Having passed a few milestones on my infolit journey, it occurs to me that I’ve accumulated some ideas about the nature of good search. Maybe it’s time to share seven of them with you.
Rather than thinking of myself as someone who’s been around the block a few times, I prefer the designation “seasoned searcher.” Not that I am a pro yet (that takes two lifetimes), but after you’ve been around search for awhile, certain best practice patterns emerge. Here are mine:
PROBLEMS NOT TOPICS
A lot of searchers start their research with a topic, as in, “I need to find out about the functions of the Mars Rover.” Google doesn’t help overcome this misperception of genuine research. Google was built as an answer machine to tell you what you need to know about topics. So you look up everything you can about the Mars Rover and you … What? Write it up? The authors of the Wikipedia article have already done that, and they did a great job of it (see “ Curiosity (rover)”). This exemplifies my first tip: Search the problem, not the topic.
Instead of replicating Wikipedia, you should let your mind move a few beats and consider a problem that can take you beyond mere facts. In an episode of The Big Bang Theory, Howard, who had charge of the Mars Rover, crashed it in a vain at tempt to impress his girlfriend with his Rover driving skills. New possibilities emerge. Not that anyone at NASA would have given Howard the controls, but there must be a means for the Rover to right itself when bad things happen.
Now you have the kernel of a problem, beyond a topic, which you can encapsulate in the question, “To what extent is it possible for the Mars Rover Curiosity to have a fail-safe mechanism against possible accidents such as getting stuck or overturning?” The “to what extent” part makes it evaluative, not simply a quest for information. You want to look at how good the fail-safe option is, including a variety of views on this, if they exist. Searching directly on the problem will focus your inquiry and eliminate all the basic stuff. Use search queries such as curiosity rover hazard avoidance , curiosity rover obstacles , curiosity rover stability , and so on. These are goal-oriented searches rather than those that simply retrieve some facts.
ELEPHANT IN THE ROOM
The adage, “When doing a sculpture of an elephant, chip away anything that doesn’t look like elephant” might not always produce good sculptures, but it makes for great searching. You can, I suppose, think of search as a means to gather all you can know. But these days, with the vast abundance of available resources, you’re advised, as a smart searcher, to get rid of whole reams of stuff you don’t want until you start seeing an elephant.
Good search is like sculpting. You are seeking a shape in side the mass of search results, a set of patterns comprising results that actually deal with the problem at hand. Paring down to fewer, but better, results beats a ton of stuff on the topic but not on the problem.
You want none of the “Take the first five citations and go with them,” which belongs to the false notion of searching topics rather than searching problems. Once you are searching for a solution to a problem, your results have to be sorted into stuff that meets your criteria. If there are limiters that can speed up the chipping away process, use them. But be careful: Sometimes a limiter will remove resources that actually are relevant, leaving you wondering why your search result elephant is missing a left ear.
I hear this constantly: “Why do I need so many books/articles in my bibliography? They all say the same thing.” My simple answer is, “If you are searching for the problem rather than just the topic, they don’t all say the same thing. No, they don’t.” If you are in the territory of seeking resolution of a research problem, you will find that the voices speaking to it rarely speak as one; they’re not identical. The key for the searcher is to find the patterns of discussion in the results, determine what the various positions are, and identify the key proponents of each position.
This isn’t easy, but good search isn’t easy. (Thank Google for lying to you that search is a snap.) Here are some helpful tips. A first clue might be a source that points out researcher A found this, but researcher B found that; or researcher A explains it this way, and researcher B that way. It might be the problem has undergone a transition in approach so that the old view was A but the modified view (maybe not held by all) is B. Literature reviews in recent articles often provide a gold mine of conversation identification guidance. Assume that the sources don’t agree, and seek out the nuances.
When you see names of position proponents, track down their actual studies through reference lists and bibliographies. Locate and read those sources, not just the first and last paragraphs. Group them by their position or approach so that you have the categorization of voices ready for your own ongoing analysis.
If all your results do say the same thing, there is a fundamental flaw in your process. If everyone agrees, you may not be addressing a real problem. You may only be searching a topic. Consensus, much as we might desire it, prevents us from advancing our research.
A true problem requires a lot of work to get to a solution. A true problem is not something whose answer can be found by looking it up. It can’t be that simple. Here you need to make the distinction between seeking information as a goal—you look it up and find an answer—and using information as a tool to solve something you can’t look up. If you are asking questions such as, “What happened? What does the data tell us?” your sources are going to solve everything without any need for your own digging and analysis. The answers you find will have a basic sameness. Problems are not supposed to be that easy.
So, if everything says the same thing, you need to tweak your goal to raise it above fact-finding and into the realm of seeking resolution.