Good writing is hard.

I have an ongoing quest to figure out how I feel about hypertext fiction . I admit that I was not thrilled to discover that so many of the entries were Twine or Twine-like pieces in this year’s IFComp. A number of those I wound up liking a lot, so I’m pleased to say that my opinions have been evolving. I’m still much more likely to seek out parser-based IF games over the next ten months than I am to look for more hypertext pieces, though, and here are some of my reasons why.

It starts from this: Good writing is hard.

Old-fashioned non-interactive media are mostly highly filtered. When I browse the fiction shelves at the library, I know that what I see is a tiny, tiny subset of the novels that have been written. Most books are never published at all. This filtering has some downsides, like potentially excluding good books from perspectives that are less mainstream or that the publisher thinks won’t sell. But it also means that the quality of writing in a random library book is higher than the quality of writing in a novel somehow randomly selected from the set of All Novels Written, or (more realistically) from a pile of manuscripts in a publisher’s incoming mail. Published works have also gone through an editing process to improve the flow of the story, tighten up weak points, remove inconsistencies – not to mention just basic proofreading. Nevertheless, I don’t like all books – far from it!

Works published online, such as Twine pieces or other hypertext fiction, don’t have as many filters in front of them. This absolutely has some upsides, primarily allowing writers to have a voice even if they don’t fit a publisher’s mold, but it also means the writing quality in a random online piece is more like that of a random manuscript than a random published book.

With an IF game, I’m willing to forgive less-than-brilliant writing when there’s an additional layer of enjoyment from puzzle-solving. Solving puzzles makes me feel clever, and I like that feeling. I can sometimes get that same feeling from non-interactive pieces. The best example is when I am reading a mystery novel and can guess who the murderer is a little bit before it’s revealed in the story. Even though I know the author intended for me to figure it out, it still gives me that moment of triumph, like the author and I are exchanging a knowing wink. Aside from mysteries, though, a lot of hypertext stories don’t have that same kind of “aha” moment to them. Other puzzles are harder to implement in hypertext than with a parser, at least in the clever-feeling-giving way. It can feel like an accomplishment to realize in a parser game, “Oh! I could go get the olive oil from the kitchen and smear it on the hinges. Then I could sneak through the door without waking up the guard!” It feels a lot less clever to simply click on “Smear the olive oil on the hinges,” even if I can click on other options instead. The moment of realization is lost.

I’ll write more another day about why making a game seems harder with hypertext than with a parser. Right now, let’s consider interactive fiction that doesn’t fall into the “game” category. After all, games are not the only kind of medium I enjoy. Of my favorite hypertext pieces in the comp, I’m not sure if I’d qualify any of them as games, but rather as interactive stories.

It’s hard to write a good story. There are some factors that I believe make creating an interactive story even harder, however.

In the books (and films) whose fiction writing I admire the most, there are ways in which the author crafts an experience for the reader/watcher: foreshadowing, themes and symbolisms that run through the whole piece, overarching messages. The author guides his audience through the story. This is something that sets fiction apart from real life, and what makes fiction satisfying in a different way from history or biography. History and biography can both be fascinating, of course, but there are reasons why people also read novels.

To accomplish these things AND have interactivity is another entire level of complexity. With branching fiction, foreshadowing multiple endings sounds like a contradiction, especially if real branches start early on in the story instead of multiple endings only coming at the climax of the tale. It effectively means the author has to write multiple stories at once. This is part of what made those Choose Your Own Adventure books of the ’80s kind of atrocious from a literary perspective. How can you foreshadow, in early pages, both an ending where you wind up as a pirate captain and an ending where your little brother turns into a dinosaur and eats you? (Of course, this didn’t keep me from devouring those books in elementary school, but no one said my palate at the time was very refined.)

Even if the interactivity is not because of a branching structure, any measure of yielding meaningful control to the reader makes foreshadowing et al more complex. By “meaningful control,” I mean that it’s not enough to let the reader click “Next” or its equivalent, or to offer choices that obviously have no impact on the story or how the reader experiences it. One way to yield some control is to allow the reader to experience the pieces of your story in multiple orders. For this to work, either any order must make sense (hard to structure a narrative with the above literary devices in this case), or you have to figure out a way to limit options until your reader has experienced certain “prerequisites.” This isn’t impossible – Solarium did this very well – but it does require thought and intentionality.

With a regular, non-interactive story, the reader doesn’t expect to control anything that she/he reads. Giving partial control can result in them feeling frustrated by limitations. To write a good hypertext story, it isn’t enough to write an okay story and add some interactive elements. It isn’t even enough to write a good story and add some interactive elements. You have to write a good story and then add good interactive elements.

And that’s hard.

And I realize that I am writing as someone who never having tried to create an interactive story myself.* I have dabbled with Inform 7, but never actually completed a project, and I’ve never touched Twine. I admire people who have put forth the effort, and I’ll keep trying to review their comp entries as honestly and fairly as possible. But while I intend to keep playing parser-based games outside of comps this year, for my non-game fiction fix, I plan to head to the downtown library instead.

*Aside from a Choose Your Own Adventure book I wrote in third grade during Writer’s Workshop, at the height of my obsession with CYOA books, and illustrated with crayon. It was adorable, but from an adult perspective I would not count it in the ranks of great literature.