use.

In recent years I’ve found that I like a lot of (graphical) point-and-click games, even though the only verb is “click on.” I’m interested in seeing some of that simplicity in a text game interface.

In The Messenger, when you type “help,” it says:

Here are all the things you can do:

  • go in a direction (n,w,e,s,u,d)
  • look around (“look” or “l”)
  • examine something (“examine” or “x”)
  • use something (“use” or “u”)
  • use something “on” or “with” something else
  • check inventory (“i”) or status (“status”)
  • save, restore, restart, quit

When you “use” a takeable object, you’ll probably take it. When you “use” a button, you’ll probably push it, etc.

The goal is to give the player a complete and pretty short command set.

It avoids “guess the verb,” but there is still potential for “guess the noun.” (Maybe analagous to “pixel hunting” in point-and-click games.)

I’m wondering, can anyone point me to examples of decent games that use a limited parser like this? (I’m not so interested in explicit-choice games or Twine games.)

Also, are there any posts or articles with insights about this concept?

There’s Treasures of a Slaver’s Kingdom, which besides navigation verbs has only REGARD, USE, SEIZE, PARLEY, and ASSAIL as verbs – which gives you an idea of the flavor of the game, though you can also use the more boring EXAMINE, USE, TAKE, TALK, and ATTACK. It’s pretty great.

I wrote part of a game with a limited command set – the commands were basically “attack”, “defend”, “move” with exactly one noun each. It seemed to be workable, and I got a tutorial area finished, but I had a hard time thinking of a way to make the body of the game fun. (Which is why it’s still just a part of a game.)

You’re recreating the UI model of a graphical adventure game, as you note. My theory is that such a game can be recast into hypertext, dropping the parser entirely. But I’d want to build it to be sure.

Another issue (as in graphical adventures) is sort of guess-the-verb in reverse; the player might not be able to divine what “use” is going to mean for a given object in a given situation. (A situation like this tripped me up in Treasures of a Slaver’s Kingdom – the use for the object was supposed to be completely non-obvious, and you were presumably supposed to find it by guessing that it would have some use in the location in question – but I fixated on the idea that it was supposed to be used in another location that I hadn’t unlocked yet.)

In this sort of game, the player never cares about predicting that. Either it’s obvious, or the player tries it just to see what will happen.

(If the resulting action is a surprise to the player, so what. If it’s a surprise to the protagonist, you’ve blown the player’s suspension of disbelief – don’t write the game that way.)

Well, maybe I hit on the unique situation in which this can cause confusion, but as I said that really messed me up in Treasures of a Slaver’s Kingdom (combined with the way the hint system works and a few other things about the game).

At some point, fairly early on in the game, you acquire a piece of stinky bait. There’s also a Pirate Ship, which you can’t get on yet. I said to myself, “Well obviously, once we get onto the Pirate Ship we are going to USE the bait to go fishing.”

Except what you’re supposed to do is USE the bait by the seashore to get swallowed by a sea monster, whereupon you will find the object that you need to get on the Pirate Ship. Ordinarily you would find that by trying to USE everything everywhere, or perhaps by associating the bait with the seashore, but I had already mentally classified the bait as something I would use later. And since the official hints are coded by objects, I was looking at the hints for everything else but the bait. “I know what I’ll do with the bait,” I said, “I just can’t do it yet.”

The other thing is that going through the whole sea monster sequence opens up some combat upgrades that you’re supposed to need to progress in the game – but it’s possible to get farther than you’re really supposed to by tweaking the difficultly level and spamming your way through combat (that is, if you try three or four times you can often win a combat you’re not really strong enough for yet). So I was able to make it a lot farther in the game than I should’ve before I realized I was totally stuck on that puzzle, and I was never getting onto the Pirate Ship.

IIRC I wound up solving this by source-diving.

It’s the unknown unknowns that get you here, not the known unknowns. If an object is obviously going to have a totally crazy use, I’ll try it anywhere that looks plausible. (I think this is what happened with the

dead frog

in that one Trapped game.) But if it looks like it’ll have a predictable use, I might not try it when the crazy effect is intended.

Could be worth looking into Jim Munroe’s Guilded Youth – I believe only four verbs are required to finish the game.