Social Games Journey
In 2010 I was looking to get into social game development. I knew in 2009 that it was a really hot area, but my MMO company, Gazillion Entertainment, was not interested in getting involved in social games – they wanted to stay the course with MMO. As I was not working on a game since I shuttered my MMORPG back in October of 2009, I needed to find something that was going to get me developing again.
My girlfriend (at the time), Brenda Brathwaite (now Romero), was the Creative Director of a social game company called Lolapps. She was working on a game called Critter Island and the company had another game, Critter Town, that wasn't going anywhere and had been rebooted a few times already because there wasn't a full-time game designer on it. Ron Gilbert had been consulting with Lolapps and giving them good direction, but they didn't know how to execute that direction as they had never made a game before.
I decided to take the consulting job of making a game for Lolapps using whatever I could salvage from Critter Town and getting it launched by September 30, 2010. I started on August 1. Yeah, that's only 2 months. Back before 1992, making games in 2 months or less was normal for me, so having a development team should help me get something good done quickly. I had 4 programmers working on the game every waking hour, and I needed them to keep doing that if I was going to ship by September 30.
The first 3 days of August I spent playing Frontierville – I had identified that game as being the most recent Zynga release, and it was designed by a veteran: Brian Reynolds of Civilization, Rise of Nations, and Big Huge Games fame. The most I learned from Frontierville was what I couldn't do in a social game as compared to a typical PC game. The audience's gaming vocabulary was so limited that I needed to make sure I was using known user interface and gameplay mechanics to capture Zynga's audience. I figured out the game's monetization pressure points and economy and decided to emulate them.
The way I design games is not straight-forward. I take all the parameters that I have for the game and my brain synthesizes them all and develops a design almost instantly. Here are the parameters I was working with:
- Must be done in 2 months
- Need to salvage as much code from Critter Town as possible
- We are using a very fast rendering engine, so it should show that off – that means we should be drawing a whole lot of something on the screen
- The player needs to chew through energy, so they must be doing something that encourages it
- There should be a good vs. evil story constantly in action that the player can affect
- The game should feel like they are a positive influence over a slightly negative environment and feel that they have the power to change and beautify their world
- The game had to feel like a storybook fantasy come to life
- There must be a Mentor character that helps the player learn how to play and provides an aspirational layout for the player to be impressed by
- The game needs to make money
- I can call the game whatever I like
The game design came to me instantly. I wrote out the game design so we could start working on it. My codename for the game was Furry Forest. I knew I needed to come up with a more quality title. "Furry Forest" was par for the course with Facebook game names, but I wanted a name that stood out from the crowd and was something you could tell was quality and had a lot of thought behind it.
I told the team the name was not final. I would know the name of the game sometime during development. The game will tell me what its name is when the time is right.
You can read the original design doc here.
Two Heads Are Better
When we had the evil forest emanating scary sounds and terrifying the fair visitors, the visitors would scream and run to the player for comfort. You just needed to click the visitor and you would say, "It's okay." and the visitor would sigh and go back to normal.
Brenda sensed a problem, however. She didn't feel good leaving the game because she was worrying that while she was gone the scary forest was terrifying everyone and they'd all be crying wrecks the next time she logged into the game. I needed to come up with a design solution that worked well within the game's structure.
I decided that having Protectors able to buy and place anywhere would be a great solution. You could choose a cool looking Protector, place them next to the forest, and when the forest made a scary sound the Protector would launch a fireball at the place in the forest where it came from. If customers got scared, they could run to a Protector who would dance around and make them happy again. It provided a great way to attack back at the forest, and added more personality to the game by giving each Protector a dancing animation and special song. It was a huge hit and showed that even though I didn't feel like Brenda did, identifying player's issues and coming up with an intrinsic solution is really important.
This part of development inspired Brenda to write an article about it.
On the 4th day I got the very small team together in a room: 4 programmers, a producer, a UI person, and a lead artist, then proceeded to tell them the entire game design so they could take notes and write down everything needed to make this game come to life: game systems that needed coding, art that needed creating, what needed to be tracked during production, etc. It was all laid down during that meeting and afterward we were off!
I got the team up to 12 people eventually. There wasn't an art team when I started, so I brought Christine MacTernan over from Critter Island as she was under-utilized. For a test, I had her create a storybook fairytale house in a painterly style – she did a great job and got the lead artist position. She had never been a lead before, so I had her sit next to me so I could help her figure it out. I Christine do the art in a painterly storybook-style, which was unlike all Facebook games up to that point. I wanted the game to look distinctive, so we threw all the art away and started over. The user interface art got revamped to match the game's art over the first month.
The team was very hard-working and did a stellar job. To tell the truth, however, making the game was a nightmarish situation because the team had never made a game before, and was second-guessing every decision I made. I told them they needed to trust me – it would all fit together and work well if they just made it the way I was telling them to. I had people telling the coders to change how parts of the game worked because there was a bug in the code while they were making something work and couldn't figure out the bug. I had to stop all kinds of shenanigans that happened like that. It was like running a daycare for a while. This team had never made a game before, so they had no idea. They just needed to follow directions.
One situation was a perfect example of the nightmare scenario of development. I don't remember the details but while Brenda and I were in Culver City for IndieCade, the producer wanted to do something horrendously bad for the game. Brenda spend a solid 7 hours on the phone with him to try to convince him why it was dumb to do what he wanted to do.
September 30 came and the game wasn't fun. It still needed tuning and balancing. The owners were flipping out. I had to calm them down and tell them the game was very, very close but still not there.
I had to argue with the owners to not release the game before it was ready – they were trying on a daily basis after September 30 to just get it out there. You only launch once, I told them, and if they launched without my approval they couldn't use my name for any promotion and I'd have my name taken off the game. They wanted me to enter crunch mode, so I told them okay and then I worked with Brenda every day until 2am.
This was all nightmarish garbage to deal with when on a tight schedule and everyone is crunching. I didn't mind crunch at all, but I did mind the annoying junior antics.
I remember clearly where I was when I came up with the name of the game as clearly as I remember coming up with the word Deathmatch during DOOM's development. I was in the kitchen at home and it just flashed into my head. Brenda was in the living room and I said,"Well, I figured out what the name of the game is. It totally makes sense. It says Where, What and When all in two words. Ravenwood Faire with an e." She said it was perfect.
If you paid attention right there you would have seen that it's spelled Faire with an 'e' at the end. Ravenwood is the place. Faire is what you're making and the spelling tells you it's during medieval times.
At Lolapps I told the team the name of the game. They shuffled around, mumbled, and an assistant producer took it to the owners and they told me they wanted to A/B test the name against other names they thought would be good. I told them they're wasting their time – Ravenwood Faire was a great name. It was quality. It stood out. Even though it was 2 words, the length of the name made it easy to see in the top 30 list on App Data where all the names were short and stupid.
They went ahead and tested it against other names for 3 days. Furry Forest was one. Ravenwood Faire beat them all easily. In admitting defeat they took one last blow at the name. They wanted to remove the 'e' because it made the game sound lame like it was a Renaissance Fair ("Ren-faire"). I told them the 'e' was fine and added a Time element to the name. No, they wanted the 'e' gone for spite. Well, at least my name won out. You know, I had named about 100 games at that point, but these juniors didn't care about experience. Even though they hired me for exactly that. So maddeningly dumb.
While the loading screen was being developed, I made sure the artist put the skeleton of Frontierville's Pioneer Jack in the dirt. That should tell Zynga that we're coming for them. Actually, they thought it was hilarious over at Zynga HQ (I found out later). Who did this little piss-ant company think they were, challenging Frontierville like that?
Social Shut Down
On October 17th, 2010, Facebook shut down Lolapps's games on Facebook. Why? Because of a story in the Wall Street Journal that said a bunch of companies were selling Facebook user's data. The user data didn't come from Facebook, but from a company called RapLeaf that connected Facebook User IDs to user data they had in their own databases. They then sold this data to many game companies. Zynga was doing it as well as several other companies. RapLeaf was just down the street in San Francisco from Lolapps.
Facebook got a letter from Congress that said they better figure it out and straighten up the mess. Facebook had to smack down some companies to show Congress they did something about the problem. Facebook didn't touch Zynga because Zynga accounted for 25% of their income back then. So they decided to punch Lolapps because they didn't like the CEO.
The CEO, on a Sunday, sat on Facebook's front door hoping to talk them into allowing just one game to launch the next day, Ravenwood Fair. The CEO told Facebook that none of RapLeaf's code was in Ravenwood and it was free of the hassle that got them shut down. Facebook decided to be merciful and allow Ravenwood Fair to launch the next day. They warned the CEO that any issues that arose after launch will be dealt with harshly.
The first couple days after Ravenwood's launch on October 19, 2010 was spent priming the pump to get everyone to notice the game. Lolapps was sending messages to all the players of their shut down games and trying to get them to play Ravenwood Fair. It worked.
A week after launch Facebook found an install bar wording issue and shut the game down for a week. Lolapps fixed the issue and began the process of getting ravenwoodfair.com ready just in case they were shut down again. They would put the game on their own site off of Facebook where they couldn't be shut down.