The latest incarnation of Vikingr is on the iPad. I’m now designing from the bottom up rather than the top down, thinking about the nuts and bolts of a satisfying core combat loop rather than the abstract motivations of narrative. Combat in Vikingr is risky and in no way a default decision, just as in the historical record (trade was a wiser move when the opponent was formidable, though less glamorous). To provide a less micro-managey outer loop, players will control a whole household rather than an individual, and this change should also help mitigate the displeasure that comes from loss of life. Once the decision to raid is made, the player selects a target (one of the Viking lands: Denmark, Iceland, Norway, or Sweden) and a match is made with another active player through Apple’s Game Center service.

During a raid, each player directs a group of Vikings in battle by adjusting their stance (reckless, prudent, cautious, or cowardly) and moving them around a hex grid in soft real-time. Characters may act whenever their fatigue counter reaches 0. Any action increases fatigue, and fatigue decreases on its own over time. When a character in an attacking stance comes near an enemy, they fight until one of them dies or moves out of the way.

On average, reckless attacks beat a conservative, prudent attack; prudent strikes best a cautious defense; and a cautious defense finds opportunity in a reckless offensive. Battle is not about probable outcomes, however: an attacker who strikes from behind or merely gets a lucky hit can pull out a victory despite a poor choice of stance. The Valkyries don’t always choose the best man to win the fight; sometimes they want him in Valhalla. Cowardly flight will lose to any attack, and it provides a much greater mobility along with its social stigma.

Vikingr's basic movement interface

On the life simulation side, the fine-grained week structure may be replaced with twelve months or four seasons per real-world day or, possibly, a system where characters become occupied and undirectable for a certain number of real-world hours while undertaking tasks such as farming, crafting, or travel.

I hope to implement the core risk/reward activity of combat and the basics of the other elements needed to support it (deciding when and where to raid and the basics of the life simulation interface) by the end of the month.

Originally, I designed Vikingr as a sort of “Viking Wars”—a turn-based, asynchronous social game in the vein of BBS or text-heavy web games (e.g. Legend of the Red Dragon, Archmage) or, more recently, hybridized graphical games like Mafia Wars. This would progress at a pace of one year per day, granting players 52 turns per day, and subject characters to the ravages of age and permanent death so that players would move on to a new character roughly once per month. I would also introduce additional limitations to emphasize the quotidian nature of most of Viking life: For example, raiding and long-distance trade expeditions had to be conducted when little could be done at home, so activities like farming would be performed during springtime turns and the expeditions would take place in summer.

My goal was to take the above, sprinkle in some social mechanics, and play up the fact that everyone was contributing to one massive shared history with their gameplay actions. By giving players a little context for these histories, we could be off to the races with a novel web-game that doesn’t just permit player-generated stories to emerge but acknowledges them and incorporates them into its gameplay.

Unfortunately for me, when I started prototyping all this—a website, a server in my programming language of choice, a bunch of equations and formulae, decisions made for scalability’s sake, research into hosting and APIs and this and that, lists of features to implement—I realized that I was thinking like an engineer, not a game designer. So I went to paper and did a sort of simulation RPG about budgeting one’s time between preparing for raids and going on them. This felt far more like accountancy or event planning than viking, so it forced me to reexamine my base assumptions.

Ultimately, I came to an inescapable conclusion: “Take a new technology and apply it to an existing genre in an uncommon fictional context” is an incremental way of thinking. There’s a time and a place for incremental improvement, but a 12-month MFA thesis is emphatically not it. So I went back to the drawing board: If I wanted players to create and share stories about being Vikings (and to use my game and my technology to do it), what must I help them to feel? The answer to that had to merge historical reality with literary fantasy.

At one point, I had planned to produce several games for my thesis to conduct a few experiments about the Ensemble computer-aided storytelling system. In the interest of providing a more complete experience and having a stronger presentation at the winter and summer shows, however, I chose to focus on a single project. To validate the hypothesis that Ensemble is worth designing around, Vikingr will be tested in two variants: One with a direct causal relationship via the Ensemble system between gameplay and narrative outcomes, and one with a weaker or even arbitrary link between the two.

So, what is Vikingr? The word is from Old Norse and means what you’d expect: a raider of the Viking sort. The topic has always interested me—not in a Wagnerian noble savage way, and certainly not in a video game-y mead-drinking hammer-swinging barbarian way—especially since learning that viking is an activity; used as a noun (vikingr) it means “someone who viks”. It denotes a profession, not an ethnic group, and a part-time profession at that: viking is a get-rich-quick scheme, a way to raid or trade in far-off markets to boost one’s income and prestige while waiting for the crops to grow in early summer or mid-fall.

Vikingr is a narrative-driven social game where players compete to accumulate wealth and renown by farming, crafting, trading and raiding. Players form bonds with each other through marriage, the fostering of children, and exchanges of gifts, and such alliances represent important obligations. Players may opt to provide additional narrative details about their characters using Ensemble, and these will be merged with the narrative facts and events that emerge during play—for example, a character who frequently tells stories to entertain guests at parties may be considered a skald, a sort of poet; another who is skilled at battle could be called “war-wise”; a Viking skilled at both would earn the title “warrior-poet”.

This emphasis on the individuality of specific player-controlled characters is an attempt to evoke the so-called X-Com Effect, where randomly-generated characters can become narratively significant to the player due to small differences in appearance and statistics and the player’s personal history with using those characters. Vikingr hopes to go even further by systematically recognizing these gameplay histories (and thus the likely player-defined goals and stories behind these characters) using an achievements-like mechanism that feeds back into further aesthetic and game-mechanical systems.