It's Pervasive. It's Computing. It Needs Testing.

Evaluating early prototypes of any design requires a little bit of know-how. Luckily for us (insert sarcasm), as the paradigm shifts towards the more pervasive end, traditional desktop-centric tests no longer apply. Well, they do somewhat, but perform so badly that it's essential we find new tests and find 'em quick.

In comes a variant of experience prototyping; instead of simply letting users loose on a design, we address the group-think strategies most people tend to fall into (something that's always an issue with pervasive computing) by becoming immersed in the target population's current practices ourselves. And it's not just faking it - it's critically confronting a design's assumptions and finding misinterpretations.

However, one should proceed with caution. When researchers are also participants, they need to be careful not to let the interaction experience descend into artifice. Now what does that mean? It means bias is readily there and it's quite tempting to get carried away with your own judgments, clouding your actual experience to the point where it's truly elusive.

To top it off, we must address the fact that pretending a prototype is more accurate than it really is poses its own set of difficulties. Which is why we need a really, really good Wizard of Oz that can make it possible. Mix experience design and a sprinkle of magic together and "Bam!" You now have a suitable alternative to true contextual evaluation on very early design prototypes. In the end, you'll be gaining some realism at the expense of impartiality, but also the reverse; impartiality at the expense of realism.

Now, isn't informed design grand?