Dancey hopes to clarify his game's outlook today. Read on for his dev blog, in which he discusses what he means by "minimum viable product," distances Pathfinder from the cripplingly expensive graphical arms race plaguing the industry, and elaborates on just how Goblinworks plans to roll out this "crowdforged" MMO.
I'm sure this is a cost effective approach, and it's probably ideal for "small, agile" indie development teams like GoblinWorks. Will it make for a full-featured persistent world, though, or just another combat lobby?We thought this was a great opportunity to talk about our plan in some detail because I think a lot of Massively's readers will find it really interesting.
Becoming a "minimum viable product"
As we have been working on Pathfinder Online, we discovered Eric Ries' work on "lean startups" and his thoughts about the minimum viable product. We've adopted his terminology as a way of communicating our plans to the public. To summarize, Ries defines an MVP as "that version of a new product which allows a team to collect the maximum amount of validated learning about customers with the least effort."
The point of an MVP is not to make one and stop work. The point is to make one and use it as the very first point of interaction with real customers, and then start iterating on the design by tightly integrating the users' feedback into the development of new features and expansion of existing features. It's a perfect fit for a sandbox MMO. If we make a successful MVP, we know we'll be able to continuously add new features and game systems essentially forever, making the game more and more complex over time, more and more beautiful, and larger and larger in scope. A "minimum viable product" is not a "minimum" product. The key word is "viable," not "minimum." To make a viable MMO, we need to deliver a compelling game loop and a way to generate revenue.
The core game loop
To make our MVP, we had to figure out what the smallest number of features were needed to implement a compelling game loop with meaningful human interactions. We decided that our loop would have two interconnected segments.
- Segment one would be finding monsters in the world, killing them, and taking their stuff.
- Segment two would be finding resources, harvesting those resources, and turning the resources into crafted goods.
- The interconnection would be that the stuff one character crafted would make another character better at killing monsters.
- Killing another character and taking its stuff would be a shortcut.
Player vs. player conflict
You can't study the history of the MMO genre and not be familiar with the way that PvP has warped players' experiences and expectations since the very start. Jef gets right to the heart of the matter in his comments about Pathfinder Online:
The thinking seems to be that all a dev team need do is build a combat engine and associated conflict systems and then throw players into the mix and let them make of it what they will.We think that PvP is a critical element needed to make the game loop function. And we think that PvP is a fascinating and incredibly rich mechanic to stimulate meaningful human interaction. We accept that there are hazards and risks to including it in our game, but we think the upside of succeeding in restoring PvP as a desirable feature of MMOs is worth the risk.
We didn't define PvP and a combat lobby as "viable." It's not enough (in our opinion) to start generating validated learning about our game, and it's not compelling enough to sustain our game over the long term.
Exploration, adventure, development, and domination
Exploration and Adventure are the focus of the first segment of our game loop. We defined "minimum" for this segment as having several different monstrous opponents with different strengths and weaknesses appear in the game world. We want to have a game world that is large enough that players have to spend time travelling to see it all and find all the interesting features we've embedded in it, so at minimum it needs to be big enough that you have to work to see it all.
Development is the focus of the second segment of the game loop. We focused on our economic system to express the minimum viable nature of this segment. There will be a small variety of harvestable resources in the game world that the characters can find and extract. Those resources will be processed by player characters into crafted items of value -- of a wide enough range that a single character won't be able to produce all of them in parallel.
Our "complication" represents the initial implementation of Domination. Characters will need to transport harvested resources from the wilderness back to crafting facilities in towns. Characters will need to equip themselves with upgraded gear to confront the tougher and more rewarding monsters. These characters will be targets for those who want to use the "shortcut." That in turn implies the potential for other characters to become guards and enforcers, and all the fractal depth that arises as players and groups of players organize to react to the actions of other players and groups of players.
Graphics, animation, and visual effects
Our game loop requires characters to give the player an interface to the game world. Character visuals are a place where many MMORPGs excel. But for the purpose of our MVP, we know that we don't need to invest the time and resources in making our characters as detailed and distinctive as the competition. If our MVP is successful, we will be able to iterate on that aspect of the game forever. But to begin we only need to give our players the tip of that iceberg.
This is one of the biggest risks of our MVP approach. Players have been taught to judge the quality of a videogame based on the quality of its graphics. Players expect a new game to meet, or exceed, the graphic quality of the state of the art. This is the reason that MMO development budgets have spiraled ever-upward: The costs of the graphics, animations, and programming required to integrate the visuals with physics, collision, and realtime interaction require large expensive teams and a lot of development time. We accept that tradeoff. To make a game fast, on a small budget, we aren't going to engage in the visual quality arms race.
The velvet rope
We said earlier that in order to make an MVP, we need to deliver a compelling game loop, and a way to generate revenue. When we make our MVP available to the players, we're beginning a stage of development we call Early Enrollment. Early Enrollment is not a "beta test" in the sense that term was once commonly used. Instead, it's a process of helping to rapidly evolve the MVP into a more fully featured product through a process we call Crowdforging, a tightly coupled feedback loop between the community and the development team. Our models for this process are projects like Minecraft and Gmail.
The experience of our Early Enrollees is going to be quite different from the way most players engage with an MMO. We are going to limit the number of players we allow in to the game for several months, perhaps as long as a year. We think we can support approximately 20,000 players with the MVP, but we're going to cautiously approach even that number, holding access to the game below that threshold for the first several months.
Our Early Enrollees will secure their places in the queue to gain access to the game through our crowdfunding systems. The first month of Early Enrollment is allocated to the backers of our Kickstarter. Soon we'll begin selling access to the game for future months via our own platform on our own website. These packages will provide purchasers with the game itself as well as some number of months of game time, varying based on the price of the package they select. The Early Enrollees will also have access to special discussion forums, polls, in-game access to the development team, and other tools that will enable the process of Crowdforging. Together, we'll begin to iterate on the MVP, with the community helping us decide the best use of our time and resources to expand the game.
Community excitement is rising, and more and more people are becoming aware of what we're attempting to do. The support we've received has been incredible! Please come join the ongoing conversation on our message board forums and check out the development blogs we've been publishing over the past year and a half to read more details about all these topics and more!
Thanks very much to Ryan Dancey for sharing his insight with Massively.