A series on prototyping
In the 4 years since our book on prototyping first came on the scene there was precious little written about the professional way to prototype. Today prototyping seems to be the hot topic, unfortunately most of the current stuff available on the internet only give an isolated tip or trick. What is especially harmful is that most of these articles rush into how to prototype without really understanding what it is. These works are rife with unquestioned assumptions and and uncritical approach to prototyping.
The book I co-wrote with Michael Arent and Nevin Berger remains still the only thorough attempt to understand prototyping. In the coming series of posts on prototyping, I want to make a compact discussion of what a prototype is and how it works from our book Effective Prototyping for Software Makers . This post will it a little more approachable and if you want the full details, by all means you can buy the book
In this series of posts I want to address 3 broad topics:
- What does a prototype do
- What is a prototype
- Raising the bar in prototyping
In this first post I want to discuss what a prototype does. For that I want to use a definition of prototyping that restricts itself to what it does not what it is. For that I turn to a definition from the book “Universal Design Principles” by William Lidwell and others:
A prototype is “The use of simplified and incomplete models of a design to explore ideas, elaborate requirements, refine specifications, and test functionality.”
For ease of discussion, I will break this definition down into its components. First, I will throw out the models business because that goes into what a prototype is, which will be the subject of the next post. That leaves us with the following uses of a prototype:
- to explore ideas
- to elaborate requirements
- to refine specifications
- to test functionality
All prototypes attempt to do at least one of the above purposes and usually more than one simultaneously, often without the prototyper even being aware of it. What is essential to know is that the prototype is first and foremost a communication medium. A prototype communicates the above 4 concepts.
Here the accent is on if the idea is desirable. Prototypes are at their best when they explore abstract concepts or ideas and makes them concrete. It is easy enough for a group of technocrats to discuss their new idea for a killer document registration, yet being able to both rapidly and interactively visualize with a prototype makes the idea come alive and often inspires and informs the whole ideation process. Any software idea can be visualized with a prototype. But here are just a few examples (a fuller list comes in a future post discussing prototyping content):
- Interactions design
- Application functionality
- Visual design
- Information design/architecture
- Rough concepts and ideas
Among the many means of using prototyping to explore are:
- a single prototyper visualizes the idea
- a group prototypes through participatory design practices
- members of a group each sketch out their ideas as a group
- a group brainstorms a prototype with a designer as facilitator
Here the accent is on, whether the prototype is possible. A prototype elaborates requirements by often illustrating what is necessary to actually put an idea into action. For example, and idea of having a running total in web interface when illustrated will make a developer realizes they need Web 2.0 technology. Or it could make the business analyst realize that discounts or other items that effect the total also need to be known upfront or somehow communicated to the end user. Once an idea is explored, software makers often look at a prototype differently. Among the types of requirements that are elaborated by a prototype include:
- End user
Here the accent is on, whether the prototype is feasible and if so how. One the idea is desirable and deemed possible, then the detailed design comes in. A prototype is often a superior form of specification than a large paper document with lots of verbiage where the requirements are difficult to ascertain let alone visualize. Furthermore, the prototype speaks in the visual language of the product itself and cross cultural and language concerns are not so big an issue with today’s global development teams.
A prototype can be stand alone documentation if it is a totally complete model. Otherwise, often some form of annotation or some lightweight document is needed to accompany it.
Here the accent is on evaluattion, for example whether the prototyped design is usable for the end user.
A working prototype (paper, digital or whatever form) can be shown to stakeholders and they can test it to see if they can work with it. If it works the way it should or they way it needs to. This way corrections to the design will cost no redevelopment costs.
So in a nutshell, this covers what a prototype does. In essence it communicates four things:
- to explore ideas — is it desirable?
- to elaborate requirements — is it possible?
- to refine specifications — how do we do it?
- to test functionality — does it work?
How and what it communicates will be discussed in my next post on what a prototype is. In that post I will discuss the characteristic parts of a prototype. Understanding these characteristic parts allows you to control how your prototype will come across to your audience.