As I was showing del.icio.us to my students last night, I realized that it's an amazingly useful tool for information architects. I regularly tell my students that asking people is a generally bad way to find out what they want, or how they want it organized (and point them to Cory's Metacrap article for examples). But if you're trying to figure out what set of labels to use for a set of domain-specific content, and you don't have a multi-thousand dollar budget for studies and consultants, how do you create a usable, appropriate vocabulary?
Here's how. Add a site to your del.icio.us bookmarks, and then look to see who else has added it. What descriptive tags did they use for it? As an example, here are the current links to Metacrap in the del.icio.us system. I used the terms metadata and semweb. Other terms used include taxonomy, ontology, ia, humanFactor, and xml. That's a great start for thinking about how to make it part of a collection, and how to organize/label that collection.
It's also interesting to watch how people's tag collections grow and change. VirtualTraveler has started using a pseudo-hierarchical tagging system, by including a / character in some tags (e.g. ComputerHistory/Books, ComputerHistory/DesignReports).
Would the system benefit at all from a collaborative thesaurus, I wonder? As an optional rather than require tool? How hard would it be to implement that?
What a great sandbox...
And as rapidly learning and collaborating ants you can admire how fast this growing web of contributors learns and evolves without any top-down coordination.
Through delicious you can actually see patterns evolve over time as information miners learn rapidly how to select, reference, categorize and post information resources of their own interest.