Graphic vs. Text-based open worlds
Jon Ingold, Creative Director of Inkle Studios, said something interesting in his 2013 GDC talk (well, he said lots of interesting things, but here's one of them): expansive, open world text-based games are cheap to produce because, unlike an open world, Ubisoft-type game, you don't need 50 coders creating a bajillion textures and graphics and playing with light in an enormous sandbox. 1 writer and 1 coder can create an Inkle-style game, and because creating a branching textual narrative is much less time and labor-intensive than creating a branching narrative *and* animating it graphically, text-based open world games will often be far more narratively complicated even than big studio games. But, he said, the thing about text and humans is, we remember phrases quite well. If you describe a character responding to an enemy's attack with the phrase "Quentin ducks, dives to the ground, and rolls out of harm's way," then the next time the enemy attacks two paragraphs later, Quentin will have to do something else, or at least it will have to be described differently. If Quentin again "ducks, dives to the ground, and rolls out of harm's way," Ingold said, players will write in and report that the game has a bug. Whereas a video game with full graphics might have a fight in which the player directs Quentin to duck, dive, and roll in that sequence fifteen times in order to win, the text-based game cannot describe the fight that way or else the player finds it mindless and repetitive.
(below: the Inkle Studios game Sorcery)
This makes a lot of sense, considering the way we teach young humans how to interpret the world. We make them read and then write as best they can in imitation of words they've read or heard. By the time a young human might get around to playing her first text-adventure game, she's become extremely sensitive to quirks in text; she's had to, having passed standardized tests demanding those very skills. This is also why quoting works the way it does—unless you've attended a performing arts school, you are probably better at remembering and saying a line from a book or movie than you are at imitating the actions of an onscreen character.
(The Inkle solution to this problem is pretty fun. They list thirty synonymous/linguistically equivalent words or phrases and then write a script to generate a sensible sequence of moves for every fight. It's insanely unlikely that the same sequence would ever be generated twice, so the player gets reasonable content without getting bored by identical text. I feel like it would be cooler if it were more Markov chain and less fill in the blank (*adjusts monocle*), but hey, I haven't seen the code so maybe it is and I just don't know. (sidebar within sidebar, a Markov chain/Markov process is a way of generating text automatically based on a program's analysis of a corpus of writing. If your corpus is the entire works of Shakespeare, you can write a program that looks at every word in that corpus and keeps track of which words are most likely to follow—maybe "Forsooth" is followed by "I" 22% of the time, "when" 7% of the time, etc. Once the program has been fed enough material, it can start writing "like Shakespeare" by generating statistically probable Shakespearean sentences.))
The thing is, in an Ubisoft-style open world game, for example FarCry 4 (pictured below), players often *do* have a bored/deja vu reaction to the repetitive nature of the activities afforded. Massive open world games tend to get a lot of credit for the amount of content they offer and the huge proliferation of adventures they make available for the player. But the experience of playing one similar quest after another (or, indeed, one open world after another) can bore players. As this thread attests, Ubisoft games can feel like the same adventure with different trimmings has been copy pasted twenty times in a row and that's the game. I'd go further and say that open world games in general often feel identical to me. Only so many times you can collect a special weapon, find bits of information, run from the bad guys, look for more good guys, enjoy gorgeous scenery, and kill a lot of things. FarCry 4 was beautiful, but incredibly boring to me.
The problem isn't that open world games are inherently boring. They're just all the same. It's not even that I mind killing games. I mind the boring, action-hero-style, shallow character affect. Shadow of the Colossus has an amazing narrative that integrates the fights with the character's development; Wander, the main character, doesn't just battle enormous monsters, he battles them 1) with a fascinating grabbing/climbing mechanic that raises all kinds of questions about embodiment and landscape and isn't just about shooting a thing in the right place, and 2) the battle does something to him. He's fighting for a reason, it turns out the reason was morally wrong, and he suffers horribly for it. This is way more interesting to me.
