prototype vs. sim—validating software and ux design

[ SXSW Bios ] #sxsw #frogsxswi (Archive)


How much smoke and mirrors does it take to validate interaction models during the software design process? When do you have to stop faking it and start making it? How do you handle the traps of realistic demos slipping into production or permanent beta? Simulation, spike, proof-of-concept, interactive demo, prototype, and other artifacts often come with loose definitions and inflated expectations or lose their primary purpose during collaborative software design and realization. Design technology experts from frog who regularly push and pull on the boundaries of art and science will define bounds and discuss challenges, opportunities, risks, and rewards of going too far in real code during design or not going far enough. Topics will include defining needs and socializing intent for code-driven design assets across stakeholders, balancing speed and fidelity during interaction design, and understanding where early target platform development best informs and validates design.


Making is designing. Use technology to make ideas real as part of the design process (not only part of the build process). Watch the rise and fall of what people change/tinker with (DIY culture) to know where you are getting it right (things that are left alone) and where you are not there yet (things that are modded). Perceptual prototyping is maturing.

UX Simulation

Model it first.

Requirements sometimes fail. UX sim offers a rapid path forward (to see if it works) when requirements are difficult to capture, when research will be too costly.

Impact of the DIY movement on design practice: rapid propagation of ideas (people already know about your idea). If someone cares more about your idea than you do, they will take it and do something with it faster than you think (“ad-hocracies”, Doctorow); except they aren’t good at publishing standards which is opposite to the expectations of the consumer. New input methodologies.

Some Process

Use expedient technologies (flash/air, wpf, html5), utilize an agreed-upon visual and interactive fidelity and an index of the vision (make sure they know what it means, that it’s a simulation, and the blue sky factor/vision), create companion documents that re-enforce core concepts and principles (but without excessive wireframing). However, the simulation is the artifact.

Generative Whiteboarding: Everyone has to put an interaction model up on the board. Select the most viable option (not the problem-child or the favorite for favorite’s sake).


Make it tangible first.

Any idea—no matter how absurd—could, and should, be made real. What you learn from making it can both surprise and inform. Apple shares intimately detailed knowledge about their iphone/ipods to allow cases to be made. When this isn’t done, fragmentation occurs amongst technology and design alike.

Proof of Concepts

Make it workable first. “Technical spikes”.