 |
|
Home > Selected Internet Resources > Colophon: About This Site |
by Edward Proctor
Entia non sunt multiplicanda praeter necessitatem. |
|
As anyone who has tried to use the Internet and World Wide Web for serious research knows, it is a chaotic jumble of ever-increasing amounts of material of widely varying quality. Many images have been invoked in an attempt to convey something of this problem, ranging from comparing the Internet to a bookstore which has been struck by an earthquake, to a library whose card catalog has been destroyed, to an enormous encyclopedia whose pages have been torn out and scrambled. Attempting to locate specific items of information in this overwhelming chaos can be like trying to find the proverbial needle in a haystack (with the added difficulty that the haystack keeps growing in size while the needle moves from place to place). What I have attempted to do with the Web site I have developed for Meyer Library, "Selected Internet Resources", is to bring some order out of this chaos, to make the Internet actually useful as a tool for doing serious research (and not just a toy to be randomly 'surfed'). As a recent American Library Association (ALA) publication observed,
The three greatest problems facing those attempting to use the Internet in a focused way are
That is, not only is there an overwhelming amount of material on the Internet with no standardized access points, but much (in fact, most) of this random assortment of information is of questionable quality. These problems are particularly acute for new users, and are further compounded by the fact that the vast majority of Internet users never progress past the third or fourth screen of hits retrieved by their search engines. [2] And at this stage in the development of search engines, a location near the top of a list of 'hits' is less likely to be a guarantee of quality than an indication that the person creating the Web page was aware of the many ways in which search engines can be 'tricked' into giving individual pages a high ranking. I have attempted to deal with this problem by sorting through and evaluating tens of thousands of Web sites, selecting the most useful and reliable among them for inclusion in my own site. I have then created standardized access points and conceptual categories to enable students and faculty to locate these materials easily. As a further refinement, I draw users' attention to the best of the best by extracting and simplifying subject-specific search engines for each particular category, placing these modified search engines at the very top of each subject page (where they will be the first thing a user encounters when he or she clicks on the page). I have tried to balance the needs of the casual user in a hurry, looking for just a few facts, with those of the scholar conducting in-depth research. Hyperlinks are presented in the form of tables which are five cells in width, in order to allow as much information as possible to be taken in at a single glance (it has been realized since the early days of telegraphy that 5 (five) characters or units are the maximum number that can be accurately scanned and processed by the human eye.[3] This apparently minor innovation has, in fact, dramatic effects on the amount of information that can be displayed on a single screen, especially "above the fold" (i.e., above the bottom of the computer screen). Equally importantly, such an arrangement and presentation of links does not require any scrolling on the part of the users. Based on observational studies of what users actually do -- and not just what they say that they do, or what Web page designers think that they do -- it has been demonstrated repeatedly that forcing users to scroll down a Web page will result in them missing much of the material contained on that page. "Only 10% of users scroll beyond the information that is visible on the screen when a page comes up."[4] For quick ready reference I have bolded and highlighted in red particularly noteworthy content pages (Highlighting in red is a practice dating back to the manuscript copyists of the Middle Ages; and bold is for the benefit of those who are color blind). I chose to make use of color and bolding to draw users' attention to particular sites in order to keep things as simple and uncluttered as possible (rather than following the more common Internet practice of describing sites as "Way cool" or "Fantastic!!!", without actually conveying any useful information). For those seeking more detailed information, I have provided a broad selection of worthwhile Web sites and links to other recommended resources (including, whenever possible, "mirror", or duplicate, sites providing similar information, in order to compensate for the distressingly-frequent phenomenon of Web pages being temporarily 'down', or inaccessible). The vast majority of previous attempts at incorporating online evaluations of Internet resources into Web sites, search engines, and directories have been too little to be of much help, but too much to allow pages to be readily scanned. That is, at the same time that insufficient information is provided to enable users to make informed choices, far too much in the way of graphics and overly-wordy recommendations with little substance to them, such as "This is a really fantastic site!!" clutter up their screens, preventing users from easily running their eyes over the page to locate relevant information. Based on several years of research into end-user information-seeking behavior, it is obvious that the major difference between the Internet and traditional print sources is that Internet users scan, rather than read. The clear implication of this knowledge for the design of Web-based publications is the crucial need to reduce the number of words appearing on a page. Even the better-known directories such as Yahoo! typically refer even to sources which they recommend highly in only the briefest of terms, e.g., the entire description of The New York Review of Books (NYRB) is: "book reviews, essays, political commentary, literary criticism, back issues archive, and much more." An Internet user unfamiliar with the NYRB would have no way of knowing from this description that it has been termed "the premier literary-intellectual magazine in the English language". [5] Far more than simply reviews of individual books, the NYRB frequently prints miniature literature surveys and thematic essays which are excellent places to begin research on a topic, and to keep abreast of the latest developments in a field (e.g. the neo-Darwinist debate that raged between the scientists Stephen J. Gould and Richard Dawkins; critiques of revisionist histories of WWII; the historical roots -- both real and imaginary -- of the current crisis in the Balkans; and so on). Another significant lack in Yahoo! and the nine other most popular search engines and directories is their failure to provide any information about the years of coverage provided by a Web site (a fact which can often be the decisive element in deciding whether or not to consult a site). This point is particularly crucial, since some studies have found that fully "98% of Internet search engine users use only the TOP TEN search engines." [6] In this case, simply saying that the NYRB maintains a "back issues archive" is of less value than would be the information that they have, in fact, archived all of their material going back to 1971 -- almost unprecedented coverage in Internet terms, where anything over two or three years old is considered virtually ancient. Even worse, the other American book review of record, The New York Times Book Review (NYTBR), is provided with no description at all by Yahoo!; not a single word to differentiate it from the hundreds of personal vanity Web pages which clog the Internet with judgements like, "I think this book is really cool!!!" At the very least, it would have been helpful to have indicated that the NYTBR is archived from 1980 onwards, and is searchable by Book Reviews only, or by Book News and Author Interviews only, or by a combination of the two. In general, far too little information is provided by most search engines and directories to be of much help to a person trying to decide which source among many to rely upon. In contrast, in my own online guides I describe the same sources (and many others) in great detail, including such information as depth and breadth of coverage, chronology, cost of access (if any), presence or absence of graphics, such as photographs, maps, charts, etc. (It is astonishing that none of the dozens of major search engines or directories indicate that the vast majority of online newspapers do not maintain images in their archives. Even commercial archive sites, such as those of the New York Times, the Los Angeles Times, and the Washington Post, which charge for access to older articles, do not provide the photographs, drawings, or charts which accompanied the original articles. It would obviously be a great help to a person looking for images to know this fact, as well as to be told that the Christian Science Monitor and the CNN Web site do provide graphics. In addition, I explain the most effective methods of searching particular databases, both their strengths and weaknesses, as well as offer suggestions for compensating for weaknesses with other sources, both print and online. In addition to gathering together some of the best news sources in a single easily-accessible location, I have made a special effort to find online replacements or equivalents for those areas in which our library holdings are weak or nonexistent. For example, Meyer Library does not subscribe to a single African or Southeast Asian newspaper. In this era of shrinking budgets, it would be difficult to justify such an expense (even if we were able to afford the cost of such subscriptions, the air mail postage alone would be prohibitive, and the news delivered most likely over a week old by the time we would receive it). In addition to being of obvious interest to our international students and faculty conducting research in these areas of the world, timely coverage of foreign news is becoming increasingly of concern to business students, in their dealings with the world as a global marketplace. And as recent events have shown, no part of the world is too remote to be of concern to Americans. Following the U.S. Embassy bombings in East Africa, much use was made of my links to direct, primary news sources, such as the online version of the Daily Nation (the national newspaper of Kenya), just one among dozens of online African news sources assembled at "Africa News Sources". Similarly, when the economy in Thailand began a rapid decline last year, triggering a domino effect in other Southeast Asian markets, I was able to provide business students with relevant newslinks to current business information through my collection of over 50 (fifty) local news sources at "Southeast Asia News Sources". Once again, this was information which would not otherwise have been available to users of the library. Similarly, I have collected hundreds of other links to resources otherwise unavailable to the users of Meyer Library, ranging from interactive car pricing guides and real-time airline flight tracking, to online catalogs of the great research libraries of the world, major statistical databases, etc., etc., thus enriching and complementing our physical collection. A feature unique to my Web site is the way in which I have placed standardized, subject-specific search engines at the top of each page, both in order to draw users' attention immediately to the best sources available, and to spare them the frustration of locating and puzzling over how to use totally different search engines on every new page they visit. In an academic library in particular, I think that it is important that there be some order and reason behind the arrangement of information resources. With this in mind, I have created a common command interface for all of the search engines I have placed at the top of my pages. That is, I have redesigned them so that they all look and function alike. In order to do this, I removed all unnecessary graphics and rewrote various pieces of HTML code and cgi and Java scripts, re-programming the very way that the search engines function. The most important modification I have made has been to change whenever possible the default setting to the Boolean AND operator; that is, I have modified the search engine interfaces so that if a user enters two or more words in his or her search, both words must be found in the Web pages retrieved. This is a particularly crucial improvement over sites which have search engines set to default to the Boolean OR operator. The importance of this can been seen in a real-life example: a student in one of my workshops was looking for information about the pollution of the Atlantic ocean. So, she entered the terms "pollution Atlantic ocean", very reasonably expecting that the URLs retrieved would refer her to Web sites dealing with this topic. Instead, she was swamped with over 14,000 hits, most of which did not appear to be relevant. What had happened was that the search engine's default was to a Boolean OR, so that it looked for Web pages containing any of the words she had entered (i.e. anything about any form of pollution -- whether of the air OR water OR soil OR ritual objects; OR anything with the word 'Atlantic' in it -- ranging from NATO to the extinct Atlantic Great Auk -- ; OR anything containing the word 'ocean' -- whether the Atlantic OR Pacific OR Indian). Obviously, this is not an efficient way to search for information. When the very same search engine is modified to do an AND search (i.e. to look only for Web pages containing all three words, pollution AND Atlantic AND ocean), a much smaller number is retrieved (and, more importantly, a number of which are likely to be relevant). Even though these modifications which I have made are not immediately obvious to the user, I think it should be clear how important they are, and how much wasted time and effort they spare users. The first such engine which I modified was that to the very useful current news site, "News Index". Their search engine is set to default to "display articles containing any of the search words" (i.e. a Boolean OR search); this results in a search for information about the effect of NATO bombing on the Kosovo refugees producing the following:
Clearly, 2,385 (two thousand three hundred and eighty-five) hits are far more than the average library user is likely to need. When exactly the same search terms are entered into my modification of the News Index search engine (located at the top of my "News & Current Events" page), the following more manageable (and relevant) results are produced:
Obviously, it is far preferable to have a search engine retrieve 2 relevant articles than for the user to have to wade through the 2,385 hits retrieved by the unmodified engine. Part of the difficulty that I have observed many library users have in attempting to use the Internet is that of forming appropriate conceptual categories with which to begin. I have made an effort to pull together materials under a variety of subject headings which will be of use to library users. Many of the categories and features on my site have grown out of my direct observation of the needs of library patrons. For example, on one occasion a faculty member assigned Jonathan Swift's "A Modest Proposal" to a class of over 30 students. Unfortunately, he neglected to place a copy of the essay on Reserve, and so the first three students to make it to the Library checked out all of our available copies. There was not enough time to order more copies by Interlibrary Loan, so I searched the Web and located a fulltext version of the essay, which was thus readily available to everyone who wanted a copy. Similarly, a few weeks later, another large class was attempting to locate copies of The Federalist Papers, with the same results (since, once again, no copies had been placed on Reserve). Again, I managed to locate some Web-based versions of this material. I realized as a result of these experiences that such needs were likely to arise again in the future, and so I gathered together several "Sources of E-Texts", in order to ease the task of locating such materials (a task which can be surprisingly tricky -- the vast majority of material retrieved whenever titles of books or essays are searched for will inevitably be secondary commentaries, citations and references, reading lists from syllabi, and so on, but very seldom the text itself). Similarly, I have assembled sources of the E-Journals at "Sources of E-Journals". Once again, I have standardized and modified the search engines which I have placed at the tops of my pages so that they will default to a Boolean AND search. The importance of this modification can be seen by the fact that the search engine which the Colorado Alliance of Research Libraries (CARL) provides for e-journals at their site, defaults to an OR search, unfortunately rendering their excellent collection much less accessible and less useful than it can be (and less useful than it is, when approached using my modified search interface). As set up by CARL, a patron searching for electronic journals on "Indian History" would have to wade through 192 (one hundred and ninety two) hits returned by a search which had defaulted to looking for "Indian OR History" (whereas my own modification, by virtue of the fact that I have set it to default to an AND search, automatically narrows down the number of hits to the 2 (two) which are actually relevant, since only those items are retrieved which deal with both "Indian AND History"). I have covered numerous other subjects, creating in the process dozens of new categories in which to place them. Among these are
I have made an effort to cover Minority issues, including All of my pages have been designed with the special needs of the visually handicapped in mind, so that they will be readable by a variety of text-only browsers and speech synthesizers. I also place within the HTML code itself detailed descriptions of any images located on my site. I anticipate that most of my pages will meet the standards currently being developed by the U.S. Access Board, which will soon "unveil standards aimed at ensuring that Web sites operated by firms doing business with government agencies are fully accessible to the disabled"[7]standards which will have an effect on U.S. Internet sites comparable to the effects of the Americans with Disabilities Act (ADA). The importance of what I have accomplished is underlined by the fact that it is increasingly becoming obvious that an entire generation of students has come of age thinking that if something is not on the Internet, it doesn't exist. Numerous anecdotal accounts from Missouri State faculty lead me to believe that we are not immune to this problem. If ( as seems to be the case) it is inevitable that students are increasingly relying on Internet sources exclusively, it is even more vital that what they rely on be of good quality and readily accessible (obviously, the more accessible an information source is, the more likely a student will be to use it). This is of importance not just for students, faculty, and other library users, but for librarians themselves. Concern over librarians keeping up with the rapidly-changing information environment was, in fact, a focal point of the 17th International Essen Symposium on Information Superhighway: The Role of Librarians, Information Scientists, and Intermediaries, where it was observed that "the development of electronic resources has been so rapid that many information professionals have been left behind" and that a "lack of information quality control, and increased forms of access require librarians to enhance resource selection activities".[8] A study of academic users of the Internet in the current (May 15) issue of the Journal of the American Society for Information Science summarizes the situation well, and underlines the need for projects such as my own:
This is what I have attempted to do. |
| Send comments & suggestions to Edward Proctor |
Acknowledgements |
|
I would like to thank Mrs. Iryna Bernatska for her invaluable help. Her imaginative and thoughtful contributions have significantly improved this site, both technically and in terms of content. Thanks to her, this project is better than it would have been. |
Footnotes |
|
1. Kennedy, Shirley Duglin, Best bet Internet: reference and research when you don't have time to mess around. Chicago: American Library Association, 1998: v. 2. DeUlloa, John R., PromoteOne,http://www.promoteone.com/author.html, May 18, 1999. 3. "In devising the symbols for the digits, Morse seems to have recognized that a receiving operator could easily read by eye up to five printed dots, but that a larger number of dots would be more difficult to read quickly and accurately". The Art & Skill of Radio-Telegraphy, 2nd. Rev. ed., Chapter 19, A Brief History of Morse Telegraphy, Part I, William G. Pierpont. 4. http://www.useit.com/alertbox/9605a.html, May 17, 1999. 5. Esquire, quoted at http://www.nybooks.com/, May 19, 1999. 6. Yahoo, Netscape, Excite, Infoseek, Lycos, Alta Vista, Web Crawler, Hot Bot, Look Smart, Go To. Advanced Access, http://www.advancedaccess.com/searchengineresultsframed.htm, May 18, 1999. 7. Seminerio, Maria, "'Handicapped' Access Hits the Web", ZDNet Tech News Now, April 18, 1999. http://www.zdnet.com/zdnn/stories/news/0,4586,2243282,00.html. 8. Helal, Ahmed H., and Weiss, Joachim W., eds. Information Superhighway: The Role of Librarians, Information Scientists, and Intermediaries. Proceedings of the International Essen Symposium (17th, Essen, Germany, October 24-27, 1994). ERIC NO: ED381176: i. 9. Voorbij, Henk J., "Searching Scientific Information on the Internet: A Dutch Academic User Survey", Journal of the American Society for Information Science, 50(7), May 15, 1999: 607. |
| Please send suggestions & comments to EdwardProctor@MissouriState.edu |
|
||
|
|
||