Sketching, Personas and the Nintendo Power Glove

At my day job, our UX team is, well, fledgling. One reason I wanted to learn about UX at this year’s SXSW is to learn more about bringing a user-focused culture to my team. Today’s first session, Bootcamp for a UX Team of None was ideal for beginning to answer these questions and the panelists gave helpful ideas on everything from sketching and prototyping to testing and validation. One of these ideas is the 6-8-5 method where you get down 6-8 ideas in 5 minutes. They handed out sheets of paper and had everyone practice. I was surprised at how fast that time goes. You have to be quick—and that’s the point. These are not meant to be any kind of finished work. This is only to get down lots of ideas in a short amount of time. After a round of this, we collect feedback from colleagues. Then, do it again. After another round of feedback, put each idea onto its own page and you’re ready to go out and test with people in the street.

I can’t wait to try this, but what about sketching in the first place? How do you communicate through a sketch? Well, the next panel, Shut Up and Draw, gave some strategies for that. Dan Roam (Back of the Napkin and Blah, Blah, Blah), Jessica Hagy, and Sunni Brown each discussed how to start sketching quickly. They described how when you’re trying to get someone to understand what you’re saying, a sketch is most often the best way. People understand images far better than more words. In other words, shut up and draw.

Personas are always a hot topic, and I want to learn how to do them better. The speakers for Avoiding BS Personas worked through the process using Bolt Peters’ work for as an example. One item that needs to be cut from personas is fluff. People will not take them seriously and it can lead the development team down the wrong path if you use them. The fluff increases the signal to noise ratio and as a result, the persona is useless. The other end of the spectrum is a stock photo with some demographic info attached. This often is too little info and does not give a full picture of who the people really are.

Instead, Jill Christ and Stephanie Carter say you need to go out into the field and test real people where they actually use the website. Taking photos and video of their environment as well as taking care to note items they use and how they use those things in their space. Then synthesize the data and look for patterns. This is not a quick task, in the study, they talked to upwards of 2,000 people (although I don’t believe they visited all those). In the end, had a very clear idea of who their audience is and are able to use that information to create products that solve their needs.

Finally, RJ Owen spoke about UI design at Don’t Build a Power Glove. Growing up, he loved the power glove, although he never played games with it. One day he read how the power glove was an example of horrible design and set out to find why. In a week and a half and on his own time, he did his own study to replicate the conditions before the Power Glove came out in 1989. He talked to families and watched them play video games. He asked them questions about the controllers and what types of controllers might be best for which types of games. Basically instead of starting with technology first (like Nintendo did), he started with looking to the user and figuring out what problem they had and set out to solve that. Nintendo didn’t have any idea what an actual user’s environment was or how they actually played the game. If they had, they may have never came out with the Power Glove. The takeaway here is that before you set out to make the next great app or website or product, talk to your users, do a little bit of research (maybe even go more in-depth as in the example) and solve the User’s problem.

Tomorrow is the last day, I really can’t believe it has gone so fast. Jeffery Zeldman will be inducted into the webbie hall of fame, so that should be interesting.

Categories: Uncategorized

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.