IFComp 2013 and the Twine Revolution

It’s been several months since I’ve posted here, but I’d really like to get back into the habit again. Luckily, the Comp is here to force the issue, late though I am to start it. This year, there are 35 entries. Strikingly, more than half of them — 19, to be precise — are listed as “Web-Based”. I’m told that about 3/4 of these were written in Twine.

And that’s as good a reason as any to go off on a rant that’s been building inside me for a while: the Twine rant. This is in large part a reaction — a visceral reaction, largely unjustified for reasons I’ll describe — to a provocative essay by Porpentine, portraying Twine as a revolutionary force against the gatekeepers and hegemons of capitalism, elitism, and the dominance of the parser-based IF community. As a member of the IF community, this stance really took me by surprise. Dominance? I thought we were the scrappy underdogs. What are we dominating? Certainly not the games industry.

As far as I can tell, our domination is of the term “Interactive Fiction”. Porpentine laments the fact that googling “interactive fiction” yields pages of results about parser-based IF, but little about hypertext. As she puts it, “Some say non-parser isn’t interactive fiction. If the words can be interacted with, it’s interactive fiction.”

Now, my gut reaction to this is that it makes as much sense as Bobby Fisher’s infamous claim that he shouldn’t be considered anti-semitic because Arabs are semites too. Words have histories, and don’t always mean what they sound like they should mean. In the case of “interactive fiction”, Infocom gave it a much more specific meaning three decades ago — essentially, “text adventures, but said with less embarrassment” — and that usage stuck well enough that the hobbyist community that sprung up after Infocom’s demise still used the term, and named the Usenet newsgroups where most of the early information-sharing took place “rec.arts.int-fiction” and “rec.games.int-fiction”. 1 Actually, there’s at least one usage of the term “interactive fiction” before Infocom’s: Robert Lafore’s Interactive Fiction: Six Micro Stories, published by Adventure International. I played this on TRS-80 back in the day. The interface here was command-line-based, I don’t think it actually qualifies as a parser, because it made no attempt at grammatical analysis: it just scanned the input text for certain key words and ignored everything else once it found one. For example, “No problem” would be interpreted as a refusal, because it contains the word “no”.

Not that there wasn’t disagreement about terminology back then! Every once in a while, someone would wander into the newsgroups and tell us all that the stuff we were writing and playing wasn’t really interactive, because it didn’t meet some criterion he had made up — say, for example, that it only counts as interactive if there’s significant plot branches and multiple endings — and that therefore we should all be doing things his way instead. These attempts at redefining our art out from under us seldom went over well, and that’s part of why I’m so defensive about this. Magnus Olsson, in a memorable 1997 raif post, compared the IF community to a jazz club, and these bossy newcomers to someone who goes to the club to tell everyone in the club that jazz is dead and they should all be using pre-programmed synthesizers instead. No matter how good his arguments are, they’re going to be ignored by the people who showed up because they like jazz.

I kind of felt like Olsson was giving these guys too much credit, though. Some years later, I came up with another metaphor that I think better captured the more cluelessly bossy newcomers: Imagine someone who has never heard of baseball. On hearing the term for the first time, this person tries to imagine what it might be like on the basis of the name alone, and comes to the conclusion that it’s probably a soccer-like ball game involving RTS-style base-building, creating structures to protect your goal from the enemy. Intrigued by this admittedly rather interesting idea, he goes off to see a baseball game, only to find that it’s nothing like what he imagined. And so he yells at the players that they’re doing it all wrong.

So anyway, the Twine Revolution strikes me as kind of similar to this scenario, except that instead of demanding that existing baseball teams abandon their ways, a whole bunch of people who are interested in RTS-style baseball got together and formed a league. And they’re getting a lot of press coverage, and people are watching their matches, and they all refer to this new sport as “baseball”. For some of them, it’s the only kind of baseball they know. And somewhere, there are players of traditional baseball (which is what they have to call it now, “traditional baseball”) looking on in bemusement and confusion.

Now, that was my gut reaction. Here’s why it’s bunk. For starters, the IF community is a tiny, tiny thing. Even at its height, the Comp never got more than a few hundred people to vote on it. That is not enough people to dictate common usage. Besides, the whole bossy newcomer scenario described above should have been a pretty good indication that “interactive fiction” was a poor choice of term for what we were doing in the first place, and if people even today are googling the term and expecting hyperfiction, well, something is wrong. If a small group like the IF community disagrees with the rest of the world about what “interactive fiction” means, we can’t reasonably tell the rest of the world that they’re the mistaken ones. Which leads to my second point: we don’t actually disagree with them. There have been plenty of Comp entries with hypertext interfaces in the past. Andrew Plotkin’s last-released work of IF has an essentially hypertext interface, although it does allow you to type the words instead of clicking on them if you really want to. Emily Short’s current project, a highly procedural and simulationist system called Versu, has a touchscreen-friendly contextual menu interface rather than a text parser. And for years now, every time “CYOA” interfaces come up, I’ve been pointing out that a simple interface doesn’t have to mean a simple world model. There has even been parser-based IF that I thought would have been improved by taking the parser away.

But I will draw one line. Regardless of the interface, the world model — the thing you interact with — is an important part of IF. I wouldn’t call static hypertext 2That is, stuff that could be represented entirely in HTML without any scripts. interactive in any meaningful way, even when it represents a branching story. You don’t act on such a work and change its state, you just look at different parts of it, change which already-present page is being displayed. That’s exactly the level of interactivity you have with a normal, linear book. Twine is of course capable of much deeper interaction than this. But static hypertext is kind of Twine’s default state, just as text-adventure-with-directional-movement-and-inventory is Inform’s. It’s the easiest thing to do in Twine, and therefore, as far as I’ve seen, it’s all that the majority of works in Twine do. And all power to them, really. If static hyperfiction is all you want to make, Twine is a pretty good way to make it. The Twine community certainly won’t disapprove. But to call the result “interactive fiction” — can I at least fight this? Or am I edging into bossy-newcomer territory here?

Ultimately, is the terminology even worth fighting over at all?

Well, there’s at least one place where it makes a substantial difference: the Comp. Because that’s where the inevitable static hypertext and the text adventures will be judged against each other. Which is kind of like having a baseball match where one side is playing traditional-style and the other side is playing RTS-style.

There have been static hypertext works entered into the Comp before. (Heck, there was one last year.) But they were always entered by isolated individuals, and voted low by irritated players, and forgotten shortly afterward. The Twine Revolution is bigger than that, and could possibly balkanize the Comp. If the Twine authors are out in force this year, it stands to reason that the Twine players will be too. This could be the first time that there’s a substantial faction of Comp voters who just don’t like parser-based IF at all, just as there is surely a substantial portion (myself among them) who will balk at static hypertext. If they’ve read Porpentine’s inflammatory rhetoric (and forgotten the bits where she says she actually “likes parser”), they may even possibly regard the parser as an enemy, as something to be defeated by staging a mass invasion and voting all the Twine games to the top. And that would make me sad, just like the patrons of Olsson’s jazz club would be disappointed if they suddenly found that the “jazz is dead” guy and his friends were not only booking half the performance slots but winning awards at jazz festivals for their preprogrammed synthesizer recordings.

But in the end, as I said in a previous post, the Comp isn’t where the really exciting stuff in IF is going on these days anyway. And I still intend to get around to talking more about that in a month or so, once my Comp writeups are done.

   [ + ]

1. Actually, there’s at least one usage of the term “interactive fiction” before Infocom’s: Robert Lafore’s Interactive Fiction: Six Micro Stories, published by Adventure International. I played this on TRS-80 back in the day. The interface here was command-line-based, I don’t think it actually qualifies as a parser, because it made no attempt at grammatical analysis: it just scanned the input text for certain key words and ignored everything else once it found one. For example, “No problem” would be interpreted as a refusal, because it contains the word “no”.
2. That is, stuff that could be represented entirely in HTML without any scripts.

6 Comments so far

  1. Harbour Master on 22 Oct 2013

    I suppose I hadn’t really foreseen Twine dominating the IFComp simply through force of numbers, but it makes sense considering the lower bar of entry for both development and play. So the question is whether IFComp needs categorisation so as not to be swamped or end up setting apples against oranges; the problem there is how far you would go with categorisation, because once you’ve opened Pandora’s Box, it could get messy pretty quickly. Parser has generally been in decline so it feels to me more like an endangered species rather than some gated community with guns at the ready. Parser has barely featured on gaming sites in recent years – even something as progressive as RPS – although some indie voices like Richard Perrin have tried to push these works back into the limelight.

    I feel sad and guilty. I just haven’t played enough parser games in the last few years even though I love parser. If you’re not part of the solution…

  2. Emily Short on 22 Oct 2013

    I see no evidence that there’s likely to be a massive downvote of parser IF — if that does happen, it’ll be something to talk about then, and in that circumstance (pretty much *only* in that circumstance) I could see an argument for making a parser-games-only comp. But as things stand, I think we benefit from communication with other related types of interactive storytelling, and it seems premature to fret about this.

    I also don’t think the gaming press is hostile to parser IF these days — I was surprised by how much uptake Counterfeit Monkey got despite my having done nothing to promote it to games journalists.

    I do think that a lot of traditionally parser-oriented authors are experimenting with new forms. Jon Ingold, Aaron Reed, Andrew Plotkin, and I are all working on related things as a living or academic pursuit; it’s just that the majority of what we’re doing is using some other interface than the parser, though often over the top of a complex world model. So a lot of the energy is directed to adjacent genres and styles.

  3. Mark on 27 Oct 2013

    In mathematical terms, there’s not much difference between a dynamic state (an “interactive fiction” model) and an exhaustion of static states (a “hyperfiction” model), as either can be made to be entirely indistinguishable from the other. That’s not a very useful argument, though, as the effort required in making hyperfiction resemble IF is usually greater than implementing it as IF in the first place.

    If the competition were to be divided into not-too-severely-misnamed “stateful” and “stateless” categories, or “simulated” and “unsimulated,” that might be useful going forward. The distinction is fuzzy, but it’s also future-proof, unlike differentiating them based on whether the game uses a parser or not.

  4. Jason Dyer on 3 Nov 2013

    I seem to remember rec.arts.int-fiction was originally started (in 1990 or so?) for hypertext.

  5. Storycade: Chris Klimas War, Pestilence, Famine, Death and Twine-StoryCade on 4 Apr 2014

    […] Carl Muckenhoupt, who maintains the venerable Baf’s Guide to the Interactive Fiction Archive, describes the traditionalist position best. To excerpt: Magnus Olsson, in a memorable 1997 raif post, compared the IF community to a jazz […]

  6. War, Pestilence, Famine, Death, and Twine – Chris Klimas on 11 Dec 2016

    […] Carl Muckenhoupt, who maintains the venerable Baf’s Guide to the Interactive Fiction Archive, describes the traditionalist position best. To […]

Leave a reply