End User's Corner - June 1996

Jack SolockSite-ation Pearl Growing

Jack Solock, Special Librarian

June 1996

Consider this scenario. You are a tenured economics professor with twenty years of experience. You are well versed in both teaching and research, and you know your sources cold. Your students, however, have started asking you why your class isn't on the net. You know that if you put your class on the net, you will not only want to put lecture notes, assignments, and test examples up; you will also want to provide links to excellent Internet economic resources.

Until now, you've avoided the Internet like the plague. How do you find out what the best sources of information in your field are? Intrepidly you fire up your browser and type in the address of Alta Vista (http://altavista.digital.com). You've heard this is one of the biggest search engines on the Internet, so you even study the FAQ a little and do an advanced query on "title:economics." About 10,000 matches come back. Hmmm. You then type in the address of Yahoo (http://www.yahoo.com/), a large Internet subject index you've heard of. You click on "Economics" and find that in addition to eleven meta-indexes, there are over 300 resources listed. Where do you start? Which ones are valuable?

Then one of your colleagues tells you about the twelfth edition of "Resources for Economists on the Internet," by Professor Bill Goffe of Southern Mississippi University. It turns out that it was at Yahoo (and at Alta Vista too, for that matter), but you missed it. Your colleague told you it was at the Clearinghouse for Subject Oriented Internet Resource Guides (http://www.lib.umich.edu/chhome.html). There you find that "Resources for Economists" can also be found at a gopher site (gopher://una.hh.lib.umich.edu:70/11/inetdirsstacks), which you feverishly link to. [Note: When last checked by the Internet Scout team, this gopher was no longer available.] You download and print it. You're on your way.

This tale is an example of the problems people face when they try to find valuable Internet resources in their field. Last month we talked about "drinking water from a fire hose." The Internet contains so many resources that a major problem with using it is finding quality resources. The title of this article refers to an information concept called citation pearl growing, which is a strategy that librarians use when querying large information databases like Lexis/Nexis or Dialog. It can also be used profitably for Internet searches. In order to understand it, two other information concepts, false drops and precision, must be understood first.

When one queries an information retrieval service or Internet search engine, query terms are sent to a database and citations are returned that either contain those terms, or (with an engine like excite!, at http://www.excite.com/) terms that are conceptually related to those terms. For a variety of reasons not necessary to delve into here, a percentage of the citations returned will not be relevant to the query, even though they meet the query specifications. The ones that aren't are called false drops. The larger the database, the more potential for false drops.

Precision is defined as the ratio of relevant citations returned to the total number of citations returned. And herein lies the problem with Internet searching. It is very easy, using Internet search tools, to get high retrieval. High precision is another matter. Unless a searcher is well versed in the intricacies of search syntax of the various engines, high precision can be almost impossible to achieve. In addition, even if precision is achieved, the question of quality has not been addressed. An educator's job is not to become expert in information searching: it is to educate.

Tools like Alta Vista and Yahoo are voluminous databases. They boast about that fact. But, as the above fable shows, they can be frustrating. Citation pearl growing takes a different line of attack to finding information. It assumes that if one (or a few) known quality resources can be found, they can be used as a beginning grain from which to grow an information pearl. And how does one evaluate Internet resources for quality?

Internet resources are no different from any other information resources, and educators must use the same professional judgement to evaluate them all. What is the purpose of the resource? Who created it? What is its scope? Is its information accurate? Is it current? Is it well organized? Who is its audience? Is it stylistically pleasant?

In the above story, Goffe's "Resources for Economists" is a grain from which to to grow the pearl. He is a professor of Economics. This is the twelfth edition of the work. Our economist friend is using Bill Goffe, not an Internet search engine, as her window to economics resources. From "Resources for Economists," she can find hundreds of quality resources, organized in a way that makes sense to economists. She can spend her time using the Internet rather than searching it.

This is why the Clearinghouse for Subject Oriented Resources, the WWW Virtual Library (http://www.w3.org/hypertext/DataSources/bySubject/Overview.html), the Bulletin Board for Libraries WWW Subject Tree (http://link.bubl.ac.uk/), your department home page pointers list (if available), or your library's subject resources Web pages, are all excellent places to start when searching for quality Internet information in your field. They have been put together by people rather than machines. A person or persons, sometimes experts in the field, sometimes librarians, have looked at and analyzed the resource, filtering it to you. Now you can begin to collect these resources, growing your pearl as you go. The non-technical method of talking to your reference librarian or your Internet savvy colleagues is also highly recommended.

In the Scout Toolkit (http://scout.cs.wisc.edu/scout/toolkit/searching/) we have a large selection of search tools to choose from. We have deliberately sorted them out by the amount of human intervention that goes into each, precisely because we feel that this is an important distinction for searchers to be aware of.

This is not to say that engines like Yahoo or Alta Vista aren't good, or don't have powerful capabilities. When you have questions like, "What was that URL I had on my screen when my machine just crashed?"; "What have my colleagues been posting to Usenet news?"; "How many new resources in my field have become available?"; or "Now that I've put my Web pages up, how many people are linking to them?", large search engines or indices are great resources. They are the best place to find very specific information. Be prepared, however, to spend some time learning their searching syntax to really use them to advantage for these purposes.

Finally, there is now much talk of "intelligent agent" technology, which would allow a researcher to use software to scan the Internet and bring back useful documents and resources. While this may or may not hold promise for the future, the only intelligent agent at present is the biological one in front of the monitor.

InterNIC News

This article originally appeared as part of the End User's Corner, a featured column of InterNIC News, which was published monthly by Network Solutions, Inc. and InterNIC from May 1996 through March 1998. As of April 1998, End User's Corner will be published by the Internet Scout Project.

Copyright Susan Calcari and the University of Wisconsin Board of Regents, 1994-1998. Permission is granted to make and distribute verbatim copies of the End User's Corner provided the copyright notice and this paragraph is preserved on all copies. The Internet Scout Project provides information about the Internet to the US research and education community under a grant from the National Science Foundation, number NCR-9712140. The Government has certain rights in this material.

Any opinions, findings, and conclusions or recommendations expressed in this publication are those of the author(s) and do not necessarily reflect the views of the University of Wisconsin - Madison or the National Science Foundation.

Internet Scout

A Publication of the Internet Scout Project

Comments, Suggestions, Feedback

Use our feedback form or send email to scout@cs.wisc.edu.

© 1996 Internet Scout Project

Information on reproducing any publication is available on our copyright page.