Uncategorized

The algorithm is dead?

About a month ago, WO posted a podcast in which he espoused an interesting new idea: that in games, the “algorithm” is dead, much as in literary analysis, where the author is (to some of us) dead.

This requires a bit of explanation of the structure of games that Gregory is using. Books typically have three components: the author (or creator), the narrative itself, and the reader. Gregory (and I’m not sure of his source here, maybe this is a common distinction in his field) notes four components in interactive works (or “games”, for simplicity): the author, the potential narrative, the realized narrative, and the player.

The potential narrative, as a simple example, is what’s on the CD or in the executable of every game. It’s the set of options, the algorithms, etc., that define what could happen in the game. The realized narrative, is, concisely, a run-through of the game. It’s the story that you “the player” make out of the game.

Now, for the meat of his argument — the “algorithm” is dead. If you listen to his podcast, Gregory argues that the intent of the algorithm might be inconsequential compared to the player’s experience. For instance, a game might have really complex algorithms to take into account what the user says or does when determining what should be next, but if the player feels that they had no agency, no real role, then all those computations were fairly moot. This parallels a reader’s experience while reading a book: if the author intended it to be a commentary on Columbian politics, for instance, but the reader sees it instead as a bitter diatribe about U.S. involvement in Latin America (having, perhaps, read a few chapters of Howard Zinn’s works), then the author’s viewpoint is going to matter less to the reader than their own take of the book.

I think this is a false distinction between the author and the algorithm, for practical purposes. Now, I realize that he’s arguing from a theoretical standpoint, but bear with me. Let’s say that I like the game Sims 2. This probably means that, among other things (like the “genre” and graphics), I like the way the game responds to my input. So, am I going to seek out other games that use Will Wright’s Algorithm #124 for Restroom Needs, or am I going to seek out another Will Wright game?

…Most likely the latter. People identify games by their authors and the supposed intent of those authors, rather than any obscure “algorithm”. Rockstar, for instance, makes games that are well-known to respond a certain way to people’s input. People know this and look for their games, even when it’s Bully instead Grand Theft Auto.

In other words, reputation is attached to the author, not to the algorithm.

So while I agree with Gregory that, theoretically, the algorithm is dead, because the user’s experience will always matter more than the algorithm’s intent, I think the same stands for the author. It’s just that in games, the author gets to suck up both their intent and the intent of their algorithm, making the distinction so barely there for players as to be useless.