Thinking about searching
I've taken to writing down my reactions to everything I read for class. It really helps to crystallize my thinking. I've sort of done it before, but making it a part of my research process is totally inspiring my brain and making me draw new conclusions and ask different questions. I should do this all the time, for everything.
Tonight, I'm thinking about two articles I had to read for my "Information, Organization, and Access" class. See bottom for links and summaries. Reaction to them:
Thinking about how people search for things. Thinking about how malleable our queries are, how malleable our brains are. Marcia Bates's berrypicking theory says our queries evolve as we sift through search results. I love that. I love that we react so quickly to new information, whether it is internal or external. It is exciting to think about search interfaces that are designed not just to deliver results, but to foster ideas and inspire innovation. Sometimes that can be as traditional as saying, Other people who have thought what you're thinking or looked at what you're looking at have also thought this or looked at that. But to think of even breaking free of what other people think. Either to throw a dart and hope it might land someplace useful, or to develop algorithms that use the same tracks for wheels of thought but constantly change destinations and starting points. To design a computer to ask, "Well, have you thought about this? Could this be useful? What if you crossed this data with that data?"
That isn't to say computers should tell us how to think — but they should be aids for our brains, providing additional stimuli that we can ignore or embrace. Because who knows? The processes for the best discoveries are sometimes first ignited by the chance meetings of information sources that seem unrelated. See: Rosetta Stone.
Articles read (somewhat dated):
- "The Design of Browsing and Berrypicking Techniques for the Online Search Interface," Marcia Bates, 1989. Summary: when people search for things, they refine or change their queries and conduct multiple searches until they find what they want. They collect information in bits and pieces rather than in one big payload. (See article for sad 1989 illustrative infographics.)
- "Historical note: Information retrieval and the future of an illusion," Don R. Swanson, 1988. Summary: the relevance of search results is problematic. How do you define relevance? It would be awesome, Swanson says towards the end, if relevance patterns could connect multiple documents from multiple fields that have unintended or unthought logical connections. "A pursuit that goes beyond information retrieval and may resemble the correlation and synthesis functions of intelligence analysis."