한국   대만   중국   일본 
Missing Links | John C. Dvorak | PCMag.com
The Wayback Machine - https://web.archive.org/web/20110806041207/http://www.pcmag.com/article2/0,2817,33326,00.asp

Missing Links

John C. Dvorak

When you analyze the strengths of the online experience, the hyperlink has to be at the top of the list, followed perhaps by convenience. In fact, the whole idea of the World Wide Web was to make the concept of hyperlinks universal so that people could jump around in a quest for knowledge or wisdom. The exact method in which these links should be employed has never become standard. Here's the problem with linking and what to do about it.

Overlinking. When the browser was first introduced, people did not understand the linking concept, and we had two wrong-headed schools of thought dominating the scene. The first was the "gone berserk with links" concept, where almost everything on the page was hyperlinked. You'd never find, for example, a reference to a city or country without a hot link to some Web site dedicated to that place. If Chicago was in the text, there would be a link to the Chicago Chamber of Commerce or some other Chicago information site, just in case the reader had never heard of Chicago. This resulted in text that was almost unreadable because of the number of underlined or odd-colored links. Worse, the reader didn't know whether any of these links were actually important in regard to the text.

Overlinking eventually resulted in some strange, surreal creations. Penn Jillette (of Penn and Teller fame) wrote an unusual column for Excite.com and epitomized overlinking by linking odd words in the text of an essay to off-the-wall sites that had nothing to do with anything: The sites were just interesting. This idea never became popular, and now only bloggers use it experimentally.

Null linking. The opposite extreme of overlinking is not underlinking: It's no linking whatsoever. This was practiced by almost all the newspapers in the world and is still employed as a technique to keep people on a site. It's almost laughable. Even if the article is a rebuttal to someone else's commentary, there's no link to the source or to the original commentary, as though the Web material were on the printed page and linking were impossible. Not understanding that the Web is a different medium than ink and paper shows a lack of understanding of the nature of media and is somewhat insulting to the reader.

In the early days of null linking, newspapers were simply afraid to link, thinking it was plagiarism or just weird. Many simply could not comprehend how it worked. Over time, this rationale (or stupidity) evolved into the concept of the closed system. If a site links to the outside, the reader might discover something more interesting and thereby miss some banner ads. This thinking does not allow Web writing to evolve into anything special.

When writing for the Web, I, for example, like to link to pertinent information that will enhance the column (this information tends to be optional). I assume that readers know how to go back and forth. If the links entertain readers, those people are actually more likely to come back to get more excellent links. Thinking that readers will zone out and disappear into the Web netherworld is silly.

Overlinking is probably better for readers than null linking. Because the null linkers include most of the large information providers, the whole idea of the hyperlink has largely been ignored in talking about the architecture of the Web. XML is an important topic, but links are not. Why not?

Dimensional linking. Instead of e-commerce companies wringing their hands over XML implementation, how about some real effort to make the informational experience on the Web more valuable? The Web is used mostly for information gathering. Just as security and other Web features have levels, why can't linking? This, of course, would require fundamental architectural changes in HTML. But wouldn't it be incredibly valuable if there were two, three, or more levels of links that could be turned on and off within the browser? The levels could include no links, pertinent links, ancillary links, definitional links, and so on. If you were reading about a subject that you knew nothing about, you could hit a function key and the page would light up with links to provide you with definitions of seldom-used words within the text and links to explain those concepts. Perhaps you could toggle to a list of links that connect to advanced concepts. The possibilities are endless, and rich text could be seriously rich.

Discuss this article in the forums.



blog comments powered by Disqus