Alright, so the idea is to make a text-based world. What is that? What does it require?
I suppose first I need to decide what it’ll look like. One option would be to make it like this…
You are within a deep forest. There are thick walls of trees to the east and west. There is a
blue ball here. A trail leads
north and
south.
>_
That’s kind of traditional, isn’t it? You type in a command. It wouldn’t be difficult to make the coloured text into links. Hell, since I’m doing this in flash it wouldn’t be difficult to encode the whole thing in html. For that matter I could probably do it all as JavaScript, but I’ll save stuff like that for the future.
One problem with the traditional format is that it doesn’t give you all your options clearly. You’d usually have to use the help command to figure out what all the commands are. I’d like to have options like that clearly displayed on the screen, ready to be clicked. Perhaps you could click or type, at your preference. I want things to be convenient. It’s common to use just one character to operate a command, and annoying when you enter a room and a command is overridden by a local object, or a local object is unexpectedly overwritten by a command you forget about. It’d be nice to incorporate auto-complete, so you know what command you’re putting in like that.
I’m already having a bit of scope creep though, thinking about autocomplete. I should start out with the basics. Rendering a few connected rooms, inside a building, which is insides a closed outside area. Maybe a country Inn. That’s always a great setting.
Another thing I’ve considered using is context menus. When you click on the Blue Ball in my earlier example, what do you do with it? Do you pick it up? Do you kick it? Do you place it on the altar of Grendaline?
When I consider that, I wonder if I should even include the text entry. After all, there’s little to do with it besides guessing at things you can do and having the game say ‘I’m sorry, I don’t know what you mean.’
I want the player to be interacting primarily with the text they’re supposed to be reading, so clicking on it could be a good way of doing that, and using context menus can be good for that. I’ve seen interactive fiction that uses context menus before but it’s been rather horrendously styled, in basic Windows format boxes. Very coarse. One of the things about reading, as with writing, is that you want a very plain interface for the reader to project their imagination on. You don’t want them focused on a dozen widgets and gadgets. I remember some time back, a group of my friends were all excited about a handy web app for writing with no distractions, and a minimum of features. Imagination can be hindered by too much visual noise.
Another thing I’m considering is breaking up the information into seperate sections. This might go against what I was just saying, but I think it’s worth consideration. This would take the D&D approach of having character sheets and reference manuals in front of you; you still have great imagination in that situation.
I’m thinking maybe I could have one area with a text description of the location you’re in, and one with sensual data like smells, moisture, gut feelings, that sort of thing, and another with significant objects and mobiles in the room, along with their positions. I’m not entirely sure how I want to do that.
One way to display the sections would be to put them in seperate quarters of the screen, and filling the extra quarter with hard data and interface stuff like common actions.
Another way I’m thinking of would put the interface and common actions in a thin collumn on the right, with the different text segments rendered into seperate paragraphs. That way it would be clear the first paragraph describes your location, and the second your senses, and the third what is important. It might be good to combine the second and third paragraphs. That’s a thought.
In addition to positioning the information, I wonder about scrolling. Some text games have only one information set on screen at a time, explaining what’s going on at that exact moment. Others will re-write everything as you perform actions, keeping a record of all your actions and everything that happened since you started the game. They both have their flaws and features. I have a personal preference for keeping the information from scrolling. For one thing, it’s easy to lose your place reading when the text scrolls up. On the other hand, if the text disappears when you act again it’s easy to miss something when you push the wrong button. One thing I’ve considered is combining the two, in the multi-pane format. The location only shows you where you are, and not where you’ve been. But at the bottom of the screen there is a record of your actions and interactions, with the important features of the space.
There’s a lot of options for how to format the text display of a world. I would like to experiment with several, to see what works out the best.
I have an appointment in a half hour, but after that I’m thinking of going to the library and putting together some graphical ideas for formatting the text output. Look forward to it!
