You have /5 articles left.
Sign up for a free account or log in.
There's a rhetorical move we could call "overstatement by search engine metrics." You occasionally see it in print, though it seems a trope especially suited to television. A narrator or reporter will ask, for example, "Is Nancy Pelosi a space alien?" in a suitably incredulous tone as the question is typed into a search bar. "We thought it was a strange idea, but our search produced more than 800,000 hits!"
Which is, technically, true (the number of hits, that is) provided the search is done in a totally incompetent way. Placing the question in quotation marks -- so that the search engine looks only for that specific combination of words, rather than every site on which they appear together in whatever order -- will, in fact, yield no results from Google, at least until this column appears.
The example is deliberately preposterous, but it illustrates a couple of problems that seem common enough. The most obvious would be a naïve certainty that the number of search-engine results on a topic somehow indicates the level of public concern with it. Scrolling through a few screens' worth of results ought to disabuse anyone of that idea: besides hits that are obviously irrelevant, there are all the links to pages generated by robots that steal content from legitimate sites, for whatever shady purpose. But people seldom browse that far. Most stay on the first page of results.
The other issue is an excessive confidence in the ability of search-engine algorithms to recognize a natural-language inquiry, determine its meaning and generate appropriate results. That capacity is improving all the time, to be sure, though progress may just compound the problem. People already joke that it's slightly creepy how our devices seem to be learning to read our minds. The next step is in the direction of reliance on them being able to do so.
Enter, then, Daniel M. Russell -- foot hesitating in midair. His official job title at Google is senior research scientist for search quality and user happiness. His book, The Joy of Search: A Google Insider's Guide to Going Beyond the Basics (MIT Press), quite naturally falls under suspicion of being one more promotional vehicle for a company ever marching toward world domination on as many information-related fronts as possible. I offered to loan the review copy to someone who took one look at the cover and waved it away on the grounds that she is a committed user of Bing. For a company that took "Don't Be Evil" as its motto, Google has somehow inspired quite a few misgivings -- even among people who use it constantly.
No such ambivalence should apply to The Joy of Search, which is not the brand-focused user's guide a potential reader might understandably take it to be. Russell does cover some of the elements of Google's search-engine grammar, as with the use of quotations marks mentioned above. Likewise, a search can focus on specific websites, look for materials in a particular digital format or find webpages that refer to one author's name or another's but not both, and so on. Similar commands work with other search engines, with numerous videos on their use up on YouTube. These techniques ought to be common knowledge by now, though plenty of anecdotal evidence would suggest otherwise. (Plug the name of your preferred search engine into the search field and add "commands" or "operators," and the instructional materials should be quickly at hand.)
Russell's tutorials are on a much higher plane but turn on the same basic principle of learning to treat the search engine as a tool that can at best assist the brain to address questions rather than effortlessly delivering an answer to your screen. "We once had a dream," he writes, "that a single system would integrate all our information resources into a single master repository. The holy grail was that we would only need to search in one place -- in one master catalog, as the one place to find everything."
The first-person plural here might well refer to his colleagues at Google, but I think he is describing what became a more or less ubiquitous long-term outlook, dating back to digital antiquity, when AltaVista and Yahoo were mighty search-engine powers. In any case, it was not to be:
Instead of a perfectly flat and universally accessible universe of information resources, the world of information remains obstinately clumpy and segregated … This means not only does a literate person need to have a comprehensive understanding of the areas of information that [are] available but also an understanding of the different ways that information can be stored, indexed, accessed and read.
Missing from that formulation but pervasive throughout the book is that said "comprehensive understanding" is not something you acquire first and then put into use. It develops from experience, from searches conducted in breadth and depth, and with a certain amount of creativity. Russell places a surprising emphasis on the importance of intuition -- disciplined intuition. He recommends writing down your hunch or best guess before beginning to search for an answer: "If you don't write your intuitions down," he advises, "you'll quickly and unconsciously adjust them to align with the facts, once you figure out what they are. It's almost as though we can't stand to think of ourselves as making a mistake, so your mind will quickly back up and retroactively adjust your intuition to what you've learned."
I can't begin to describe the author's case studies in answering difficult questions about botany, architecture or history or determining the location of a building in a photograph taken in an unnamed country. Not everything he consults is online, but most of it is; the trick is to analyze the question, the evidence and the availability of relevant silos or pockets of applicable data. Faced with a question, you can always google it, but Russell makes explicit that you'll never learn anything if you leave it at that.
