Convergence Culture

Part Three: If you build it, they will come.

Ever since I can remember, I have loved using level editors. I got my start before I was even a teenager building Lode Runner levels on the Vic 20, by my mid twenties I was making levels in UnrealEd to play with friends, and my first game industry job was as a level designer. The game level editor was at least as important to my personal creative development as was the word processor. I believe a good level editor can be as engaging and entertaining as a game itself and - when combined with a great game - has the potential to create a positive feedback loop between playing and creating. This is important if you believe - as I do - that playing and creating are fundamentally the same thing.

But designing and building level editors is hard. On one hand, you want to give players unchecked freedom to design and build any environment they can imagine, from a space station to an African savanna to the Eiffel Tower, or even all of these things at once. Conversely, you also want player created environments to play well within the context of your game. Designing a level editor that connects creators who want to express themselves with players who want tightly designed levels is a challenge I believe can be overcome by making level building into a game itself.

The first time I saw a screenshot of the isometric grid of flowerbeds and fences, cornfields and cattle that we call Farmville, I was stricken with clammy-skinned flashbacks of Floaters and Chrysalids. I needed to know where 'Sniper' Sato was, and who was packing the Blaster Launcher. But regardless of superficial similarities, Farmville is nothing like X-Com. Farmville is essentially a level building game, and X-Com is a game that needs an unending stream of new levels. What a match.

Farmville affords player expression through the design of farms, and a forum for sharing those expressions with others. The actual gameplay of Farmville is either shoddy, evil or non-existent depending on your worldview. X-Com, by contrast, is a hardcore, single-player strategy game in which players express themselves through the way they confront the game challenges. The forum they have for sharing that expression is non-existent (but at least it's not evil).

One of the amazing achievements of X-Com is that it uses a procedural 'level assembler' to construct appropriate levels on demand, offering an infinite variety of terrain in which players battle to repel the alien invasion of Earth. Underlying this 'assembler' are simple rules which determine what makes a valid level. Farmville farms, by contrast, are player generated. Because Farmville explicitly rewards time spent with experience and profit, many Farmville farms are rigorously optimized with dense fields of high-yield crops. But because Farmville is also a platform for socialization and expression, there is also an incredible variety of farms that reject progression entirely; from eclectic need-one-of-everything collections, to idyllic pastoral scenes where some kind of virtual Feng Shui reigns.

In order to give players the feeling of scope required by an X-Com game, a modern remake must confront a difficult challenge. Building an entire planet worth of content by hand is not feasible. Procedural generation risks feeling wooden, and lacking in creative flourish. Relying simply on user generation risks undercutting serious themes with an overwhelming percentage of penis-shaped levels. But a game that incentivizes players to make appropriate X-Com levels can potentially solve this problem.

A casual game for mobile devices or the web that puts players in the role of Mayor, Farmer, or Ranger, and gives them the tools to build all the urban, rural and wilderness landscapes needed, while explicitly rewarded their time spent with experience and profit would quickly generate the needed content. Adding another reward axis selecting for designer-defined 'appropriateness' would lead to player created maps suitable for playing X-Com in. Those maps could then be published to the cloud and pulled down by the game on demand. These casual players cum level designers would be rewarded with experience, gifts, rare items and prestige for designing X-Com appropriate environments. Friend management, sharing, and publish/subscribe tools would begin to bridge the sadly widening gulf between two important groups of players.

It's been almost twenty years since X-Com and an entire generation of similarly hardcore games excluded a mass-market gaming audience a hundred times the size of the audience they chose to service. Today, the pendulum is swinging the other way. The casual revolution - as exemplified by Farmville - is excluding the hardcore gamer. It doesn't need to be this way. We do not have to accept the cynical segregation of diverse audiences when the technology exists today to unite us. By linking our games, we create new domains in which all kinds of players can create, co-operate, compete, collaborate, and ultimately converge.

Comments

[Mike Dexter]

The best level editor I ever tried is the one you did for Far Cry 2.

[nine]

I would be fascinated to see a game try this. The closest thing to *3d* user-generated content forming the main game has been Spore. Are there any other AAA-title examples?

[JOhn]

The guy who architected the engine for far cry, made a game called Cube, wich basically lets the player enter a "create level mode"

[BeamSplashX]

Minecraft is among those ideal tools for level design. You could even have a similar kind of survival mode + architecture style of play set in a nomadic/early mankind period, then "advance the levels in time" to produce layouts for modern game levels.

[Darius K.]

It's a pretty neat idea, though I wonder: once you design an algorithm to determine "appropriateness" of a level created by a user, couldn't you just use that to inform a procedural generation system instead?

Thinking it through, there may be some value to allowing aesthetic wiggle room within the appropriateness measure. A human would (probably) naturally gravitate towards creating an aesthetically pleasing level in ways that no fitness function could predict. I'm thinking maybe:

Design a fitness function for grading X-COM appropriateness of a level, where the function allows room for creative/aesthetic expression without totally killing the fitness score. Combine that with a quick in-game poll: "Did you like the level? [thumbs up] Click here to give detailed feedback on the level [link to a 5-question poll]". Combine a level's fitness value with user ratings and use that as the ultimate rank for a level that gets factored into your level-choosing algorithm for a particular gameplay session. You could even weight the ratings higher than the fitness if a level gets a lot of positive votes, but you could still rely on the fitness function to promote new levels without votes. That way you might avoid the issue of popular levels staying popular and it becoming hard to break into top levels.

[Gareth Mensah]

I for one would really like to see that happening, it would make senses that the merging of casual and hardcore gamers would happen along the lines of gameplay allowing diverse gaming objectives yet promoting cooperation between the different brands of gamers.

If game designers are gods to the virtual worlds they create, then level editors are alike to angels and devils, and following that metaphor, instead of incentives, strong set of rules ought to be established to prevent the generation of degenerated levels/maps.

[hendo]

collaborating a game between casual gamers and avid addicts using inter-media platforms is a grand idea. i'll buy your GTA:GD "spore clothing designer" created clothes from my facebook daily energy fashion industry manager with real time GTA marketing purchases from the streets of player generated maps. these are brilliant ideas, but where are you going with this?

i don't mean to sound like "the jackal" here, but you sound like there is a crime being committed here and no one is doing anything about it. don't get me wrong, i would say that the majority of us agree with your ideas. you are pouring so much passion in to these columns, but to what end? are you rallying the troops or clearing your head?