The influence of text and typing on parser game design

I’ve been thinking about some of the issues in this essay for some time, and I don’t have all the answers. Part of the reason I’m writing this is to see contrasting opinions.

==The influence of typing==

I’ve noticed that certain types of interactions work very well in parser games, and others do not. Parser games traditionally tend to favor ‘aha’ moment puzzles, exploration, and story-heavy sequences that guide you on what step to take next.

Parser games traditionally do poorly with randomized combat and other CRPG-type things. Kerkerkruip is perhaps the most successful in this area. Parser games also do poorly with any kind of repetitive gameplay (like grinding), even though repetition is central to many other game types.

I’ve realized that both of these facts may be related to the cost of typing. Most game systems let you interact with a single button press, but typing commands takes much more mental and physical effort. This means that every command needs to count. Puzzles like the letter-remover in Counterfeit Monkey or finding the right color to use in Coloratura involve a lot of mental effort followed by a single, high-yield command. On the other hand, there was an IFComp game a few years ago where you had to go down a street and deliver a newspaper to 8 different people, talking to each of them, every day for a week, and this involved a lot of typing of arduous commands.

The exceptions are directional commands (n,e,s,w) and other fast commands (i,x,l). These are so easy to type that you can, in fact, recreate console-type gameplay. I experimented with this in Ether, and many other people have included fast-paced directional commands in their games.

Hadean Lands is a pleasure to play because it avoids the time-cost of typing by automating actions for you. Other games lower time cost by adding short cuts (A for ASK in some Emily Short games, and so on).

Edit: Story-driven games also have a high yield per command, because the commands are easy to guess and generally result in a large amount of exposition.

==Lack of graphical feedback==

The time-cost of combat in text RPGs can be reduced, and has been in many games, but such games still remain unpopular (outside of Kerkerkruip and perhaps a few others). Why? Such games tend to focus on numerical scores, and it can be difficult to gauge the effects of your actions.

This has long been addressed in console games. In those games, critical hits are shown in larger fonts, perhaps shaking the screen or your controller. Colors change to indicate low health or status effects. And it’s easy to judge the effectiveness of attacks when you can see ice shards raining down on a fire elemental.

A picture is worth a thousand words, and without graphical effects, it is difficult for players to keep up with combat. Even Fallen London uses graphics such as bars shooting up or down to emphasize the effects of your actions.

For a parser combat system, or (even easier) for a Twine game, finding a way to visually represent the results of your actions would make any RPG more effective, whether through bolding and italicizing, color choices, font changes, or other text effects.

Edit: Other repetitive actions can have a similar effect; it can be difficult to judge whether the response text changed at all, and so it can incur a large reading cost if you have to repeat an action many times in a row.


The traditional style of parser games is affected by the greater cost of each command, and by the lack of quick visual feedback on actions. Typing shortcuts and visual text effects can enhance a non-traditional parser game.

I’d be interested in hearing any feedback you all have on these ideas.

Repetition is just as bad in choice-based IF.

At first I thought it was endemic to all text-based games, but there are a number of fun “clicker” games with text-based UI, like A Dark Room,.

I now think the problem is prose. ADR has some ASCII art, but the important part of the UI is represented in tables. Compare also Horse Master, in Twine. The tables take the burden off of the text. Without those tables, repetitive games like those would be unreadable.

Hey, that’s a great point; tables really do help simplify things.

Just one thing about CRPGs. Eamon recognised back in the early 80s that in a game where you may want to spam a command (like ATTACK SO AND SO), nothing is faster than pressing a single key to repeat that command. So they made it that you hit return at a blank prompt to repeat the command.

I like this scheme so much I put it in all my games. I do have a bit of an accessibility gripe/mania about unnecessary typing, emphasis on the unnecessary. This came out of major RSI troubles I used to have back in the early 2000s. With the volumes of typing I did, distributing the workload across both arms equally and reducing keystrokes had a big impact. So anytime I can personally reduce typing (don’t type YES or NO, just press Y or N - don’t type G then RETURN - just hit return – etc etc) I try to.


I do agree with your overall conclusion. But one of your assumptions is faulty.

I have to disagree with you here. MUDs are parser games, and they demonstrate just how successful and popular text-based randomized combat, repetitive gameplay, and grinding can be. Dedicated MUD players care just as much about combat systems as graphical MMORPG players do, and I suspect MUDs are more popular today than single-player parser games are (while noting the general niche audience for both kinds of games.)

Pertinently to your overall conclusion, MUDs traditionally employ colored text as a quick way of giving players feedback.

Hey, this is great. What MUDS are popular right now? That’s an area I don’t know much about at all.

I don’t know much about it these days, either. (I was fiercely loyal to GemStone IV and DragonRealms back in the day, and I didn’t branch out much.) But the MUD Connector appears to be alive and well:

Probably the most popular RPG muds right now are the IRE games, particularly Achaea, and then other independent/hobby games like BatMUD, Aardwolf, and Alter Aeon. All of those basically are grindy hack and slash games. There’s a whole other mud genre of roleplay games. is not bad for numbers, though I’ve heard people comment that its stats aren’t accurate.

I’ll second Carolyn’s point that saying parser games don’t do well with RPG gameplay is mostly an issue of familiarity among IF players.

It is possible to replay a large chunk of a text game quickly, by typing fast and skimming the game output. (Not necessarily comfortable, but possible.) I strongly suspect that this allowed games with unwinnable states to persist as the default through the commercial IF era.

In graphical adventure games – with walk animations, animated transitions, and (often) long CD load times – replaying is both slower and more tedious. (You spend your time waiting for the game to let you perform the next click.) This, I believe, was a big motivation for the never-unwinnable style of adventure game design.

I would like to suggest that parser games need, at least some, system world model to be effective. This is because you need responses for the input permutations that you otherwise would not have in, say, choice based games.

The system world model allows mundane operations to be performed entirely by the system rather than preprogrammed.

The more parser there is, the more the player can express concept that need answering. This is a challenge for game design.

Monotony doesn’t kill the desire to play a parser game, unrewarded monotony does. MUDs are being discussed in this thread, and that’s the first thing that also popped into my mind upon reading the original post. I’m a long-time MUD player, and when I think about and design IF systems, they’re heavily influenced by the amount of time I’ve spent grinding away on MUDs. What is so compelling about typing “kill rat” over and over again? It’s because in a multiplayer environment where competition is the spirit of the game, you’re rewarded with experience and maybe gold and maybe an item for killing that rat. Every once in a while you might get a rare rat artifact. If you’re a crafter maybe you can make some nice rat boots with all those corpses. But no matter how you look at it, you’re typing “kill rat” over and over again, and whether or not it seems fun, it at least feels mildly rewarding. Eventually you might start typing “kill dragon”. Slowly, the gameplay is more rewarding because it acknowledges the time and effort you’re putting in.

In a single player IF game, however, this conversation changes entirely. No longer is the gameworld constantly moving whether or not I’m typing. No longer am I competing directly with other players. It doesn’t matter if I kill one rat or kill one hundred, I’m still simultaneously the best and the worst rat slayer in the game, because I’m the only one playing it. There’s no real opportunity for reward beyond the words/prose the game gifts me.

Most popular MMOs (think WoW and GW2) thrive on this same mechanic, but make it pretty with graphics and sound. Players are still willing to kill a hundred rats, but maybe it’s more engaging to do so. But even with all that jazzed up presentation, if it was single-player, it would need to be more rewarding on the cinematics and plot in order to properly reward players for killing all those rats by themselves.

The issue with having cumbersome or repetitive actions in an IF doesn’t stem from the monotony or the weariness (so much typing, ugh!) of the action, it stems from the lack of competition. The world stands still and reacts only to what you do, turn by turn. There’s no chance of anything else happening outside of that window. Alone, that’s boring. Why am I spending every turn killing a rat, when I’d rather be saving the princess? Why not let me save the princess? There’s no inherent competition with another princess-saver. Make me feel like the protagonist.

You could have incredibly repetitive (or at least, very simple) gameplay in IF, but you’d have to design it with a multiplayer angle (think Guncho) in order for it to be rewarding in any way to players. I don’t want to kill a hundred rats by myself, but maybe I’d kill a hundred rats with my friends, or against my friends, or in an environment that is living and breathing around me.

I would be extremely happy for a choice-based designer that included world modeling and a robust inventory system. That can nearly be achieved using Inform 7 and Hybrid Choices at the moment. I think Quest also permits this but I’m not sure.

Choose Your Story has a simple inventory system, but I haven’t been able to test it out extensively because their documentation is piecemeal.

I think the difference between console games and text games is a matter of scope. In a text game, you’ve got a little blinking cursor and you can type anything. In a console game, you have five or six buttons at the most. If you look at it that way, it makes sense that parser games are puzzle-based and console games are repetition-based, because parser games are kind of built upon the idea of discovering what commands work.

On the flip side of that, there’s the world of the game. In a text-based game, you have to describe everything in a room, and each object represents a significant amount of work for the author to write as well as for the player to read. Because of that, there tends to be a very limited number of things you can interact with, and a high percentage of those things will likely be crucial to the story simply because it isn’t efficient to create superfluous stuff. In a console game, on the other hand, you can go around and stab every tree, rock, rat, NPC, cabbage cart, and bright orange yoga ball you see in the hopes that it’ll do something at least visually interesting.

I do agree with your idea that visual interest is good for repetition. If I stab a bright orange yoga ball and it explodes into confetti, that’s fun to watch every time it happens. On the other hand, reading “the yoga ball explodes into bright orange confetti” is only really fun the first time. I think the fact that console games are usually real-time has something to do with it: In a parser game, there’s no need for an action to take time, so in theory you could be type “stab one thousand rats” instead of typing “stab rat” a thousand times. In a console game in real-time, you can watch the stabbing action happen and it adds believability to the action so you accept that you can’t carry it out a thousand times instantly.

This isn’t at all a distinction between “console games” and “text games”, but between adventure games and CRPGs. CRPGs cover everything from Nethack to WoW, and it’s not at all true that they all use five-or-six-button input. (Plenty are turn-based rather than real-time, too.)

CRPGs are a well-developed genre. Repetition is a core principle, but making repetition interesting is what the genre is all about.