This month’s Game Developer magazine contains a look back at LittleBigPlanet by the key members of Media Molecule, reflecting on the development process, what they thought went well, and what they thought went… not so well.
The four page article is a fascinating look into the processes, ideas and company culture at Media Molecule, and gives us some detailed insights into their working methods. Below are a few of the points brought up by Siobhan Reddy, the executive producer of LBP. There are a whole bunch more of in the magazine itself, which should be on shelves now, and a digital version can be purchased here for $3.95.
• Pop-it’s accessibility
Pop-it is the name of the in-game editor. There was a time when pop-it was not a single editor but was a set of tools, a paintbrush for painting, a hair dryer gun to blast away material to sculpt, a decoration placement tool, a paint scrubber, and so on.
We had the project greenlighted with pop-it like this, but afterward, it was one of the first areas to evolve. Initially, we didn’t know how versatile the game was going to be or how we would feel about the editor, but as time went on and the conceptbecame more concrete, we became more confident that people would want a more focused editing experience that prioritized functionality over “cuteness.” We quite literally went back to the drawing board. Our new design became a pre-visualization video, which we later tried out in the game.
Making this decision marked a shift in pop-it in that we now expected people to be able to create bigger and more complex things. Between December and May (2007) the pop-it design was worked through until we rested upon the “stamping” concept we finally shipped with. Back then, this was all happening at thesame time that the team was working out what they wanted to make! The switches (basically cause-and-effect nodes that the player can use to construct mechanisms) were the last big feature to be added into creation toolset, and this unlocked the game design enormously. Creating pop-it was a massive job, and one that had constantly moving goal posts. Amazingly, it was primarily coded by one man, Jonny Hopper.
• Molecular structure
A key part of our company culture is to acknowledge that the production plan doesn’t belong to the production team—it belongs to the creator of the work. Each member must make his or her plan visible to everyone else so that we all know what’s going on and can understand how it affects other areas of the game. Once the team grew to be more than 10 people, we started using a structure that we call “molecules,” which was inspired by Valve’s cabal structure and our own experiences. We agreed that our workforce should be organized into small—and ideally cross-pollinated—molecules: pop-it, character, levels, online technology, and player experience, while our audio designer was included in every molecule as well.
Members changed over time, depending on who was working on what. Some people were in two or three molecules at a time, and each molecule works slightly differently, with the only requirement being that their plans need to be visible and shared. Early on we established something we called Friday features. On Fridays, people show the work they created during the week. This allows the whole team to see what’s going on and helps us end each workweek on a high note.
•Settling on servers
We spent a long time talking with Sony about how to proceed with the server technology to support the online components. We had proposed a LittleBigPlanet -specific solution, but it was rejected because, we were told, improving an existing tried-and-tested server would be safer than relying on an entirely new server that would be a perfect match, but would require a lot of testing and not map well to Sony’s process. As a result, a whole section of the game had to be managed and developed collaboratively between Sony and Media Molecule. The launch has been rocky, but that said, the team working on the online components has worked incredibly hard to support the huge amount of online content that has been generated to date.
•Building a studio and making a game at the same time
Our focus was often split between building the game and establishing the studio. We recently conducted some team surveys and learned that there were a lot of things that worked well enough when we were a very small team, but broke down as we became bigger and busier. These issues ranged from the trivial to the serious: employees not knowing who to get direction from, starting meetings on time, making visible the reasons people failed to come to work, office space not being used efficiently, inconsistency in work-life balance—the list could go on!
There is an expectation from the team, and a good one, that every area of the company be the best it can be. We underestimated the amount of support we needed to ship LittleBigPlanet and run the company to its best simultaneously.
Clicked on both links but was not able to access the article. The site was restricted to subscribers only. Any way we could get a few more excerpts from the article?
Oh yeah, didn’t notice as a friend sent it over, with his subscription link…. I’ll get some more quotes out of it and adjsut the article accordinlgy, sorry about that! 
Really appreciate it. I might have to go ahead and get a subscription, seems like a good magazine. Thanks again for the added info.
