Notes from the Technology Trenches - Are You Practicing What We Preach about Searching?By Cindy Curling, Published on November 29, 2003
Cindy Carlson is the Electronic Resources Librarian at Fried Frank Harris Shriver & Jacobson in Washington, D.C., a web committee member for the Law Librarian’s Society of Washington, D.C., and organizer of its Legal Research Training Focus Group.
Lately I have searching on the brain. I've been reading many, many articles about how people should search. I've seen tons of tips on using advanced features of Internet search engines (Google, Altavista, AllTheWeb), both those that are visible and those that take a little insight or hunting in the help pages to use. Really, there have been so many lately that I was becoming a little jaded about them. And then I read a short article from TVC Alert about a "trends" discussion at the recent Internet Librarian conference. One of the speakers, Stephen Abram of Micromedia ProQuest said that he thought librarians were spending too much time focusing on teaching people how to search, as opposed to teaching them how to choose the right resource, and how to evaluate the information they find. Now, granted, I wasn't there and may be misinterpreting Mr. Abram's intent based on a summary, but that's how it reads to me.
Boolean search logic, he claims, is a waste of time to teach. The theory is that in future syntax will be transparent and your computer will understand what you mean from your context, so it will be able to create a query that will get you the information you need without you having to do much if any of the query construction.
Boolean Outdated? No WAY!
What is your immediate reaction to that? No way, can't be, gotta know Boolean? My question is, how long will that be? I don't doubt that every effort is being made to make Boolean an invisible force for search logic, but I don't see it coming along so quickly that we can just quit worrying about syntax now. Well, mostly.
Sure, you may not need to worry overmuch about the syntax of searches that are shooting for a particular, specific result. Take the major Web search engines noted above, for instance. It's truly an uphill battle to get users to pay attention to an advanced search when for most of their queries they get great, relevant results. They get those results, though, because they tend to be searching for a specific page belonging to some entity like an association, company, school or government agency. I doubt that even librarians use the "always search using more than one search engine" rule in a situation of that type. Once you've got your answer, you've got your answer.
The tricky bit comes when users are searching for something topical and it's either very broad or very specific. For those queries, a simple word search will often return results that include their search terms, and they may even be relevant. The trouble is that there are often simply too many to sift through, or that there are many that are duplicative or that mention the right words but don't give further information. Google has certainly brought us a long way toward combating those problems with its popularity ranking, but what happens when all those considerations are equal? Then users are well rewarded by using advance search features. (For an eye-opening comparison of search engine features, see the chart at Search Engine Showdown.)
And how about Boolean driven fee-based services like Lexis and Westlaw? Frankly, I would hope that after years of free use in law school that even a very new attorney would be familiar with the conventions of Boolean search syntax, so I don't worry too much about them. But what about legal assistants? What about the secretary whose attorney says, "Get me this. It's on Lexis. Here's my password?" Does your current contract really make you feel comfortable with that scenario? And while using the natural language queries on those services is sometimes extremely valuable, it is not always the best option. Sometimes you really do need to see all the records in which a particular term appears, not just the "best 100." Besides, just finding the option to search in plain English can sometimes be a challenge. And what about the Web-based subscription loose-leaf services? Users REALLY need to have a decent idea about how those tools work and of the correct strategies to use. If not, they may not even know that a shared passwords defaults were reset by the last user and think they are searching for this or that when they are actually searching this w/25 that. I have a headache now just thinking about it.
On the Other Hand
I do think Abrams has an excellent point about source selection and analysis. Another problem with Web searches is that even when you get relevant results, you often still miss reams of data that isn't accessible in a broad search but could be out there if you searched directly in the right database. People do need to be more aware that it is sometimes inefficient to search by keyword across all available Web pages. A keyword search for a broader topic AND a resource type (annual plants and database) or (chemistry and biographies or people) may get you to a resource that is infinitely more helpful then a specific topic search alone. Or they may need to use a hierarchical list like Yahoo's. Once you are at a well designed, rich resource (think Amazon or the Internet Movie Database) you don't need to worry as much about search syntax because you're already dealing with a relevant, discreet pool of possible results, and if you go through a service like Yahoo where things are already categorized, you may still be dealing with loads of pages, but at least they tend to be genuinely relevant.
And are the relevant results you get on the Web reliable? No one over-arching entity monitors what gets put up on the Web. Even old reliable print sources like the New York Times have dealt with embarrassing issues of reliability lately, so don't believe it unless you can verify it or at least be fairly sure that the person providing the information is trustworthy. For more on evaluating Web sources, see Getting It Right: Verifying Sources on the Net and Evaluating the Quality of Information on the Internet.
Frankly, I don't dwell on Boolean searching most of the time as I do resource training. What I spend more time teaching our new associates these days is to think more about how the documents they need are structured and how they can leverage that to their advantage. They understand that search relevancy sometimes increases based on where you search within a document. Some simple examples: searching the title and first few paragraphs of news, law reviews and sources like the ALR or searching an online legal periodicals index rather than all law reviews in full text.
We also concentrate some on getting them beyond their default habit of searching online. Easier said than done, but they do at least get the idea that it may help them with their online work if they get a little background in print or through a subscription service that we don't charge back to our clients. And speaking of source selection, why is it that no one wants to call anyone anymore? Here in Washington there a thousands of government offices, libraries, associations and non-governmental agencies filled with experts who would probably be thrilled to their toes if someone actually called and asked them for information on their favorite topics. While you are relying on the Web, don't forget that it is a great source for contact information too.
It's Feedback Time
So, many of you out there are librarians and are well aware of the current wisdom on searching. Do you practice what you preach? How much of your searching day is spent exclusively on Google? How often do you really use other search engines? How about your other resources? Do you use your own catalog? Intranet? (Do you even have an intranet? I was at a meeting this week espousing the advantages of making training materials available on the office intranet and out of a group of 20 there were still a few people who didn't have the option.) Do you have some kind of knowledge management software? Do you actually use it? Do you use your print resources? Direct online subscriptions? Do you feel that Boolean logic is still important to teach? Why?
And how about you attorneys? If you're reading this, you know what we librarians think is important. Are you a good searcher who practices what the librarians preach? If not, why not?
Believe it or not, I really want to know; I'm just exactly that kind of geek. So, please email me with your thoughts on searching issues and tell me how it really is with you.