Note: This post is one in a series of posts that seeks to examine ways in which information visualization (infovis) can be game-like, or gamey (an unfortunate, but fun, term we default to when discussing this topic as a group).
Moving on from my earlier discussion of narrative infovis, I will now engage what Bobby and I identified as free-form infovis.
Free-form infovis allows users to engage large data sets in relatively open terms. A nice game analogy might be something like the experience of playing Grand Theft Auto. Players are free to roam within a large gamescape and adhere to a set of play guidelines only if they choose to do so. Structure is available as something to adhere to, react against or simply ignore and these choices, or options, are inherent in the game's system.
Like the examples I discussed in the narrative infovis post, free-form infovis is another extensible means of engaging data sets, but the user experience is more open-ended. Free-form infovis is less prefab analysis and more investigation via interaction. Analysis is generally a by product of interaction.
Martin Wattenberg's NameVoyager – a visualization I've brought up in a previous post – provides an interesting case study for this discussion of the gamey properties of infovis. I say this because of the visualization's popularity and that it has been recognized as being game-like by its creator. The white paper is available from IBM.
NameVoyager is powered by a data set culled from lists of the 1,000 most popular names for boys and girls compiled by the Social Security Administration (SSA). The SSA has published these lists every decade from 1900 to present. The interactive visualization allows users to type in names and view the resulting trends of said name's popularity over time. It's an engaging piece that has been remarkably popular since it was released in 2005.
Through an analysis of user comments, Wattenberg concludes:
that usage patterns are strongly social and seem more closely related to those of online multiplayer games than to conventional single-user statistical tools. Indeed, users seem to fall neatly into Richard Bartle's well-known categorization of online game players as explorers, achievers, socializers or killers. This stands in contrast to the traditional view of information visualization as a task-oriented problem-solving activity.
Wattenberg hypothesizes that the broad popularity of the NameVoyager stems from features that give it a game-like sense of fun and make it suitable for social data analysis.
I'm mostly concerned with the "features that give it [NameVoyager] a game-like sense" coupled with the "data analysis" associated with information visualizations, be they journalistic or academic. For Wattenberg the level of a user's engagement is key to arriving at the "game-like" designation.
Relevance is one factor that determines a level of engagement with any artifact, be it a good news story, information visualization or game. With NameVoyager the point of entry, or relevance, for an individual user starts with their knowledge of names or their own name. Wattenberg posits that the "common ground" of name knowledge is what facilitates expression of individual perspectives. This "common ground" appeal could manifest itself in the content and subject matter of a game.
Additionally, in the same way that "killer graphics" draw audiences to games, smooth transitions, mouseovers, and multiple filters and views might make an infovis more appealing to a wider audience. While that might be a simplistic observation, eye candy goes a long way in attracting audiences. How many times do you have to look at a static bar graph to determine that it's boring? Graphic elements also facilitate spectator involvement, which Wattenberg believes is a component in generating social activity around NameVoyager.
I'm not going to leave it at eye candy... I think when considering the video game as an expressive medium, one of its best affordances is the concept of multiple plays. Unlike a blog post, news article, or a photo, users are more likely to return to a game and get more out of it the more they return. Free-form infovis shares this affordance with video games.
Take GoodGuide's Vote With Your Dollars for instance.
Granted the transitions and mouseovers aren't on the same level as the interactivity available in most contemporary game releases, but this means of engaging data is generally more appealing than the standard pie chart or bar graph mentioned above. As an aside, this visualization was built with Jeffrey Heer's Flare toolkit, an ActionScript library that can add nice transitions, multiple views and filters to almost any properly formatted data set. This is just one of a few emerging visualization tools that are making data more engaging for casual users.
But let's return to the concept of multiple plays. When Wattenberg mentions the level of user engagement and its relation to game-like qualities, I think he's on to something. The more a player plays a game, ideally, the better she becomes at the game. She can attain a better level of familiarity with the gamespace through multiple plays. I think the same can be said of free-form infovis. The more a user engages a particular visualization, the better his familiarity with the data set. Upon repeated engagements he might recognize certain correlations and comparative insights that weren't previously apparent.
I'm not going to take the these comparisons too far, but I do think there are more opportunities for data sets and games to intersect. I also think there are more lessons designers and statisticians can learn from game designers to create more engaging and informative visualizations.