Egger, F.N. (2000). Lo-Fi vs. Hi-Fi Prototyping: how real does the real thing have to be? "Teaching HCI"  workshop, OZCHI2000, Sydney (Australia), December 14-19, 2000

Low-fidelity (lo-fi) prototyping is characterised by a quick and easy translation of high-level design concepts into tangible and testable artefacts. Lo-fi is also know as low-tech, as the means required for such an implementation consist, most of the time, of a mixture of paper, cardboard, post-it notes, acetone sheets etc. A clear advantage of lo-fi prototyping is its extremely low cost and the fact that non-programmers can actively be part of the idea-crystallisation process.

At the other extreme, high-fidelity (hi-fi) prototypes are characterised by a high-tech representation of the design concepts, resulting in partial to complete functionality. High-tech, however, implies higher costs, both temporal and financial, and necessitates good programming skills to implement the prototype. The main advantage of hi-fi, high-tech prototyping is that users can truly interact with the system, as opposed to the sometimes awkward facilitator-driven simulations found in lo-fi prototyping. Obviously, there is a continuum from low to high-fidelity prototyping that usually stretches out from early to late design.

Whatever the background of the students, it was observed that putting students in a condition where they do not have the time for high-tech implementations forces them to focus on the conceptual design stage and explore design alternatives before deciding which one to carry forward to the next stage. This is especially important for students with an IT or SE background, since spending more time on the conceptual design stage forces them to pay more attention to the psychological aspects of HCI design. In addition, experimenting with lo-fi prototypes prevents them from getting tunnel vision and going for a type of design they are familiar with and know how to implement, as opposed to the design that best meets the user requirements.

For example, a group of students was asked to present the conceptual design of a hand-held communication device for children aged 9 to 12. Given that they had only two days to come up with a first design, the requirements were defined in a brainstorming session within the group, as they did not have the time to go and talk to actual children. Despite the fact that the user needs and preferences were ill-defined, the group opted for a medium-fidelity prototype. They designed a desktop simulation of a palm-top device, which looked more like an adult's expensive gadget than a communication device for children. When asked about the functionality their system would support, they replied that it would run Windows CE and thus, that the children would be able to install whatever they wanted... The point was that these students rushed through the conceptual design stage only to be able to present a neat simulation.

Unleashing the students’ creativity by exploring alternatives can best be achieved by promoting throw-away prototyping. By presenting it as an exploration phase, students are more likely to focus on the iterative refinement of their ideas rather than on the details of their technical implementation. Although some students prefer to see their ideas turned into a visually pleasing computer-based system straight away, one should bear in mind that the refinement of their medium-fidelity prototype might take precedence over testing the relevance of the system for its target users. If the educational objective is to teach students that HCI design consists of several well-defined stages, then care should be taken that the students do not neglect or skip parts of the process just because they want to demonstrate their programming skills.

Along the same line of reasoning, one should be careful not to encourage the use of whatever sophisticated software package or piece of expensive technology one’s institute happens to possess. That would qualify as technology push, an idea that does not really fit the philosophy of HCI. Indeed, students might use a high-tech infrastructure only because it is available, even though the resulting system might not respond to any actual needs people might have (thus, the system would be high-tech but not useful...).

The advantages of lo-fi or low-tech are numerous. In addition to being cheap, fast and accessible to non-programmers, these rudimentary techniques can yield a maximum of feedback on design ideas at a minimal cost. Insisting that students should consider adopting low-tech throw-away prototyping instead of going straight away for a computer-based evolutionary prototype might help them explore alternatives and, above all, focus more on HCI aspects of the target system.

Florian N. Egger has a first degree in Psychology & Philosophy, an MSc in Ergonomics/HCI, and is currently doing his PhD in HCI. His doctoral research deals with human factors in business-to-consumer e-commerce systems, such as trust-inducing features and factors contributing to a positive user experience. He has been involved in the active coaching of students enrolled in a Master course in HCI. These students have a one-year HCI design project in which they have to go through all phases of design to end up with a fully-functional system.