Warren Spector calls "failure" on Uncharted and other big hits: does he have a point?

One of the first posts I've written for Videogames Beyonder was about Warren Spector and his 'One City Block' design dream - I'm going to translate that one from the original italian text soon, as it's a very interesting description of some of the highest objectives attainable in the realm of game making. And I'm happy to return to Spector in the wake of his PAX Aus 2015 keynote, where he made a distinction between "low", "medium" and "high expression games" based on players' agency and the potential for emerging, unpredicted situations. For instance fighting games, sport games and sandbox experiences (Dishonored, Fallout, Deus Ex: Human Revolution, The Sims are explicitly mentioned) would fall into the high expression category, while things like Uncharted and Heavy Rain would respectively be low and medium affairs.

Taking such successful brands and associating them with a "low rate of expression" has caused a bit of unease among the fans, as if Spector was trying to imply that certain genres were somehow "better" than others. While some of his words about Uncharted suggested that this was not the case...

"It's not that games like these are bad, but they limit your ability to interact with the game world, so the story can unfold the way the storyteller wants it to unfold.
You have very limited ability to express yourself; it's about how you accomplish a predetermined path to get to the next plot point.
It's a great story - a better story than I'll ever tell in a game - but it's not a player story; it's not your story."

 ... some others weren't chosen as wisely, causing an apparent contradiction:

"If all you want to do is show off how clever you are, get out of my medium! Go make a movie or something, because that's what you should be doing."

Game developers are not required to be subtle or elegant, but such statements creates unnecessary friction rather than helping to get the real point of discussion across, and that is: certain genres are somehow better than others at doing what? There's no misinterpreting Spector here, as he talks about offering engagement through chances, variety and reiterated unpredictability, rather than an underlying scripted narrative that justifies the gameplay.

In fighting games, the moveset associated with a character is a tools allowing the players to build their own emerging narrative, match after match: the story you tell your friend about how you beat a boss or a strong contender online, is something the game creators may or may not have predicted. The same goes for RPGs and adventures where open mechanics (= interactions) can be used to resolve particular problems and puzzles in clever and unscripted ways.

Offering engagement through chances, variety
and reiterated unpredictability, rather than
an underlying scripted narrative
that justifies the gameplay

That's what it means "making the most out of the medium" in Spector's terms, which are quite a departure from the kind of experiences he is known for; storytelling is one of the original Deus Ex (2000) strong points after all*. With the rise in computing power for both PCs and consoles, a merge of complex IA and physics could contribute towards making Spector's design dream a reality: games where the objectives are only loosely defined, and the players can use the world rules to attend them while weaving a deeply personal, active and surprising narrative all along.  

Sounds rather futuristic, right? Maybe not that much, but there's no denying the need for extensive (and expensive!) research on how the underlying systems of such games should work, and that's something that is simply not possible in the current AAA space without a radical change of philosophy from the publishers - call it courage if you will. If anything, it should be up to people like Spector to find ways to carry on such research, possibly within the scope of smaller projects: his assertions at PAX Aus may be interpreted as a wake up call to whoever has similar ambitions at heart, but the monetary problem remains.

Nowadays, it is much easier and cost effective to tie specific gameplay systems into rigid (or semi-rigid) scripts, and there's actually nothing wrong with that - both Uncharted and The Walking Dead are compelling experiences that people loves AND expects to spend a finite amount of time with. Their existence has perfectly valid commercial and design reasons, in spite of a scope that is barely comparable to what Spector would aim at - a goal that is unquestionably worth pursuing somehow.

* Quite curiously, one title that better aligns with Spector's recent views would be Thief: The Dark Project (1998), a particularly gameplay-centric experience that he produced until the mid-1997, when he dropped out of Looking Glass Austin to set up Ion Storm: in the words of project director Greg LoPiccolo, the goal in Thief was to "build a type of simulator where object interactions are correct and physics are tied in correctly". Lead designer Jeff Yaus wanted everything in the game to "behave as it should. For example, things that burn will burn, and then is up to the player to burn things, whether or not we've anticipated it" (sources: Wikipedia's "Thief: The Dark Project" and "Warren Spector" entries).

Should you be interested, here's Naughty Dog's timid response to Spector's opinion about Uncharted

"One City Block": the future of RPGs or a far flung utopia?

In very recent times, I stumbled upon an idea by Warren Spector (Deus Ex, Epic Mickey) known in the game development circles as One City Block RPG. The best definition comes from its own creator:

"My ultimate dream is about finding someone fool enough to bankroll the creation of what I call a One City Block RPG, where we simulate a building or a small neighborhood in the greatest detail. I strongly long for worlds with deep interactions. My ultimate game would take place in a single neighborhood, and I'll get to do it one day or another."

In other words, according to Spector, simulating every single detail of the reality and life of a definite space (like a building, for instance) would be more than enough to sustain a high level RPG experience. It's a mighty interesting concept, as it envisions highly interactive stages where every single action the player does, even the most trivial one, can lead to enormous consequences on the surrounding context: the player would be given the chance to manipulate each and every single object in the world at his leisure, in their normal functions or in creative and unusual ways. The same goes for social interactions, with non playing characters (NPCs) able to respond to the player's interactions according to fully simulated personalities and behaviours.

This kind of description immediately sets the One City Block RPG up as a system where a considerable amount of procedural processes takes place within the general boundaries established by the authors (the rules of the world and the goals of the game), generating for the most part unpredictable events. It's a system that implies and glorifies the so called 'emergent narrative' (*) as a mean to diversify and branch out the game experience. Previous examples shows that this is a viable solution (I think about the Radiant AI algorithm used in Bethesda's role playing games, though still limited to the creation of small linear subplots hinged on small sets of variables).

On the other hand, waiting for research to get to a point where deep, rich and nuanced plots can be generated automatically, and in financial compatibility with both the budgets of AAA productions and the final retail price, doesn't seem as viable. The main concerns are that such an achievement may be dozens of years away, with devaluation of human literary talent as an unpleasant side effect, which is certainly not among Spector's desires. A realistic One City Block RPG will have to make do with "simply" marrying the procedurality of its simulations and the high granularity of the interactions with story elements provided by a writer, ensuring the necessary narrative quality, with stricter rules for the game's world.

Luckily enough, Warren Spector is not the only one researching the One City Block RPG - on the contrary, he's at risk of being predated. France based Arkane Studios has stated that similar principles lays at the heart of Arx Fatalis (2002 - PC, Xbox), and more recently, Quantic Dream got remarkably close to the idea with its sandbox adventure Heavy Rain (2010 - PS3). The Yakuza series contitutes a far approximation. Furthermore, 2013 will see Fullbright Company releasing Gone Home, a project solely based on exploration of a solitary, deserted mansion; environmental interaction will be the only way to unravel its mysteries.

* Under the name of 'emergent narrative' goes the events caused by the user's interaction with certain automatisms implemented in the game software. As the effects are entirely dependent on the player's actions, they are not always predictable by the designers and thus constitutes a new, 'emergent' story.