The basic idea is that rapid prototyping with the parallel splatter model could lead to more thorough testing of design ideas. This is the difference between normal iterative designing and parallel prototype iteration. You want a map system, then have an iterative cycle of map systems on the main prototype. With at least a few to test and analyze, the team should have plenty to analyze and learn from.
While I take this in a game development direction, that's no limitation. Playable can be replaced with functional. This is a general idea for creating better results.
STEPS:
So, now that I've introduced the general idea, I'll get to the step processes and breakdowns.
Analyze:
This is a team activity including diagramming, discussions and determining the next design goals for the group. Not a whole lot to this, besides the facts that it is the only always team step and that it should be kept seperate from the design step. That's to make sure all the results are analyzed and understood by all the team members.
Design:
This step can be either a team, group or individual step, depending on what design goals are to be tackled. Either way, this step decides the ideas that are to be made into prototypes.
Test:
This is the heart of the concept, where the prototypes are made and altered. Yet, the explanation of why will come later.
UNIQUENESS:
Perhaps it doesn't sound sound too impressive right now, but the iterative process and prototyping ideas explain the effectiveness of those part already. No, it's not the process that is so interesting to me so much as the emergent patterns.
Main Prototype:
In rapid prototyping it makes sense to have a main set of build versions as the main prototype. These incorporate the selected code, features, interaction, gameplay and such from the parallel prototyping rounds. This means that each iteration you start with the culmination of the efforts so far, and don't need to redo the whole thing from scratch every time.
Branch Prototypes:
These are all the prototypes that are not a part of the main prototype version history. Some would say these are a waste of time, but these are not just "wasted effort", they are very useful as I tell more about later in this post.
Parallel Prototyping Round:
This is a round of iteration that focuses on designing and prototyping several ideas. That's to generate creative options with the data needed to make informed design decisions.
Updating Round:
Using the selected prototyped ideas from parallel prototyping rounds, the main prototype version is updated. This is much like refining the integration of selected idea(s) into the main prototype.
BENEFITS:
There are several benefits to this method theoretically. They are a combination of those normally found in iterative designing and parallel development.
Version Tree:
In most open source projects I know of there is a branching history of versions. The many branches give fall back options, code and feedback to be incorporated into the project as needed. Sometimes you will find that a major decision needs to be changed. The branches help speed up the process.
Information:
More so than either concept on it's own, this combined idea should give mounds of information. In fact, that's the reason for having the idea as a written up idea. As a formal design approach it should be able to handle the information that comes per prototype success, failure or in-between. Comparative play testing of versions and ideas, development feasibility and more come naturally out of this process.
Something to Show:
Iterative prototyping and rapid parallel prototyping both give results, but this combined method delivers playable prototypes of the ideas that didn't make it in. It encourages keeping each and every individual prototype created during the project. When people want to see what you have done, project progress, why you made a choice or anything else where a playable prototype would be helpful, you have them.
Have fun, spread the word and tell me what you think,
Igen Oukan
Read more!
View Blog Reactions
