During the 21st, 22nd and 23rd of September, I was in Florida, together with some other fellows GNOME contributors, for the User Observation Hackfest held within this year’s openSUSE Summit. As some other people already reported, the central event of the Hackfest was a visit to the City of Largo, home of one large GNOME deployment within its public sector. Dave (Richards) presented us his very impressive SUSE-powered thin clients and also some of the rationale behind his well-known UI customizations to GNOME.
One point I found particularly interesting was that most customizations were designed as responses to observed behavior of users (via sysadmin logs) or/and to collected user feedback (via tech support requests), and that, in fact, users indeed perceive them as solutions to their real day-to-day problems. This is an example that brings us to why I think events like the User Observation Hackfest are crucial to GNOME in its mission: Design is both problem and solution, and problem is out there. I strongly believe good interface design starts with understanding of people: who they are, why they use our software and how they communicate with it. The more we know about our users, the more effectively we can design for them.
User observation is, among other tools, one of the ways we can perform research to gain understanding about the users’ world. During the hackfest, we had the opportunity to talk to users that use GNOME in a day-to-day basis and hear about their goals, their tasks, their attitudes and about how they communicate with our software. Of course, the guys at the City of Largo share, despite some quirks, a set of common behavior patterns and, therefore, constitute one specific user group. More User Observation Hackfests would be great to observe other user groups so that, maybe, in the future we can develop a set of personas that we can use as a tool for our designs.
As those personas document trends in user interaction, they would also help the construction of our pattern language and of our HIG. During the day that followed our visit to Largo, we did some work on extracting behavioral trends from the evidences we observed. From now on, we should keep working on those trends (and, if possible, collecting more) in order to turn them into patterns or guidelines for GNOME design. I think it’s also important to observe that, as from now, the work we’ve done is not prescriptive, focusing more in high-level patterns than in interface elements. With this in mind, I think they can be useful not only for the Design team, but also for the entire project, meaning documentation, feature planning, marketing, architecture, translations and everything else.
Attending this hackfest was truly great and I immensely thank the event sponsors (GNOME Foundation, SUSE and City of Largo) for making it possible for me to attend it.