Understanding exactly what information is needed when confronted with a reference question or research request remains a challenge to information professionals regardless of technology-driven changes in the search process. It’s still true that misunderstanding the intent of a research request will send us off in the wrong direction and frustrate our clients. Terminology is important, and words often have multiple meanings.
Take taxonomy, for example. We think of it in terms of a body of literature and documents. Taxonomies could be applied to bibliographic databases or to an enterprise’s collection of internal information. To information professionals, it’s synonymous with thesauri, controlled vocabulary, and index terms. But search the web for the term, and results will pop up first from biology and its classification of organisms (plants and animals).
What do we mean when we say something? How easy is it for someone to misinterpret what we’re saying? In a restaurant recently, a person at a table near me replied to their waiter, “We’re OK,” when asked if they were ready to order. The waiter walked away only to be summoned back. Apparently, to the person uttering the phrase, “We’re OK” meant the diners were actually ready to order their meal and was not an indication that the waiter should come back later.
Here is another misleading statement. “I read an article that said … ” Did the person actually read the original article? Or was it a newspaper report about the original article? Or a mention in social media directing people to the article that included a brief summary? Even if the person had the article in hand, did “reading” it consist of scanning the first and last paragraphs? Chalk the latter up to the TL;DR (too long; didn’t read) phenomenon.
When searching traditional library databases, information professionals rely on controlled vocabulary and complex search strategies to force the databases to cough up the best results. We are masters at Boolean search strings, adjacency operators, and limiting results. Web search engines build in the complexity on the back end. Innovation in search stems from intuiting searcher intent without an extensive reference interview. Machine learning, as a component of web search, enhances the relevance of results, as do its personalization aspects.
How does this apply to the Internet of Things? When objects “talk” to us, might we misunderstand the data? A heat map in the library tells us that one particular section is visited often. But during what time period? Could a Twitter post that money was hidden there be responsible? A chart shows a spike in interest in a political topic—is this a trend or a reaction to a one-time, offhand remark?
Before we use information gleaned from the Internet of Things for decision making, let’s consider language and the possibilities for misunderstandings. In an always-connected world, don’t discount the value of human intelligence to understand the nuances inherent in research data and terminology.