We finally did it - the game is online and contracts are signed. Aside from the obvious of actually being able to play, this has a few added benefits. First, there is very little left that I am not legally allowed to tell you, meaning I can disclose more information than I did before. I hate secrets, so I see this personally as a good thing.
Second, Apocoplay (my company) now has much more control over how we manage things. Not total control, but enough freedom to do the high level things we want. This means that small rapid iterations aren't something we have to fight for anymore, they just are part of how things work now. This will be our first real iteration.
The rough notes our goals for the next two weeks are as follows. They will be solidified by the end of the day:
* Make it easier to report bugs. I don't care if this is a link to a forums post on the My Iczer page or something else that is simple, but we need a more intuitive way for users to report bugs. We can make it more elegant later.
* Eliminate all bugs marked as Urgent or higher.
* Come to a decision on the reward system for finding bugs. It should be as generous as possible without detracting from the worth of other prizes.
* Users periodically get gran based on the amount of wins vs amount of losses they have. So long as the user gets at least 10 gran, every loss detracts from the total number for gran granted for that cycle. This is unacceptable. Users should not be penalized for playing the game. If a user plays well and loses, they should still be rewarded, just not as much as the winner.
* Move to a web based Kanban system. Users should be able to easily visually track our goals and progress.
* Improve our infrastructure based upon observations of the launch.
* Before we were in agile we accrued some technical debt on our test network to get things done on deadline. I don't like technical debt, so we are closing that now and upgrading our test network.
* More fire drills. While our disaster recovery, scalability, and the like all are proven I'm not happy with their response times. Maybe this is just my previous habits from designing medical or enterprise level government software talking, but should the worst case scenario perfect storm hit I don't want to wait hours for a full recovery, I want minutes or seconds. The second the test network is upgraded I want us to run more fire drills on the test network and cut down disaster recovery times.
* Fix all bugs with normal priority.
* Make a prototype of a new card list page with a better filter and design. The goal here is to prove a basic understanding of the old database infrastructure as much as it is making a new card list.
* The game should have the ability to drag and drop cards. Create prototype pages showcasing various drag and drop options for users to test. As a note, all new code is modular so if we find something better in the future we can always swap out.
* Reach out to contacts at MIT Media labs to improve UI design. (My friends at Google spoke highly of them)
* Continue improving our networking with other gaming companies (Turbine games, EA refugees, etc.. even Zynga). The network has proven a valuable resource in the past.
Obviously the ones that allow us to get ready for a full launch have the highest priority.
As a reminder, this is our first real sprint and we have not determined our velocity yet. It will take at least three or four full sprints to get a more accurate expectation of how much new development can be done and in how much time. Also note that as we spend less time fixing the older more complex architecture and more time creating new and improve code (that doesn't touch the old architecture) the more will get done per hour per person.