When we made Eressea: the third age, the design took place with four different people in three different countries. I think the development went really well, because of some new technology that hasn’t been around when we made the original game ten years ago. Our main collaboration tools were Skype, Google Docs, Mediawiki and email, and I find that each of them helped us in its own way.
Skype is wonderful. Even though there is always somebody who cannot get the voice chat to work, you can still conference him in via a landline at almost no cost, and we found speaking way more efficient than typing. Having your hands free also lets you take notes, browse email archives or do other things without losing touch with the conversation because yo’re not constantly looking at the chat window. We ended up with one collective chat every week in the beginning, and I think it really helped get everyone psyched about the project.
The fact that Skype group conversations don’t disappear after everyone logs off, and that they give you logs for what happened in your absence also helps bring both the asynchronous development and synchronized meeting-style chats together.
Google Docs were probably our biggest help. Initially I though of setting up adesign wiki, but it turns out that a lot of game design is easier to present in a spreadsheet, and our master design notes ended up as two-dozen sheetsin a single document. Another advantage over the wiki: Thanks to google gears, I could take the documents with me when I went offline on the plane or in a park. What they lack is change tracking, so you need to communicate a little about that.
Where the Google docs shine is in collaborative editing. We would constantly have the main document open during our chats, and edit them together to suggest something, to tweak it, discuss it. Out of all the electronic whiteboard software I have used, Google’s spreadsheet is hands down the most useful product.
We did not write very many text documents, but we did use the forms a couple of times. Being four designers that worked mostly on consensus, we would for example set up a form where everyone could fill in values for recruitment costs, each fill them out, and then just use our average. We also ended up using them as our signup page, and plan to use them for surveys when the game launches.
Our Wiki became the outward-facing part of the process. When something was decided and implemented, we would add the new rules to a wiki page, and watch as the discussion page generated feedback or questions. It’s a very fasst way to catch oversights, but also to make sure that external communication is as clear as possible. Here, change history was beneficial for both readers and writers. We liked that fact that even though we didn’t want too many voices in the design process, we could still let the players contribute in a meaningful way.
Last but not least, email is not dead. Over 6 weeks of development, we sent more than 300 messages to our mailing list, which served many purposes. As an idea breeding ground, it is better because testing random ideas can derail a chat, but in email it’s cheap to do. As a repository for the results of a chat session, it is easily searchable, and discussion is easy. Over a wiki, it has the distinct advantage of being available when you’re offline or on the plane.
As the implementor of the team, I used tagging of messages a lot. I would use a specific tag telling me that the message contained something that was ready for implementation, and once I had some quiet time I would work them off and untag them when done. It’s a small thing, but it helped me keep track and meant that we avoided any kind of task tracking for the project. I have no idea what we would have used for that either, because I don’t know of any good trackers that work offline.
All in all, the situation was almost perfect. Sometime during development, Google released the plans for Wave, and I immediately realized how awesome that would be for us. If it works offline as well as the documents do, this could let us cut the email list out, and merge it with chat. I’m really looking forward to doing another projectlike this and trying out Wave, but even without it, I’m pleased with how easy it is to work on something so verbal as game design across continents and timezones.