A small group of friends (including myself) decided to develop a simple 2D game. We are all fairly new to software development, so I wanted to ask which software development model would likely work best for such a group. From the list on Wikipedia, the Waterfall and Spiral models seem to be the two best options.

Our plan is to use C# with XNA for development.

  • 1
    If it's small and impersonal go Agile. You won't need all the ceremonies but will probably end up working Agile-ish after a while no matter what you choose
    – James
    Commented Jan 31, 2013 at 4:06
  • This is a list answer question and as such is generally considered unacceptable for our Q&A format. Commented Jan 31, 2013 at 5:42

1 Answer 1


If you're new to software development, then the different software development models probably won't make much sense. Although I had studied (in school) and later read books about software development methodologies in the past, it probably took me about two decades (and ending up in a suitable industry) to truly understand what the so-called "Waterfall" process really meant and what information gets transferred between each step.

As a novice programmer, you are going to have to dedicate so much effort to actually developing software, that most of the "process" that you read about will seem like frivolous annoying bookkeeping. Believe me, you're not going to want to worry about cross-referencing the sections in the Software Design Document with the requirement numbers in the Software Requirements Document, when you've just got jumping working but there's a bug when you try to fire at the same time.

  • Definitely true. One particular thing I do need to figure out, however, is how much planning & design to go through before we begin the coding process. Would it be better to try to make a fairly complete map of the objects & methods & such, or just go over a vision of what we want and the basic necessities of how to start working towards it? Commented Jan 31, 2013 at 4:50
  • I would start with the simplest thing that can possibly work. Get a framework doing something useful, and then build on that. If you spend a lot of time designing your object model and methods and whatnot before getting anything actually happening, you're going to lose motivation and momentum fast. Commented Jan 31, 2013 at 4:56

Not the answer you're looking for? Browse other questions tagged or ask your own question.