Text Games That Aren’t Parser IF

I know I’ve posted about this a bunch already, but I continue to be addicted to Echo Bazaar, especially as more people I know have taken it up — which means that there are more people with whom to share the social actions and (more importantly) to compare notes about what’s really going on with the setting.

If you’re interested in reading more about the game from a couple of other perspectives, Pissy Little Sausages gives a wonderfully entertaining overview; Dan Shiovitz talks in detail about the game design, with a bunch of suggestions about balancing it for better gameplay (1, 2, 3, 4; possibly more to come). Edited to add: and Sam Kabo Ashwell’s take, too.

As is often the case when I become interested in a game, I wish I had access to a version of the underlying system so I could experiment with telling my own stor(ies) in it.

It is possibly just as well that I don’t right now, though, as I am also still poking around with ChoiceScript. This is something of an experiment (how much of a world model can I fit in there? how does a CYOA interface work with puzzles and a somewhat more game-like feel? if there is a world model of some depth, and the individual choices are more predictable in effect, is the CYOA interface still a bit distancing, a bit unimmersive?).


I’ve gotten to the point now where I acknowledge that the system is remarkably flexible for a CYOA system, but I miss having a full-powered programming language at my disposal. There keep being all these loops, and variables, and it would be nice if I could have some lists to keep track of things.

At the same time, because ChoiceScript works are presented in the browser but in a very bare-bones layout and style, I keep wanting to stop working on the game and start working on CSS to make it look prettier. For straight text in IF I am mostly forced to trust the player to have gotten and be using an interpreter that suits his personal typographical preferences. Once I can publish my Glulx works online, I’ll probably fuss a lot more about the presentation side of things. (Speaking of which, Wei-ju Wu has made some more progress on his browser-based Glulx interpreter. There are rough edges — most annoyingly, to my mind, the fact that the text jumps up to the top before scrolling down at each command entry — but I am impressed by how much of the functionality does work, with graphical windows and whatnot.)

The ChoiceScript project is focusing me a lot on exactly what I do and do not like about IF as an expressive medium.

I miss the groundedness and texture of an IF world. I miss the sense of creating a place. I miss being able to tuck odd details into the scenery, details that aren’t that important (and that the player knows aren’t that important) but that contribute to the worldbuilding nonetheless. I don’t have to implement all the furniture by hand in a ChoiceScript scene, which is a mercy and makes the project a lot faster than it otherwise would be, but also removes a lot of visceral appeal.

Good IF often has some quality of thereness that cannot be described as a purely writerly virtue. This is a point that seasoned static-fiction writers coming (or being brought) to IF do not always grasp; it’s easy to focus on the transcript of a playthrough because that represents the closest thing to the pages of a novel, and say “This prose isn’t very good! It keeps repeating itself! The player has knocked on the wood paneling three times in the last five minutes, and his character isn’t meant to be an idiot!” But this misses something fundamental. Perhaps a tactile fixation is, in some sense, out of character; but the player character is both protagonist of the story and the player’s proxy within the sculptural space of the game world. In graphical videogames it’s impossible to miss that you’re playing inside architecture and alongside sculpture. This is still true of many text games. That’s not necessarily a bad thing: the created space can be aesthetically pleasing and convey meaning in itself.

CYOA doesn’t feel like it has the sculptural/architectural quality, because it doesn’t let me engage directly with things.


At the same time, I’m savoring the freedom to include actions that would be quite hard to hint fairly to a player if we were in parser-land. The story I’m writing up now is one I’ve tried to code three or four times in conventional IF, and each time I’ve been overwhelmed by the scope of what I was trying to produce: it’s a piece very much about large social actions (bribery, betrayal, learning and keeping and giving away secrets, manipulating characters and groups of characters), played out in a very sizable physical space. When I went about this in a conventional way, each scene took so much effort to construct that I gave up on completing the whole. On the other hand, when I tried to kick the interaction up a level to something more abstract (using commands like BLACKMAIL), it was too difficult to keep the player aware of what all his options were.

After running through several disappointing IF prototypes, I started thinking about vastly different interfaces for the piece — like, say, a Flash game with pictures of the various characters, where you’d click on someone to bring up a Sims-like radial menu of things you could do to him at the moment. (Poison? Betray?) But the output has always wanted to be primarily text, and that seemed like a bad fit for a GUI. Moreover, my imagined interface would mean ditching a lot of elements of the original concept, which involved getting certain people together in the same place at the same time in order to do actions that involved both of them. I could never quite convince myself that that extra work and expense (implementing my world model anew in ActionScript, hiring an artist to help with the character portraits and graphic design) would be worthwhile for the awkward hybrid I suspected would result.

The CYOA version is less granular, and more forgiving if I decide I want to summarize something in order to move ahead to the next interesting choice. Something about the command line suggests that you ought to be offering the player control over every snippet of dialogue if he has control over any. But it also takes harder work on the writing to make things feel present and alive, because the level of attention required to click a button is so much lower than the level of attention required to analyze a situation and then type in a fully-formed new instruction that responds to it.

13 thoughts on “Text Games That Aren’t Parser IF

  1. URQ appears to be Windows-only :( There are few enough games that run under Linux, and one thing I’ve always liked about Interactive Fiction is that it’s highly accessible. QSP runs fine (although I haven’t found a txt2gam exporter yet that I don’t have to run under wine), and I’m not sure what Instead is. ChoiceScript at least as the benefit of running directly in a browser, so anyone can use it.

  2. Do you mind if I self-promote here a little and suggest Twine? http://gimcrackd.com/etc/src/ It is my neglected stepchild, but I wonder if you would find it more intuitive.

    Also, I got into Echo Bazaar on your Twitter-recommendation and liked it quite a lot at first, but the grind got unbearable after about 2-3 weeks of steady daily play. Is it worth it to push on through? I was disappointed at how long it seemed to take to make progress on my Ambition — I think the next step was going to take like 20 levels more of one trait to reach.

    • I did try setting up Twine a few months ago, but found that I couldn’t get it to run at all. (This is on Mac OS X.) I’ve now forgotten what the exact problem was.

      That said, I don’t find ChoiceScript counter-intuitive at all: it’s using conditions and gotos and switch statements in a way that’s all very familiar from when I learned BASIC as a kid. I know how to get it to do things, and some of those things would be pretty hard to do in other CYOA formats I’ve seen. The problem is that it becomes more and more unwieldy as I go on, it gets easier to lose my place in the code, and I have more and more global flag variables hanging around.

      Echo Bazaar does pass through periods of infuriating grind. In my opinion it does get better again in the higher levels: after level 50 the stat levels stop getting further apart (though I wish they did that a lot earlier, like say level 30 or so), but more importantly there start to be more interesting interactions between storylets and ambitions, and some of the long-running mysteries about the setting are partially paid off.

      I don’t know, I’ve come close to giving it up a few times; and I’d possibly feel differently about it if I weren’t discussing EBZ with people on ifMUD fairly often, which keeps up the communal interest in what is going on.

      I’m pretty sure that much less grind-y things *could* be done with the platform, though presumably not on a freemium model.

      • Yeah, there was an unfortunate bug that took much too long to fix, where it wouldn’t run on any other version of OS X but the one I built it on. But that is fixed now (fingers crossed).

        I agree that one of the hardest parts of writing a branching narrative is managing that complexity. I used to split sections of story out into separate files, similar to I7′s own chapters/sections/etc. I do, actually, wonder about that. I remember leafing through Aaron Reed’s printed copy of the Blue Lacuna source code at the IF suite at PAX and wondering how in the world he was able to work with a single source code file that large. I know there’s the outline, but still…

    • Chris, may I suggest that you should also self-promote Where We Remain? It’s a game that apparently tells its story in a very interesting way, and it involves some pretty obscure mythological figures. What more would Emily want?

      (I say “apparently” because I kept fat-fingering and getting killed on Easy, so I didn’t actually experience most of the narrative for myself. I should try again.)

  3. “…a Flash game with pictures of the various characters, where you’d click on someone to bring up a Sims-like radial menu of things you could do to him at the moment. (Poison? Betray?) But the output has always wanted to be primarily text, and that seemed like a bad fit for a GUI.”

    You’ve probably thought about this, but I can’t resist suggesting that something like this might be workable within an IF game. Glulx is capable of providing the graphical interface you describe, alongside IF’s standard text interface. Of course, that doesn’t solve the larger design problems you’re wrestling with, or make it but it may open up an interesting space in which a graphical interface can be used to teach people what’s possible in the text interface.

    A quick example to clarify what I mean. A keyword-based IFs like Walker and Silhouette allows both standard IF input and the use of the hyperlinked keyword interface. Clicking on an underlined keyword enters that keyword on the command line on behalf of the player. But a minor change to the interface would allow clicking on those keywords to print a full command, and in this way the player learns something about IF syntax, the ground rules of interaction, and the general possibility space of IF.

    Graphics UIs could potentially offer something similar, particularly in games that have more verbs/types of action than the standard. Again, this may or may not be a good fit for the game you have in mind, but I thought I’d put it out there.

    Of course, I’m also curious to see whether you can make a CYOA that I find interesting! The official ChoiceScript games definitely haven’t been doing it for me…

    • I did think about doing something with a Glulx GUI, yeah. I was put off by the lack of browser support for Glulx, since part of the point here would be to try out alternate interfaces on people who aren’t hardcore IF-players; and by the fact that I haven’t been totally happy with my attempts to do anything more than the most basic illustrations in Glulx sidebars. That said, the tools you’ve been working on may very well address that concern, and browser Glulx support is coming along fast as well.

      • Hm, I’ve been thinking about what to include as a simple UI example for the documentation for Glimmr. Maybe I’ll do something similar to your example (i.e., a pop-up radial menu), since that may be something that a number of folks could adapt and use…

  4. Very interested to hear that you’re playing around with ChoiceScript.

    Not sure exactly what you’re looking for, but you might check out my very short, unfinished (dead ends) version of Little Red Riding Hood:

    http://doiop.com/red_example

    Full address: http://horacetorys.weebly.com/uploads/4/0/8/8/4088687/red_example.html

    It’s all hand-coded in Javascript, but really just using two or three coding bits over and over. It allows you to throw in small details, exploration, etc. Doesn’t solve the problem of a million variables and things. With some imagination and patience, a very satisfying world model could be constructed, while still being cross-platform and intuitive for new players, and engaging for veterans. I also think it de-emphasizes that it’s a game, and reads more like the page of a book.

  5. Pingback: The week in IF #6 « IF URLs

  6. Pingback: IF: Choice of Broadsides (Choice of Games)

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s