Vol. 19, No. 3/Spring 1997
Nuts & Bolts: Caveat Websurfer
By Michael Jesse
Whenever I teach an introductory Internet class I almost always describe the Internet as the "world's largest vanity press."
The point, which new users need to understand, is that the old rules of publishing no longer apply. All of the financial hurdles to traditional printing are gone -- and gone with them are the layers of editors and proofreaders and libel attorneys who generally ensured that what ended up in print had passed some fundamental rules of scholarship -- or at least journalism.
Since anyone can publish anything on the Internet we need to be able to distinguish quality from the questionable. There are a number of websites out there which address the issue of evaluating websites. (You can find a list of them on Yahoo at www.yahoo.com/Computers_and_Internet/Internet/World_Wide_Web/Information_and_Documentation/Evaluation/) Most of these sites are academia-based, but the criteria they list are generally applicable to anyone trying to judge websites. They include:
Authority: Who is the creator of the information? Who is putting it on the web? What credentials, if any, do these people claim?
Affiliation: Is the producer of the information affiliated with specified organizations or institutions? Does this add to the website's authority level? Does it suggest bias?
Accuracy: If we spot-check some of the factual information against our own knowledge or against other sources on the same topic, does the website appear to be accurate? Are there spelling or grammar errors which undermine our trust in the site's overall concern for accuracy?
Intent: What is the stated or implied intent of the site? Why does it exist?
Accessibility: Is the site well-organized and easy to navigate? Or is it weighed down by needless heavy graphics? Are there "back to home" buttons on individual pages?
Currency: Can you tell when the information was created, and when it first appeared on the site? Does the site indicate when it was last updated, or how frequently it is supposed to be updated?
Recommendations: Is the site recommended by any website review services? Is it listed on Yahoo, or other directory sites? Is it recommended by subject experts in the field.
Reliability: Will the site still be around next month if we bookmark it? Is the server frequently down or too overloaded to accept further requests.
While these are all fine website qualities, don't expect to find them all clearly covered on each site. But depending on the circumstances you may not care. For example, a Shakespeare quotation site doesn't need a daily update. And the organizer of a site devoted to "Brady Bunch" trivia need claim no special credentials (other than having way too much free time).
On the other hand, if a site claims to monitor recent developments in breast cancer research, or to provide current statistics on capital punishment across the country, then you'll want to take extra steps to establish the credibility of the source and its claims.
It may be handy to keep in mind the same "five Ws" which reporters learn in journalism school:
Who is behind the site and do they have sufficient expertise to make these claims?
What is the actual content? Is it a technical or controversial topic? Is it at variance to other sources? Can I, at least partly, confirm this information elsewhere?
Where can we contact the source? Is there an address? Where can we confirm or refute the information?
When was it created? When updated?
Why is it on the Web for free?
The "who" and the "why" may be the most important. Very few websites support themselves via paid subscription or (enough) advertising, so why is someone putting this out for free? Does it exist to advocate a certain point of view? There's nothing wrong with using an advocacy source -- journalists have done so for generations -- but it's important to be able to recognize the bias in order to counterbalance it.
Interpreting and backtracking in URLs:Under most circumstances domain names are fairly accurate descriptors. If the domain name ends in .gov or .edu then it's likely the site is supported by a government agency or a university and is part of the overall mission of that institution. Similarly, if the domain ends in .org it is apparently associated with a non-profit organization. (Several new domain suffixes will start showing up soon, including .firm for companies, .info for information services, etc.).
One technique I always show reporters is how to backtrack in a URL. If you've found a page through a search engine and it seems to be useful yet contains no identifying information nor a link to its homepage, you can generally find your way backwards through the URL. Just go to the location window and start lopping off parts of the URL, backing up to the next earliest slash each time. If there's a tilde in the URL, it often indicates a personal page. For example, "www.whatever.net/~jones/welcome.html" probably belongs to someone named Jones who bought space from an Internet provider with the domain www.whatever.net. Personal pages are also common at .edu sites, and indicate a student or professor.
For more on this, see "Understanding URLs" at milton.mse.jhu.edu:8001/research/education/url.html.
Hoaxes, legends and parodies:Hoaxes and urban legends are more common on newsgroups than on the web because by their nature they thrive in being retold. On the more static web you are more likely to find parody sites, which aren't really intended to be believed (though sometimes they are). For example, there's "Stale," (www.stale.com) the parody of Michael Kinsley's disappointing "Slate" zine (www.slate.com). During the 1996 presidential campaign there were parody sites on Clinton, Dole, Pat Buchanan and Ross Perot. There's a Real/Audio parody site called "Real/Aroma," and countless jibes of Microsoft, including "Microsnot," and "The Diary of Bill Gates."
While you do have to be alert to intentional deception, the greater danger is simply poor scholarship. Recently I found a visually beautiful site on the filmmaking Coen brothers, but there were so many spelling and grammatical errors that I couldn't trust a word. Backtracking in the URL I discovered that the parent site builds its content by offering free space to individuals to build pages on their favorite topics. This is a nice concept -- the Internet at its democratic best perhaps -- but without some kind of quality control it's a difficult source to trust.
Website reviews and recommendations:One good way to increase your chances of finding a quality website is to work from lists recommended by others, especially those you know and respect. There are several "review" sites on the Internet, including Excite (www.excite.com/Reviews), Magellan (www.mckinley.com/) and Lycos Top 5% (point.lycos.com/categories/). Unfortunately most of the "reviewers" for these sites seem able to express themselves only in terms of whether a site is "cool."
Perhaps more reliable, though less descriptive, are lists of recommended sites from recognized experts. In our field, some of these would include SLA News Division (sunsite.unc.edu/slanews/), Barbara Shapiro (www.gate.net/~library/index.html), FACSnet (www.facsnet.org), the Poynter Institute (www.reporter.org/poynter/home/index.htm) or E&P (www.mediainfo.com/). Plus there are many reliable print sources such as the Internet Newsroom and Internet World. Also many traditional library publications, including Library Journal, now have regular features on recommended websites.
Don't cancel your Dialog account:Although it's possible to answer a lot of routine questions using free websites, I still use Dialog and Nexis virtually every day. When I need to provide a reporter with background on a topic, I'm not going to rely on the Internet alone. That being said however, it's remarkable how often you can find just what you need for free on the Internet. And while the danger of misinformation certainly exists, you really needn't fear that every site is an elaborate hoax set to trap you. The overwhelming majority of websites are honest efforts to share information. With common sense, some experience and occasional detective work you can arrive at reasonable judgments about which websites to treat as reliable sources of information.
AOL access tip:By the time this is published AOL may have solved its access problem. If not, here's a way around it. If you have Internet access through a local provider, connect to it first, then launch your AOL software. Before launching the login, change the pulldown menu from dialup to TCP/IP. This links you via the Internet instead of via AOL's overloaded dialup.
Naturally, this only works if you have Internet access independent of AOL. But if you've been depending on AOL for Internet access you really should reconsider your priorities. For $19.95 a month you can get reliable Internet access from a local provider and use AOL on an as-needed basis for its proprietary content.
Michael Jesse can be reached at email@example.com. He is Library Director at the Dayton Daily News.
The page was reformatted on 9/17/02. The content, including URLs and contact information, on this page has not been updated to preserve the historical record.