A few years after hearing him speak in Toronto, I finally got around to reading David Weinberger’s Everything is Miscellaneous. The book is partly a history of classification and library science, and partly an investigation into how the web enables new approaches to organizing information. I enjoyed his writing very much, especially his accessible summaries of research. Today’s blog post focuses on how human perceptions of categories can shape how we develop and deploy classifications at the City of Toronto.
One of the more interesting insights Weinberger uncovered was that abstract categories are often artificial and difficult for people to use. They do not reflect the loose or messy ways in which people typically categorize or associate similar things. On the other hand, “concepts can be clear without having clear definitions if they’re organized around undisputed examples, or prototypes” (pg 185). Examples like chair or car have characteristics that are quickly identified: chairs are for sitting on; cars have a steering wheel, an engine and 4 wheels, etc. These intuitive characteristics make it easier for people to group together similar things. More general concepts like furniture or vehicle are not so easy to pin down.
Toronto’s 311 online service requests show how we can use easily identified prototypes or examples to make information easier to find:
One of the more interesting insights Weinberger uncovered was that abstract categories are often artificial and difficult for people to use. They do not reflect the loose or messy ways in which people typically categorize or associate similar things. On the other hand, “concepts can be clear without having clear definitions if they’re organized around undisputed examples, or prototypes” (pg 185). Examples like chair or car have characteristics that are quickly identified: chairs are for sitting on; cars have a steering wheel, an engine and 4 wheels, etc. These intuitive characteristics make it easier for people to group together similar things. More general concepts like furniture or vehicle are not so easy to pin down.
Toronto’s 311 online service requests show how we can use easily identified prototypes or examples to make information easier to find:
In this screen capture, the service request for bulky items – missed collection is accompanied by an image of examples (a couch, a refrigerator, a lamp) as well text stating what is covered (furniture, metal items/appliances). The old term “white goods” is not seen anywhere. We might be able to make the request even easier by listing other examples like stoves, washing machines, dryers, etc.
Now let’s look at New York City’s 311 online service categories:
Instead of a Transportation category, we see the more explicit Transportation, Streets and Sidewalks. The categories are made even more obvious by listing examples (motor vehicles, traffic, parking and towing, buses, subways….). The public doesn’t have to guess which category to click on for subway information, because “subway” is right there.
When we organize information, we can’t lose touch with this central goal: help users find things. As Weinberger quotes psychologist Eleanor Rosch, “the task of categorization systems is to provide the maximum information with the least cognitive effort” (pg 186).
No comments:
Post a Comment