Weekly Links #50

It had to happen sooner or later. This week I could barely scrounge up a couple of links, and I have little to write about the most important of them. To wit, the already famous Twine has reached version 2.0 — a huge leap forward as it now runs in any modern web browser, making it available on new platforms such as Linux and Android, and more casually accessible to just about everyone.

And since we're talking Twine, remember when the default interface for interactive fiction wasn't hyperlinks, but a command parser? Turns out, experiments are ongoing, as Emily Short points out on her blog. But while experiments are good as a general rule, the examples in the article fail to get me excited, for reasons I'll explain below.

For now, let's talk a little about card games.

I've looked at the issue before, years ago, but nothing came out of it. Perhaps because I focused on implementation details rather than what makes playing cards what they are. Ironically, I came to this realization by thinking of them as objects of art. After all, when you can draw anything on any card, or nothing at all, what makes them playing cards? Are StoryNexus titles card games?

I say no, not really. If you look at the standard deck, or a Tarot deck, Mahjongg stones, even Magic: the Gathering, you'll notice they all have a few things in common. Cards are ordered along two or three dimensions, say rank and suit, or equivalent. There may be some court cards that stand apart from the others, or not. You can have a set of trump cards on top, themselves ordered — or multiple sets of "jokers" that don't fit in with the rest. Fiddle with any of these parameters, and you get something new. But you need something to fiddle with in the first place...

For now I've only toyed with a couple of concepts for educational playing cards — nothing too interesting, but that's the way it is at first. Still, this time I intend to keep at it — I don't do enough stuff in the physical world as it is.

But I was telling you about user interfaces in interactive fiction. Adventure, the very first game in the genre, was driven by a command line parser mostly because it had to run on teletypes. Subsequent titles kept the parser because it had qualities above and beyond that (much like Unix is still a good idea, all these decades later). But after a while, it all became stale, and a lot of people could never wrap their heads around a command line interface at all. Plus, the raise of mobile devices pretty much forced authors to move on -- even with helpers, typing on a tiny touchscreen is just too clumsy.

Naturally, the first alternative explored was the hyperlink, which harkens back to good old gamebooks as well as being a natural match for the modern Web. But a choice-based interface tends to show its limitations a lot sooner than even parsers. What to do?

Much has been tried in the hope of bridging the gap. Clickable word lists were one of the earliest — a case of missing the point as it turned out. Typing keywords? Dragging verbs on top on nouns? They're isomorphic to hyperlinks and two-word parsers, respectively. (Whether there's a world model underneath or not is another story.) I do seem to remember an experimental game that involved rearranging passages of text to create your own narrative, but I'm not sure how it worked.

And you know what? None of this has helped interactive fiction get any better. Oh, the genre is popular again, and even sells, but that's mostly because text adventures are fresh to an audience born after 1990. Nobody in mainstream gaming seems aware that there was anything at all between Zork and Twine. Which is too bad, because interactive fiction authors have put a lot of effort into researching level design, natural language processing, puzzle structure, AI...

None of which have led to any real breakthroughs. What we have nowadays is a highly refined artistic medium that just can't seem to say anything new anymore, no matter how hard people are trying -- and there are brilliant people working on it. We've hit a glass ceiling at some point, and it took years for the community to acknowledge that there's a problem at all, let alone its nature.

To be fair, all computer games suffer from the same problem. But games in other genres can be compelling without saying anything.

Text adventures don't have this luxury. What, with being made of words and stuff. And after many years, I might have an idea or two about how to move forward.

But that's a story for another blog post. Merry Christmas!