I have begun work on a game. My initial focus is currently on the game mechanics, but I also have a background for the setting. But I believe that the same mechanics would work in a variety of settings, just as any gameplay genre (shooter, RPG, RTS, etc) can support a variety of different settings.
The motivation for the game design comes from the fact that while I am a professional programmer and software developer, and an experienced writer, I am not an artist (discounting amateur work), and have virtually no experience making real graphical assets for games. More importantly, I am working on this alone, for now, and don't have time to also learn computer illustration and make enough artwork to fill an RPG (even a top-down 2d one).
My starting point was the idea of the multi-user dungeon, or MUD, but for one person. Which led back to old-school, room-based text adventures. But I also have experience with pen-and-paper RPGs. These are primarily text-based, but also use some graphics, like maps and drawings. Some players also use miniatures, but that is primarily to assist in combat mechanics (line of sight, cover, reach, etc).
Unlike classic text-based games, I do not want to require the user to type input. I want a modern user interface with an intuitive layout and modern GUI controls. But the content of the experience is text-based: descriptions of places, characters, objects and events.
In some ways, I think what this is describing is perhaps a bit like Bard's Tale-type games from the late '80's. But, again, with much richer, more modern user interface design and more detailed and nuanced NPC interactions.
The UI would consist of the following main components:
- a live journal: this is a running list of events that are happening to the player, or to other characters nearby
- a description of the current focus: the currently selected object or item, or the current location (if nothing is selected)
- a character inventory screen, probably with the usual paper-doll equipment viewer and a list of contents of backpack and other containers
- a character status screen, including skills, abilities, physical and mental state (health, etc) and other status information
- a context-senstive command section, which contains action buttons relevant to the current focus
The command section and a minimized character status will always be visible. The journal will toggle with the enhanced character status. The focus description will toggle with the player inventory. There will also be some way to toggle between text and maps, or full-window map views, and images of objects (if any).
As it is an RPG, it would also have quests, so it's likely that some kind of Quest UI (in addition to the book-like journal) would also be necessary.
The player's main play activities are what you'd probably expect:
- open/close, lock/unlock and look inside containers
- search for hidden items, hidden containers, hidden doors, traps- move from place to place through available exits
- inspect NPCs/monsters and objects
- pick up or drop objects
- talk to non-hostile mobs
- trade with friendly mobs or merchants
- attack hostile mobs
The game would be turn-based, unless a real-time approach was feasible and the player preferred it. But the idea is that the play experience is non-urgent, relaxing, and mostly made up of exploring the world and meeting and interacting with characters. The challenges would mostly be strategic or involve text or logical puzzles, or conversation. It would not require battle tactics like squad management and terrain utilization, but mostly avoiding unnecessary combat, otherwise dealing with it by learning an enemy's weakness and correctly exploiting it.
One area of possibly originality that I am exploring is that of generalizing the conversation mechanics by using an underlying uniform idea representation consisting of simple atomic action statements, on the order of
<subject> <action> <target> eg "Bob hit goblin"
with optional annotations for additional detail, an adverb characterizing the action, time and place the action occurred, whether the speaker was a witness or heard it from someone else, and any known original witnesses, the likelihood or reliability of the statement. The <action> would be virtually identical to the kinds of actions that players could perform themselves.
I'm also considering whether or how it would work to allow making interrogative statements or predictions of the future, which listeners could then annotate with their opinion of time/place/likelihood.
It's possible that initially the user wouldn't see these mechanics, because they would be too complex and too limited, but if the game was scripted with such action statements in advance, and they had natural language descriptions, it would be a starting point, and NPCs could have at least a rudimentary logic system which would allow them to use knowledge of such statements to determine relationships, trustworthiness, animosity, or other things which would affect what information they would share with the player, or attitudes to trading, to quest giving, to determining quest completion, etc.
This is very theoretical and will quite possibly prove untenable once I start to experiment. But most of the other mechanics are fairly tried and true.
Feedback, criticism, incredulity, similar projects etc. are welcome.
Currently developing for Mac and probable adaptation to iPad.
Most games rely on text to some degree. As Poigahn said, RPGs rely on text a lot. And even spoken dialogue isn't that different from text, in terms of gameplay. So there are lots of examples of how to use text.
As MarkS said, it's important to consider the game environment. Very few people consider a purely text-based user interface to be "normal", unlike when the first text adventures were released. Are you thinking of 100% text-only game? That would seem very antiquated today.
A console-like input system where the user has to type everything is missing out on all the benefits of and knowledge about contemporary UI and UX design. Text input requires memorizing or guessing commands, and can be very frustrating.
I recommend separating the content from the interaction when you design. Even word processors use buttons, tables, lists and other modern UI elements, although the content is predominantly text.
One thing you can't avoid in a text-centric game is that you will have to do a lot of writing. And ideally, the writing will be good on its own: with memorable characters, an evocative setting and convincing dialogue. This will go a long way to ensuring your game is fun.
Ultimately, you have to think of your audience—your players—and what they like, and what they will want to do in the game. Will they want to do a lot of typing, or would they rather click buttons every now and then? How much interaction is required? I mean, how much time is spent reading versus actively inputting commands (either typing or mouse-based)? Different people will like or dislike different designs.