Pete Coco’s post on discovery systems and “Googley” search mechanisms is a great one; I have been interested in this topic lately, because my students at Birmingham-Southern college were introduced to EBSCO’s Discovery Service this year.
Coco points out that users like the interface–of course they do! We casual researchers love the one-search-box, googley feel to discovery systems. Sometimes when you are faced with a long list of database options, choosing just one can be overwhelming! Average Joe Researcher likes slamming his query down into one search box–its so simple! (Coco uses the pizza-craving student as a metaphor here).
However, the average researcher also doesn’t carefully analyze his search results. He simply takes the first few hits and assumes they are the best. Coco questions if this really equates to a “successful” search. Granted, students are finding resources, but the usefulness of their resources is dubious, and they are likely not finding the best resources through this system. At least not without digging a little deeper. His searches are faster, but they do not yield better results.
On the other hand, as Coco notes, at least students are using the system! Maybe they aren’t really embracing dicsovery systems efficiently, finding all of the best information on the first try. However, usage numbers are up.
The question remains, does the upward spike in resource usage (and downward spike in straight-up google–based research) balance the possible drawbacks in discovery systems?
A theoretical way to eliminate this question is for metadata associated with these systems to be so perfect–so mind-readingly perfect– that users get the perfect match for their sloppy search strings on the first try! It is a great idea, but not a practical one.